Files
msd-core/tests/planning-inspect.test.cjs
Tom Boucher ab69b9ce56 enhance(#3987): guard slug re-derivation and the swallowed-precondition shape — §8.5 was guardable after all (#3999)
* feat(#3987): guard slug re-derivation, and record why the swallow shape cannot be guarded

Epic #3473's Decision 1 requires the wrong call site be UNREPRESENTABLE. #3984
measured that two of the nine §8 rules had no guard at all and recorded both as
"Shipped - test-covered". This closes one of them, proves the other cannot be
closed the same way, and corrects two false claims I merged yesterday.

1. §8.3 - scripts/lint-slug-derivation-drift.cjs.

   generateSlugInternal (src/core-utils.cts) is the canonical owner; #3883 removed
   11 inline copies. Nothing prevented a twelfth: no slug guard existed in
   scripts/ or eslint-rules/.

   The detector is STATEMENT-scoped and matches the shape the real copies took -
   one statement carrying BOTH .replace(<negated class>, '-') and
   .replace(/^-+|-+$/, ''). Statement scoping is what buys the precision: the
   loose LINE-level form yields 18 hits with 7 unrelated, a material
   false-positive rate. Measured on the tree: 5 flags, 2 TRUE, 3 SANCTIONED,
   0 FALSE.

   The three sanctioned sites are allowlisted with a reason each, following
   lint-phase-enumeration-drift's form rather than a bare denylist. The owner
   itself is listed explicitly even though it escapes by construction - an
   implicit escape is a latent bug, and the next person to touch line 192 would
   not know the guard depended on it.

2. Both TRUE positives were live defects, not style.

   scripts/qa-smell-ratchet.cjs reproduced the canonical formula including the
   60-cap but trimmed BEFORE truncating - the #2849 bug - and never
   transliterated. The divergence is total, not cosmetic:

     canonical  "privet-mir-privet-mir-privet-mir-privet-mir-privet-mir-prive"
     inline     "tail"

   Cyrillic collapsed to nothing and only the ASCII remainder survived, so the
   ratchet was keying on wrong identifiers for any non-ASCII input.

   tests/planning-inspect.test.cjs carried a helper whose comment claimed parity
   with getPhaseDirFromPhaseId. That function now transliterates; the helper did
   not, so the test asserted against a stale formula while looking correct. Both
   now route through the seam.

3. §8.5 - measured, and deliberately NOT shipped.

   A candidate detector (swallowing catch + errno-retry-set test in the same
   function) gives 26 flags across 11 functions: 0 TRUE, 26 FALSE. Every one is
   best-effort unlink/rm/close cleanup, lost-rename-race backoff, or a deliberate
   null fallback. The file-scoped variant is worse at 71.

   Worse than the noise: the only known true instance was removed by #3885, so
   there is NO POSITIVE CONTROL - the guard cannot be shown capable of failing,
   which this repo requires of every drift guard. Shipping it would add a guard
   nobody can trust and nobody can test.

   The ADR now records the measurement and the reason, keeps §8.5 at
   "Shipped - test-covered", and points at the #1884 regression test as what
   actually enforces it. An honest "not detectable at acceptable precision" beats
   a guard that only ever passes.

4. Two claims I merged into the ADR yesterday were wrong.

   §8.9 said 17 of 19 subsumed children have a test citing their issue number,
   and that #3364 and #3812 have none. Both halves are false, and the claim came
   from a NUMBER-GREP - inside an amendment whose own subject is that a text match
   is not a fact.

     #3364 IS cited: tests/runtime-marker-resolution.test.cjs:107,
       T3 installMarkerResolvesWhenEnvAndConfigAbsent_3897 (#3364), asserting at
       :115-119.
     #3812 IS covered: tests/gen-state-md-docs.test.cjs:374, asserting at :382.

   Corrected to 19 of 19.

   #3812 does carry a real finding, though a different one: it is PARTIALLY
   DELIVERED on a CLOSED issue. The shipped fix declares cardinality for
   frontmatter keys, but #3812's stated acceptance was about the
   ## Current Position BODY section, and docs/reference/state-md.md:196-208 still
   has no normative single-valued/overwrite sentence and no pointer to
   ## Performance Metrics for history. Recorded in the ADR and left for #3812 to
   re-open - fixing it here would bury a scope question inside an unrelated PR.

Note on B6: this ADDS a guard, and B6 said the net count must fall. #3951 already
amended that clause - a guard ledger is a claim about COVERAGE, not count - which
is what makes adding this one honest rather than contradictory.

Verified: the guard flags 0 on the fixed tree, and PROVES IT CAN FAIL - a fresh
inline copy planted in src/ makes it exit 1 naming the exact statement. All three
sanctioned sites were confirmed exempt BY the allowlist, not by accident of the
pattern, by re-attributing each to a non-exempt path and watching it flag.
build:lib, lint and lint:ci all exit 0.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* chore(#3987): add the changeset fragment

Doc-only, so it carries forward from the verified sha rather than costing a
second matrix run.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(#3987): §8.5 IS guardable — I was wrong, and the guard found a live defect

Two orthogonal reviews. The correctness review overturned my central judgment,
and it was right.

1. I concluded §8.5 was "not detectable at acceptable precision" and recorded
   that in the ADR. False.

   My evidence was 26 flags / 0 TRUE / 26 FALSE. The reviewer pointed out what I
   had not: all 26 false positives are CLEANUP verbs - rmSync 54, unlinkSync 43,
   closeSync 17, chmodSync 12 - and the obvious narrower predicate was never
   tried. A swallowed cleanup is legitimate best-effort. A swallowed CREATION is
   a precondition silently lost, which is exactly the #1884 shape.

   Measured properly, in three stages:
     swallowing catch                                     911
     + try-block calls a CREATION verb                     24
     + enclosing function references a *_ERRNOS set         0

   0 flags, 0 false positives. The `*_ERRNOS` naming key is empirically total -
   all 10 retry/tolerate sets in src/ follow it.

   My second claim was worse. I wrote that no positive control exists because
   #3885 removed the only true instance, so the guard "cannot be shown capable of
   failing". That is self-refuting: this very PR's slug guard proves-it-can-fail
   on a synthetic tree, and the pre-#3885 blob is available as exactly such a
   fixture. It is now the control, and it works in both directions - the rule
   flags 0c43d853e^:src/planning-workspace.cts at line 210, the line the fix
   commit's own message cites, and reports zero on the post-fix code.

   I stopped at the first negative result on the option that meant less work.

   Shipped as eslint-rules/no-swallowed-precondition.cjs, wired into the existing
   src/**/*.cts ESLint block rather than a scripts/lint-*-drift.cjs: no script in
   scripts/ requires typescript/espree/acorn, and scripts/ ships to consumers, so
   a .cts-parsing standalone guard would add a devDep at consumer runtime. The
   ESLint block already parses .cts for free.

2. The guard immediately found a live defect of the same class.

   src/capability-lock.cts swallowed a mkdirSync on the lock directory, then
   acquireLock classified the follow-on failure as `code !== 'EEXIST' → return
   null`. A real EACCES/EROFS makes openSync(lockPath,'wx') fail ENOENT, which is
   not EEXIST - so a fatal filesystem error was laundered into "lock
   unavailable". Same defect as #1884, different laundering target.

   Fixed the way #3885 fixed #1884: the creation failure propagates. Regression
   test proven fail-first by hand - with the fix stashed, EACCES was laundered to
   null; restored, it throws.

   The strict rule does NOT catch this shape (its errno classification is an
   inline literal, not a named set). The rule is deliberately left strict: the
   broadened form had 2 false positives - capability-lock.cts:408, the deliberate
   EEXIST steal protocol, and commonjs-marker.cts:131, which returns a distinct
   documented outcome. The gap is noted in code rather than papered over with a
   noisy predicate.

3. The security review found the slug guard's exemption FAILED OPEN.

   currentFunction was never reset, and only a column-0 `function` declaration
   updated it, so exemption bled from an allowlisted declaration to the next one.
   generateSlugInternal exempted 50 lines for an 11-line function. A
   re-derivation planted anywhere in that window was silently exempt - the same
   fail-open shape that produced a blocker in #3897, and an allowlist is a
   SUBTRACTION so a mismatch fails open by construction.

   Extent is now tracked by real brace depth, and a test plants a violation after
   each allowlisted function's real closing brace and asserts it IS flagged.

4. Also from the security review: the guard was a CI-DoS and narrower than I
   claimed.

   Its unbounded [^\]]* was re-scanned from every `.replace(/[^` start: 54.3s on
   a 1.28MB line. It imported MAX_REGEX_LITERAL_LEN and never called
   readRegexLiteralAt - the bounded tokenizer that exists for exactly this. Now
   routed through it with a 2MB file cap: ~200ms.

   15 of 25 genuine re-derivations evaded. Widened to catch replaceAll, {1,},
   \s*-wrapped classes, escaped ], literal new RegExp(...), five trim spellings,
   .split().join(), and multi-line .replace( args - still 0 false positives.
   Two forms still evade and are documented as deliberate gaps with negative
   tests: the two-statement/temp-var form and new RegExp built from a variable.
   Both need data flow, and guessing at it is how a guard becomes noisy.

   Also fixed: // inside a string truncated the line, a ; inside the collapse
   regex split the statement (a one-character bypass), and SCAN_EXT omitted
   .mjs/.tsx/.jsx.

5. A regression I introduced, caught by the same review.

   qa-smell-ratchet.cjs top-level-required a build output that is not
   git-tracked, so the script hard-failed MODULE_NOT_FOUND before build:lib -
   including for --help, which previously had no build dependency. The require is
   now lazy at the point of use.

6. Four of my own tests were vacuous or weak.

   T9's input yielded an identical string under the buggy formula, so it passed
   on the implementation it was meant to catch. T12 compared maxLen null vs 60 on
   an 18-char name, where they agree trivially. T9-T12 all asserted
   generateSlugInternal directly, so they would pass unchanged if both call-site
   fixes were reverted. And prove-it-can-fail was scoped to scanRepo, never the
   CLI - dropping main()'s exit-code line would have kept every row green.

   All rewritten with discriminating inputs, per-call-site rows that red when the
   fix is reverted, 59/60/61 boundaries, an entirely-non-alphanumeric row, and a
   CLI row asserting the real subprocess exit code and both sanitizeForReport
   sites.

Verified: both guards flag 0 on the tree and both prove they can fail. The
swallow rule's control is confirmed in both directions - pre-#1884 shape flagged,
post-#3885 shape clean. build:lib, lint and lint:ci all exit 0.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* docs(#3987): record that §8.5 IS guardable, and correct a correction that made a ledger worse

Three ADR corrections, two of them to text this branch wrote hours ago.

§8.5 advances to Enforced. Its previous entry said the rule was not detectable
at acceptable precision. That was wrong twice: the 26 false positives were
uniformly CLEANUP verbs, which is a reason to narrow the predicate rather than
abandon it, and the claim that no positive control exists was self-refuting - the
pre-#3885 blob is available as a fixture and this repo's own guards prove-it-can-
fail on synthetic trees. Narrowed to creation verbs plus a *_ERRNOS reference:
911 -> 24 -> 0 flags, 0 false positives, control confirmed in both directions.
The entry keeps the wrong reasoning visible, because a high false-positive count
being evidence the predicate is wrong - not evidence the rule is unguardable - is
the transferable part, and the first negative result is most seductive when it is
also the answer that means less work.

§8.9's correction is itself corrected. The original 17-of-19 claim was CORRECT
for the predicate it stated; this branch silently swapped cited -> covered and
declared 19 of 19. #3812 appears in zero test files. Changing what a word means
to make a ledger read better is a worse failure than the miscount it claimed to
repair. Both predicates are now reported separately - 18 of 19 cited, 19 of 19
covered - because §8.9 asks for a test NAMING each child, so 18 is the number
that answers it. #3812 is also re-opened for real, rather than the first draft's
promise that it could be.

§8.3 stays Shipped - test-covered rather than advancing. The slug guard catches
the copy-paste class and a dozen variants, but two forms still evade by decision
(temp-var split, new RegExp from a variable) because both need data flow. Naming
them keeps the status honest: the wrong call site is much harder to write, not
unrepresentable.

Closes #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* chore(#3987): backfill changeset pr number

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(#3987): replace my own wall-clock assertion, and close the guard that let me write it

CI went red on ubuntu shard 2/3. The failing test was mine, and the failure was
the test, not the code.

  a 1.28MB line ... scans in well under a second (was 54.3s pre-fix)  7368ms

It asserted ELAPSED TIME. ~200ms locally, 7.4s on a shared CI runner. The bound
introduced for the MAJOR-2 DoS fix works - 7.4s against a 54.3s pre-fix baseline
is the fix doing its job - but an absolute wall-clock threshold on shared
hardware is a race, not an assertion. CLAUDE.md says so directly: "Clock Seams:
Do not assert on wall-clock time." I wrote the anti-pattern the project bans, in
a PR about guards.

Raising the threshold would only move the flake. The row now asserts a
DETERMINISTIC bound instead: an instrumentation seam on drift-scan.cjs counts
readRegexLiteralAt calls and characters examined, and the test asserts
charsExamined stays under an absolute ceiling. Measured on the same 1.28MB
fixture: 120,000 calls, 48,000,000 chars - two orders under the ceiling. The
pathological fixture is kept; only the thing being asserted changed.

Proven to still discriminate: with MAX_REGEX_LITERAL_LEN raised to simulate the
unbounded pre-fix behavior, the same fixture does not complete in 120 seconds,
versus ~0.3s bounded. It is a real regression test, not a tautology.

Then the second half, which is the same defect class as the rest of this PR.

  eslint-rules/no-elapsed-assertion.cjs matched only the EXACT identifiers
  ^(elapsed|duration|took|ms)$.

I used `elapsedMs`. It evaded the rule entirely. tookMs, durationMs,
elapsedTime and msElapsed evade the same way. A guard that cannot see the
violation it exists to catch is exactly what this PR is about - it just happened
to be an existing rule rather than one of the two I came here for, and it was
found because I committed the violation it should have blocked.

Widened to /^(?:elapsed|duration|took|ms)(?:[A-Z]\w*)?$/ plus a narrow
start/endMs delta pair. Deliberately NOT a blanket *Ms suffix: a first draft did
that and produced 2 false positives on `timeoutMs` in
plan-phase-stall-detection, which is a configured timeout and not a measurement.
Verified negative on params, items, forms, terms, dirnames, timeoutMs,
cacheTtlMs and staleAfterMs.

Measured over the five files carrying camelCase timing identifiers: 0 true
positives beyond my own, so nothing else needed rewriting. The rule's own test
file gains a row asserting `elapsedMs` flags, proven to fail against the
pre-widening rule - the same prove-it-can-fail standard both new guards in this
PR are held to.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(#3987): a comment I added leaked a Claude reference into every runtime install

The runner went red with 4 failures in tests/install.test.cjs:

  Leaking: .hermes/scripts/lib/drift-scan.cjs
  Leaking: .qwen/scripts/lib/drift-scan.cjs

The instrumentation seam added for the deterministic bound carried a comment
naming CLAUDE.md as the source of the no-wall-clock-assertions rule. scripts/
SHIPS to consumers, so that comment was installed verbatim into hermes and qwen
trees, and the install suite scans for exactly this - a Claude-specific reference
reaching a non-Claude runtime.

The rule is real and worth citing; the filename is not portable. The comment now
says "this repo's test rules" and states the rule inline, which is what a reader
of an installed tree actually needs anyway.

Worth noting what caught it: not lint, and not the two guards this PR adds - the
install suite's full-tree scan, which exists precisely because a shipped file is
read by runtimes that have never heard of CLAUDE.md. Same lesson as the rest of
this PR from the other direction: the check that matters is the one that can see
the surface where the defect actually lands.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(#3987): a test fixture swallowed 46 git exit codes and produced a silent false negative

CI red on ubuntu shard 3/3:

  tests/health-validation.test.cjs:2029
  expected exactly one W024, got [{"code":"W006", ...}]

Not caused by this branch, and the evidence is decisive rather than a hunch: the
SIBLING test at :2039 builds the IDENTICAL fixture with the identical
commitsAhead and asserts the same thing, and it PASSED in the same process, same
file, same run. Same input, both outcomes - which rules out logic, ordering,
sharding and environment, and leaves a per-invocation nondeterministic failure
inside one fixture build.

The mechanism is an unchecked exit code, 46 times over. The W024 fixture performs
~46 runGit spawns and never checks a single one. runGit returns failures as DATA
and never throws, so one silently-failed `git commit` yields 19 commits instead
of 20, or a silently-empty `git rev-parse HEAD` yields a blank state_head. Either
drops readStateHeadFreshness below the advisory threshold, W024 never fires, and
only W006 remains.

Reproduced exactly: 20 commits -> ["W006","W024"]; 19 -> ["W006"]; blank
state_head -> ["W006"] - byte-identical to the CI assertion dump.

The arithmetic is what hid it. At threshold-1 and threshold+1 a lost commit still
produces the asserted answer; only the exactly-at-threshold cases sit one commit
from a false negative. Two of the seven tests are in that position, and CI hit
one. That is why it had never been seen before, and why it surfaced now: this
branch adds three test files, which reshuffles the cost-weighted shard partition
and moved this file into a chunk where the latent flake fired.

My files were checked as suspects first and cleared: all fixtures mkdtemp-unique,
no process.chdir, no .planning/ writes, no git spawns, and node --test gives
per-file process isolation regardless.

Fixed at the cause, not the symptom. A mustGit wrapper throws on a non-zero exit
with the command, exit code and stderr, and all nine call sites route through it.
The fixture now asserts its OWN preconditions before the assertion under test
runs - the seed head is non-empty, and `git rev-list --count <seed>..HEAD` equals
the requested commitsAhead - so a fixture that did not build what it claims fails
loudly as a FIXTURE ERROR naming got-versus-asked, instead of quietly handing a
weaker input to the assertion.

Proven: dropping one commit now raises
  FIXTURE ERROR: requested commitsAhead=19 but git rev-list --count reports 18
where it previously produced a silent ["W006"] pass-for-the-wrong-reason. 64/64
tests in that block pass unperturbed.

Deliberately NOT done: no threshold change, no retry, no loosened assertion, no
skip. The assertion was correct; the input was silently wrong.

Worth naming, because it is the same shape from the other side: this PR ships
eslint-rules/no-swallowed-precondition.cjs, whose entire subject is a swallowed
precondition failure being laundered into a plausible downstream outcome. This
fixture is that defect in test code - the swallowed git failure was laundered
into a legitimate-looking "W024 did not fire". The rule does not cover test
fixtures, so the connection is noted at the fix site rather than enforced.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(#3987): two tests wrote to committed files; the shard packing decided when that mattered

CI red on windows-latest shard 1/3 only:

  "gen-exit-code-registry: CLI" > "a --write run redirected to a tmpdir leaves
  every committed artifact untouched"
  AssertionError: hooks artifact must be untouched

The Linux runner passed the same sha at 40425/40425. It is Linux-only, so a
Windows-scheduling defect is structurally invisible to it.

Root cause, established by measurement rather than inference.

tests/cli-exit.test.cjs appended a corruption marker to the REAL COMMITTED
hooks/lib/exit-code-registry.js, held it corrupted across a full subprocess, and
restored it in a finally. tests/exit-code-registry.test.cjs reads that same real
file before and after its own subprocess and asserts byte equality. If it samples
while the other test holds the file corrupted, it fails. The landmine is
pre-existing, from 2ea5efc15 (#3911).

What this branch changed is WHEN the two run together. scripts/run-tests.cjs
shards by cost-weighted LPT over the sorted unit list, so adding three test files
repacks the bins:

  merge-base c3e667df3 (838 files): cli-exit -> shard 1, exit-code-registry -> shard 3
  HEAD       03b342601 (841 files): BOTH in shard 1, same argv chunk, one
                                    node --test process, concurrent

Co-location is necessary but not sufficient - Linux shard 1/3 also had both and
passed. Windows loses because TEST_CONCURRENCY defaults to 2 there against 4
elsewhere, spawn cost is ~10x, and the sibling corruptor holds one of only two
slots through a ~90s tsc compile. That turns a sub-second overlap into seconds.

Not a path-separator or case-sensitivity issue, and not CRLF - .gitattributes
pins * text=auto eol=lf. Redirection was not at fault either: ensureScriptsOut
derives all five --out flags correctly and gen-exit-code-registry.cjs honours
them with no __dirname escape.

Fixed at the cause: no test writes to a committed file any more. Both corruptors
now copy to a mkdtempSync tmpdir, corrupt the COPY, and point the generator at
it. Repinning or reordering the shards would have turned CI green while leaving
the landmine armed for the next reshuffle.

That required closing an inconsistency between two sibling generators.
gen-exit-code-registry.cjs already accepts
--out/--scripts-out/--hooks-out/--dts-out/--sh-out and honours them under
--check; gen-hooks-cli-exit.cjs hardcoded OUTPUT_PATH and had no flag surface at
all, so its corruptor could not be redirected anywhere. It now takes --out in the
same style, honoured by both --write and --check, and is a no-op when absent -
verified: a bare --check on the default path still exits 0.

ensureScriptsOut moved to tests/helpers/exit-code-artifact-flags.cjs and both
test files import it. Hand-rolling a second copy of the flag derivation would
have been a re-derivation of exactly the kind this PR ships a guard against.

Verified: both tests still detect corruption (proven by defeating the check and
watching them red, with a positive control showing an uncorrupted copy exits 0);
SHA-256 of hooks/lib/exit-code-registry.js and hooks/lib/cli-exit.js identical
before and after running both rewritten bodies, and git reports nothing under
hooks/ modified - that is the property that was violated. A repo-wide search for
the corrupt-then-restore-in-finally shape against hooks/ found no other
instances.

One detail worth recording: the tmpdir test keeps --declaration pointed at the
real committed declaration rather than copying it, because the generated banner
embeds path.relative(REPO_ROOT, declarationPath) - copying it would produce a
false drift unrelated to the injected corruption. The declaration is read-only on
that path and never written.

Refs #3987

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

---------

Co-authored-by: sim <sim@local>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-28 14:40:02 -04:00

2927 lines
126 KiB
JavaScript
Raw Blame History

'use strict';
/**
* Tests for `src/planning-inspect.cts` — the schema-v1 canonical planning
* snapshot (`query planning inspect` / `query planning.inspect`, #2790).
*
* Design: .gsd/phase/feat-2790-planning-inspect/40-design.md
* Test matrix: .gsd/phase/feat-2790-planning-inspect/50-test-matrix.md
*
* Fixture provenance (CONTRIBUTING.md "Fixture provenance (#2371)"): every
* `.planning/` document shape written by this file's fixture builders is
* derived from the SHIPPED templates the product author wrote —
* `gsd-core/templates/requirements.md` (checkbox bullets, `## Traceability`
* table), `gsd-core/templates/phase-prompt.md` (the `<task>` XML grammar and
* `## Task N` legacy heading fallback), `gsd-core/templates/summary.md`
* (`## Files Created/Modified`, `## Deviations from Plan` / `**Found
* during:**` / `**Files modified:**`), `gsd-core/templates/state.md`
* (`## Current Position`), and `gsd-core/templates/roadmap.md` (`## Phases`
* checkbox list, `### Phase N: Name` headings) — never from
* `planning-inspect.cts`'s own parsing model. `planning.inspect` has no
* writer of its own (it is read-only), so the property-test generator in
* this file (`propertySchemaIsTotalOverDocumentShapedInputs`) is
* document-shaped from those same templates, not seeded from the module
* under test.
*
* `runGsdTools(['query', 'planning', 'inspect'], tmpDir)` is the invocation
* shape used throughout — array form, shell-bypassed, safe for hostile
* fixture values (CONTRIBUTING "CLI and command routing").
*/
const { test, describe } = require('node:test');
const assert = require('node:assert/strict');
const fs = require('node:fs');
const path = require('node:path');
const fc = require('fast-check');
const { createTempProject, createTempDir, cleanup, runGsdTools, toPosixPath } = require('./helpers.cjs');
const { generateSlugInternal } = require('../gsd-core/bin/lib/core-utils.cjs');
// ─── Fixture helpers ──────────────────────────────────────────────────────────
function planningDirOf(cwd) {
return path.join(cwd, '.planning');
}
function phasesDirOf(cwd) {
return path.join(planningDirOf(cwd), 'phases');
}
function phaseDirOf(cwd, token) {
return path.join(phasesDirOf(cwd), token);
}
function writeAbs(fullPath, content) {
fs.mkdirSync(path.dirname(fullPath), { recursive: true });
fs.writeFileSync(fullPath, content);
}
function writeFile(cwd, relPath, content) {
writeAbs(path.join(cwd, relPath), content);
}
function frontmatterDoc(frontmatterLines, bodyLines, eol) {
return ['---', ...frontmatterLines, '---', '', ...bodyLines].join(eol);
}
function writeRoadmap(cwd, lines, eol = '\n') {
writeFile(cwd, '.planning/ROADMAP.md', lines.join(eol));
}
function writeState(cwd, frontmatterLines, bodyLines = [], eol = '\n') {
writeFile(cwd, '.planning/STATE.md', frontmatterDoc(frontmatterLines, bodyLines, eol));
}
function writeRequirements(cwd, content) {
writeFile(cwd, '.planning/REQUIREMENTS.md', content);
}
function writePlanDoc(phaseDir, fileName, frontmatterLines, bodyLines, eol = '\n') {
writeAbs(path.join(phaseDir, fileName), frontmatterDoc(frontmatterLines, bodyLines, eol));
}
function writeSummaryDoc(phaseDir, fileName, frontmatterLines, bodyLines, eol = '\n') {
writeAbs(path.join(phaseDir, fileName), frontmatterDoc(frontmatterLines, bodyLines, eol));
}
function writeVerification(phaseDir, phaseToken, status, eol = '\n') {
writeAbs(path.join(phaseDir, `${phaseToken}-VERIFICATION.md`), ['---', `status: ${status}`, '---', ''].join(eol));
}
function writeUatDoc(phaseDir, phaseToken, bodyLines, eol = '\n') {
writeAbs(path.join(phaseDir, `${phaseToken}-UAT.md`), bodyLines.join(eol));
}
/**
* Slugify a phase name the same way `getPhaseDirFromPhaseId` (`src/phase-id.cts`)
* does. Routed through the canonical `generateSlugInternal` seam (issue #3987)
* instead of a hand-rolled copy: `getPhaseDirFromPhaseId` does not truncate,
* so `maxLen: null` is what makes the parity claim in this comment true —
* the prior copy also never transliterated non-Latin phase names, unlike the
* real seam it claims to match.
*/
function slugify(name) {
return generateSlugInternal(name, null) ?? '';
}
/**
* The on-disk directory token for phase `token`/`name` — zero-padded numeric
* prefix + slug (`"01-auth"`), matching `tests/planning-snapshot.test.cjs`'s
* fixtures and every real project (phases are never named by a bare `"1"`).
* Deliberately DIFFERENT from the bare numeric token the ROADMAP prose itself
* carries (`"Phase 1"`) — that mismatch (directory `phaseKeyFromDir`-keyed,
* ROADMAP `phaseKeyFromToken`-keyed) is exactly what a real project looks
* like, and exactly what `planning-inspect.cts`'s roadmap-checkbox lookup
* must reconcile through the phase-id owners rather than raw string equality.
*/
function slugPhaseDirName(token, name) {
return `${String(token).padStart(2, '0')}-${slugify(name)}`;
}
/**
* Declare one phase in STATE.md + ROADMAP.md (milestone window), matching
* `gsd-core/templates/roadmap.md`'s `### Phase N: Name` heading shape and
* `gsd-core/templates/state.md`'s frontmatter shape. The ROADMAP prose
* (heading + checkbox bullet) carries the BARE numeric token, exactly as real
* ROADMAP.md documents do — `buildRoadmapPhaseCheckboxesField` and the
* requirement-traceability parser both capture this bare form. The returned
* directory is the SLUGGED convention (`slugPhaseDirName`), matching real
* projects and `tests/planning-snapshot.test.cjs`'s own fixtures.
*/
function declarePhase(cwd, token, name, { checkedInPhaseList = false } = {}) {
writeState(cwd, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
const phaseListLine = checkedInPhaseList
? `- [x] **Phase ${token}: ${name}** - stub`
: `- [ ] **Phase ${token}: ${name}** - stub`;
writeRoadmap(cwd, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
phaseListLine,
'',
`### Phase ${token}: ${name}`,
'',
]);
// phaseDirOf only computes a path; only writeAbs creates directories, as a
// side effect of writing a file. A declared-but-absent directory silently
// produces an empty `phases[]`, so create it explicitly here.
const phaseDir = phaseDirOf(cwd, slugPhaseDirName(token, name));
fs.mkdirSync(phaseDir, { recursive: true });
return phaseDir;
}
/** A healthy two-phase project: both phases complete, requirements mapped. */
function buildHealthyFixture(cwd, eol = '\n') {
writeState(cwd, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0'], [], eol);
writeRoadmap(cwd, [
'## v1.0 Current 🚧',
'',
'### Phase 1: Foo',
'',
'### Phase 2: Bar',
'',
], eol);
writeRequirements(cwd, [
'# Requirements: Test',
'',
'## v1 Requirements',
'',
'- [x] **AUTH-01**: User can sign up',
'- [ ] **AUTH-02**: User can log in',
'',
'## Traceability',
'',
'| Requirement | Phase | Status |',
'|-------------|-------|--------|',
'| AUTH-01 | Phase 1 | Complete |',
'| AUTH-02 | Phase 2 | Pending |',
'',
].join(eol));
for (const [token, name] of [['1', 'foo'], ['2', 'bar']]) {
const phaseDir = phaseDirOf(cwd, slugPhaseDirName(token, name));
writePlanDoc(phaseDir, `${token}-01-PLAN.md`, ['wave: 1'], [
'<objective>',
`Ship ${name}`,
'</objective>',
'',
'<tasks>',
'',
'<task type="auto">',
` <name>Task 1: Build ${name}</name>`,
` <files>src/${name}.ts</files>`,
' <action>Build it</action>',
' <done>Done</done>',
'</task>',
'',
'</tasks>',
], eol);
writeSummaryDoc(phaseDir, `${token}-01-SUMMARY.md`, ['status: complete'], [
'# Summary',
'',
'## Files Created/Modified',
`- \`src/${name}.ts\` - ${name}`,
], eol);
writeVerification(phaseDir, token, 'passed', eol);
}
}
/** Recursive {relPath -> {size, mtimeMs}} snapshot of `.planning/`, for read-only proof. */
function snapshotPlanningTree(cwd) {
const root = planningDirOf(cwd);
const snap = {};
function walk(dir) {
let entries;
try {
entries = fs.readdirSync(dir, { withFileTypes: true });
} catch {
return;
}
for (const entry of entries) {
const full = path.join(dir, entry.name);
const rel = path.relative(root, full);
if (entry.isDirectory()) {
walk(full);
continue;
}
let stat;
try {
stat = fs.statSync(full);
} catch {
continue;
}
snap[rel] = { size: stat.size, mtimeMs: stat.mtimeMs };
}
}
if (fs.existsSync(root)) walk(root);
return snap;
}
/** JSON round-trip with the fixture's own absolute cwd replaced by a stable placeholder. */
function stripCwd(payload, cwd) {
const cwdPosix = toPosixPath(cwd);
const json = JSON.stringify(payload).split(cwdPosix).join('<CWD>').split(cwd).join('<CWD>');
return JSON.parse(json);
}
const EXPECTED_TOP_LEVEL_KEYS = [
'active', 'diagnostics', 'generated_from', 'milestone', 'orphan_phase_dirs',
'phases', 'progress', 'requirements', 'schema_version',
].sort();
const EXPECTED_PHASE_ROW_KEYS = [
'complete', 'dependencies', 'dir', 'goal', 'phase_id', 'plan_count', 'plans',
'roadmap_acceptance', 'scope', 'summary_count', 'uat', 'verification',
].sort();
function sortedKeys(obj) {
return Object.keys(obj).sort();
}
function runInspect(tmpDir, extraArgs = []) {
return runGsdTools(['query', 'planning', 'inspect', ...extraArgs], tmpDir);
}
function parseInspect(tmpDir, extraArgs = []) {
const result = runInspect(tmpDir, extraArgs);
assert.strictEqual(result.success, true, `planning inspect should succeed: ${result.error}`);
return JSON.parse(result.output);
}
function runInspectJsonError(tmpDir, extraArgs) {
const result = runGsdTools(['query', 'planning', 'inspect', ...extraArgs, '--json-errors'], tmpDir);
assert.strictEqual(result.success, false, `expected failure for args: ${extraArgs.join(' ')}`);
let parsed;
try {
parsed = JSON.parse(result.error);
} catch (e) {
throw new Error(`--json-errors must emit valid JSON on stderr; got: ${result.error}\nparse error: ${e.message}`);
}
assert.strictEqual(parsed.ok, false);
return parsed;
}
// ─── 1. Schema contract ─────────────────────────────────────────────────────────
describe('planning inspect — schema contract', () => {
test('emitsSchemaV1SnapshotForPopulatedProject', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const payload = JSON.parse(result.output);
assert.strictEqual(payload.schema_version, 1);
});
test('locksTopLevelSchemaKeySet', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(sortedKeys(payload), EXPECTED_TOP_LEVEL_KEYS);
});
test('schemaKeySetIsDecoupledFromPlanningSnapshotShape', (t) => {
// Row 4: the top-level key set is a FROZEN mapping this module owns, not
// a reflection of `PlanningSnapshot`'s own (additive, still-growing)
// shape. Proven by locking the map itself — EXPECTED_TOP_LEVEL_KEYS is a
// hand-authored constant in this file, not derived from the snapshot
// module — so a field added to `PlanningSnapshot` cannot silently widen
// what `planning.inspect` emits without this test also being edited.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const planningSnapshotLib = require('../gsd-core/bin/lib/planning-snapshot.cjs');
const snapshot = planningSnapshotLib.buildPlanningSnapshot(tmpDir);
// PlanningSnapshot's own key set is whatever it is (additive, churning) —
// asserted only to prove this test is exercising the real module, not a
// stub.
assert.ok(Object.keys(snapshot).length > 0);
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(sortedKeys(payload), EXPECTED_TOP_LEVEL_KEYS);
});
test('locksSchemaVersionConstantAgainstTheExportedModule', () => {
// Loading a module's exported runtime value, not source-grepping text.
const planningInspectLib = require('../gsd-core/bin/lib/planning-inspect.cjs');
assert.strictEqual(planningInspectLib.PLANNING_INSPECT_SCHEMA_VERSION, 1);
});
test('locksDiagnosticTaskStatusProvenanceAndAgreementEnums', () => {
const planningInspectLib = require('../gsd-core/bin/lib/planning-inspect.cjs');
const expectedDiagnosticKeys = [
'PLANNING_ROOT_ABSENT', 'ROADMAP_UNSCOPED', 'REQUIREMENTS_ABSENT', 'REQUIREMENTS_UNREADABLE',
'REQUIREMENT_DUPLICATE', 'REQUIREMENT_UNMAPPED', 'REQUIREMENT_PHASE_UNKNOWN',
'REQUIREMENT_COMPLETION_UNKNOWN', 'ORPHAN_PHASE_DIR', 'PHASE_SCOPE_DEGRADED',
'PLAN_UNREADABLE', 'SUMMARY_UNREADABLE', 'TASK_SHAPE_CHECKPOINT',
'TASK_CHANGED_FILES_PLAN_SCOPED', 'TASK_CHANGED_FILES_CONFLICTING',
'UAT_ABSENT', 'UAT_UNREADABLE', 'PERCENT_WITHHELD',
].sort();
assert.deepStrictEqual(sortedKeys(planningInspectLib.INSPECT_DIAGNOSTIC), expectedDiagnosticKeys);
assert.deepStrictEqual(sortedKeys(planningInspectLib.TASK_STATUS), ['DONE', 'PENDING', 'UNKNOWN'].sort());
assert.deepStrictEqual(sortedKeys(planningInspectLib.PROVENANCE), ['ABSENT', 'PLAN_SCOPED', 'TASK_SCOPED'].sort());
assert.deepStrictEqual(sortedKeys(planningInspectLib.AGREEMENT), ['AGREED', 'CONFLICTING', 'UNKNOWN'].sort());
});
test('dottedAndSpacedInvocationsAgree', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const spaced = parseInspect(tmpDir);
const dottedResult = runGsdTools(['query', 'planning.inspect'], tmpDir);
assert.strictEqual(dottedResult.success, true, `expected success: ${dottedResult.error}`);
const dotted = JSON.parse(dottedResult.output);
assert.deepStrictEqual(dotted, spaced);
});
});
// ─── 2. Dispatch / usage ──────────────────────────────────────────────────────
describe('planning inspect — dispatch and usage', () => {
test('rejectsPlanningFamilyWithNoSubcommand', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const result = runGsdTools(['query', 'planning'], tmpDir);
assert.strictEqual(result.success, false);
});
test('rejectsUnknownPlanningSubcommand', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const result = runGsdTools(['query', 'planning', 'bogus'], tmpDir);
assert.strictEqual(result.success, false);
});
test('rejectsStrayPositionalArgumentWithUsageReason', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const parsed = runInspectJsonError(tmpDir, ['extra']);
assert.strictEqual(parsed.reason, 'usage');
});
test('rejectsUnknownFlagWithUsageReason', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const parsed = runInspectJsonError(tmpDir, ['--nope']);
assert.strictEqual(parsed.reason, 'usage');
});
test('rejectsScopingFlagsNotSupportedInV1WithUsageReason', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const parsed = runInspectJsonError(tmpDir, ['--phase', '3']);
assert.strictEqual(parsed.reason, 'usage');
});
test('neverPrintsStackTraceOnUsageFailure', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const result = runInspect(tmpDir, ['extra']);
assert.strictEqual(result.success, false);
// Structural "did our own error envelope leak a raw stack" proof — the
// established repo pattern (tests/commands.test.cjs, tests/config-get-default.test.cjs).
assert.strictEqual(/\n\s*at\s/.test(result.error), false, `stderr must not carry a stack trace: ${result.error}`);
});
test('rejectsEmptyAndValuelessFlagForms', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 12: `--phase ""` and `--phase` (no value at all) both land on the
// same "planning inspect takes no arguments" usage rejection as any other
// unrecognized flag — v1 takes no flags at all, so neither form is
// special-cased into a different failure shape.
const emptyValue = runInspectJsonError(tmpDir, ['--phase', '']);
assert.strictEqual(emptyValue.reason, 'usage');
const noValue = runInspectJsonError(tmpDir, ['--phase']);
assert.strictEqual(noValue.reason, 'usage');
});
test('toleratesDuplicateGlobalFlag', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
// Row 13: `--raw --raw` — v1 takes no flags, so `--raw` (recognized by
// other query commands) is itself rejected by this command's own usage
// check. The row's contract is "no crash; deterministic", which this
// proves by running twice and asserting the two failures are identical.
const first = runInspect(tmpDir, ['--raw', '--raw']);
const second = runInspect(tmpDir, ['--raw', '--raw']);
assert.strictEqual(first.success, false);
assert.strictEqual(second.success, false);
assert.strictEqual(first.error, second.error);
assert.strictEqual(/\n\s*at\s/.test(first.error), false, `stderr must not carry a stack trace: ${first.error}`);
});
test('rejectsFlagShapedValueWithoutStackTrace', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 14: a value that itself looks like a flag (`--pick --weird`) must
// still fail as a plain usage error, never crash with a raw stack trace.
const result = runInspect(tmpDir, ['--pick', '--weird']);
assert.strictEqual(result.success, false);
assert.strictEqual(/\n\s*at\s/.test(result.error), false, `stderr must not carry a stack trace: ${result.error}`);
});
test('emitsTypedReasonCodesUnderJsonErrors', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 15: every usage/dispatch failure in rows 7-11 must carry a TYPED
// `reason` under `--json-errors` — never require a caller to regex the
// human `message` prose.
const noSubcommand = runGsdTools(['query', 'planning', '--json-errors'], tmpDir);
assert.strictEqual(noSubcommand.success, false);
assert.strictEqual(JSON.parse(noSubcommand.error).reason, 'sdk_unknown_command');
const unknownSubcommand = runGsdTools(['query', 'planning', 'bogus', '--json-errors'], tmpDir);
assert.strictEqual(unknownSubcommand.success, false);
assert.strictEqual(JSON.parse(unknownSubcommand.error).reason, 'sdk_unknown_command');
assert.strictEqual(runInspectJsonError(tmpDir, ['extra']).reason, 'usage');
assert.strictEqual(runInspectJsonError(tmpDir, ['--nope']).reason, 'usage');
assert.strictEqual(runInspectJsonError(tmpDir, ['--phase', '3']).reason, 'usage');
});
});
// ─── 3. Read-only proof ───────────────────────────────────────────────────────
describe('planning inspect — read-only proof', () => {
test('mutatesNothingUnderPlanningDirOnASuccessfulRun', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const before = snapshotPlanningTree(tmpDir);
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const after = snapshotPlanningTree(tmpDir);
assert.deepStrictEqual(after, before);
});
test('mutatesNothingEvenWhenAPlanDocumentIsUnreadable', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
// Directory-in-file-position (no chmod; root-proof, cross-platform):
// readDocument() sees `!stat.isFile()` and reports unreadable.
fs.mkdirSync(path.join(phaseDir, '1-01-PLAN.md'), { recursive: true });
const before = snapshotPlanningTree(tmpDir);
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const after = snapshotPlanningTree(tmpDir);
assert.deepStrictEqual(after, before);
});
});
// ─── 4. Degradation and scope ─────────────────────────────────────────────────
describe('planning inspect — degradation and scope', () => {
test('degradesCleanlyWithNoPlanningDirAtAll', (t) => {
const tmpDir = createTempDir();
t.after(() => cleanup(tmpDir));
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(sortedKeys(payload), EXPECTED_TOP_LEVEL_KEYS);
assert.ok(payload.diagnostics.some((d) => d.code === 'planning_root_absent'));
});
test('distinguishesAbsentRequirementsFileFromEmpty', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
// REQUIREMENTS.md deliberately not written.
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(payload.requirements, []);
assert.ok(payload.diagnostics.some((d) => d.code === 'requirements_absent'));
});
test('treatsAnEmptyRequirementsFileAsARealEmptyAnswerNotAnAbsentOne', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, '');
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(payload.requirements, []);
assert.ok(!payload.diagnostics.some((d) => d.code === 'requirements_absent'));
});
test('flagsCheckboxOnlyRequirementWithNoTraceabilityRowAsUnmapped', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [ ] **REQ-01**: Something to build',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
const row = payload.requirements.find((r) => r.id === 'REQ-01');
assert.ok(row, 'REQ-01 row must be present');
assert.deepStrictEqual(row.mappedPhases, []);
assert.ok(payload.diagnostics.some((d) => d.code === 'requirement_unmapped' && d.subject === 'REQ-01'));
});
test('carriesTheRequirementUnmappedCodeOnTheRowItselfForCorrelation', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [ ] **REQ-04**: Something to build',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
const row = payload.requirements.find((r) => r.id === 'REQ-04');
assert.ok(row, 'REQ-04 row must be present');
// Per-row diagnostics is a correlation convenience over the SAME global
// diagnostics array — never a second, independent answer.
assert.ok(row.diagnostics.includes('requirement_unmapped'));
assert.ok(payload.diagnostics.some((d) => d.code === 'requirement_unmapped' && d.subject === 'REQ-04'));
});
test('flagsRequirementMappedToAPhaseNotPresentOnDisk', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [ ] **REQ-02**: Something',
'',
'## Traceability',
'',
'| Requirement | Phase | Status |',
'|-------------|-------|--------|',
'| REQ-02 | Phase 9 | Pending |',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
const row = payload.requirements.find((r) => r.id === 'REQ-02');
assert.ok(row, 'REQ-02 row must be present');
assert.deepStrictEqual(row.mappedPhases, ['9']);
assert.ok(payload.diagnostics.some((d) => d.code === 'requirement_phase_unknown' && d.subject === 'REQ-02->9'));
});
test('flagsDuplicateRequirementIdNamingTheDuplicatedId', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [ ] **REQ-03**: First occurrence',
'- [ ] **REQ-03**: Second occurrence (duplicate)',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
assert.ok(payload.diagnostics.some((d) => d.code === 'requirement_duplicate' && d.subject === 'REQ-03'));
// Row 30: the SECOND occurrence, specifically — the first wins (its text
// survives), there is exactly one row (never two), and the row itself
// carries the correlation diagnostic naming the collision.
const matching = payload.requirements.filter((r) => r.id === 'REQ-03');
assert.strictEqual(matching.length, 1, 'a duplicate ID must produce exactly one row, not two');
assert.strictEqual(matching[0].text, 'First occurrence');
assert.ok(matching[0].diagnostics.includes('requirement_duplicate'));
});
test('reportsUndeclaredPhaseDirAsOrphanNotAsAPhase', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
fs.mkdirSync(phaseDirOf(tmpDir, '99-stray'), { recursive: true });
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(payload.orphan_phase_dirs, ['99-stray']);
assert.ok(!payload.phases.some((p) => p.dir === '99-stray'));
assert.ok(payload.diagnostics.some((d) => d.code === 'orphan_phase_dir' && d.subject === '99-stray'));
});
test('withholdsPercentWhenRoadmapAbsent', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 20: ROADMAP.md deliberately not written (STATE.md alone is not
// enough to scope a milestone). getMilestoneInfo's own absent-file path
// reports SCOPE.UNREADABLE, not COMPLETE.
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
const payload = parseInspect(tmpDir);
assert.notStrictEqual(payload.milestone.scope, 'complete');
assert.strictEqual(payload.progress.accepted_phases.percent, null);
});
test('doesNotInventAMilestoneVersionForFreeFormRoadmap', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning']);
writeRoadmap(tmpDir, [
'# Project Roadmap',
'',
'Some free-form notes with no version token anywhere in this document.',
'',
'### Phase 1: Foo',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Foo')), { recursive: true });
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.milestone.scope, 'unscoped');
assert.strictEqual(payload.milestone.version, null);
// The absence of any version evidence must never be filled with a
// plausible-looking default such as "v1.0".
assert.notStrictEqual(payload.milestone.version, 'v1.0');
});
test('reportsTruncatedIdentityForProseOnlyVersionToken', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning']);
writeRoadmap(tmpDir, [
'# Project Roadmap',
'',
'We are targeting v1.4 sometime this quarter.',
'',
'### Phase 1: Foo',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Foo')), { recursive: true });
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.milestone.scope, 'truncated');
assert.strictEqual(payload.milestone.version, 'v1.4');
assert.strictEqual(payload.milestone.name, null);
});
test('derivesNameFromTheHeadingsOwnVersionToken', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 23: STATE.md declares "v2.0"; ROADMAP's heading is "v2.0.1 —
// Portability" — the milestone name must derive from the heading's OWN
// version token ("Portability"), never the raw remainder text after a
// naive prefix strip (".1 — Portability").
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v2.0']);
writeRoadmap(tmpDir, [
'## v2.0.1 — Portability',
'',
'### Phase 1: Foo',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Foo')), { recursive: true });
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.milestone.name, 'Portability');
assert.ok(!payload.milestone.name.includes('.1'));
});
test('treatsWhitespaceOnlyRequirementsAsEmpty', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, ' \n\n\t\n');
const payload = parseInspect(tmpDir);
assert.deepStrictEqual(payload.requirements, []);
assert.ok(!payload.diagnostics.some((d) => d.code === 'requirements_absent'));
});
test('emitsUnknownForRequirementWithNoCheckbox', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
// Row 28: a Traceability-table-only requirement — present as a table row
// (`parseRequirements`'s pipe-table path), with NO checkbox bullet
// anywhere. `text` is a real non-answer (null, not ''); `complete` is
// `unknown`, never inferred `false`.
writeRequirements(tmpDir, [
'# Requirements',
'',
'## Traceability',
'',
'| Requirement | Phase | Status |',
'|-------------|-------|--------|',
'| REQ-05 | Phase 1 | Pending |',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
const row = payload.requirements.find((r) => r.id === 'REQ-05');
assert.ok(row, 'REQ-05 row must be present from the Traceability row alone');
assert.strictEqual(row.text, null);
assert.strictEqual(row.complete, 'unknown');
assert.ok(payload.diagnostics.some((d) => d.code === 'requirement_completion_unknown' && d.subject === 'REQ-05'));
});
test('acceptsPrefixAgnosticRequirementIds', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
// Row 31: the ID format is prefix-agnostic — AUTH-01 and INSP-04 are
// accepted exactly like REQ-01, sharing the SAME `[A-Z][A-Z0-9]*-...`
// pattern `parseRequirements` (gap-checker.cts) uses.
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [x] **AUTH-01**: User can sign up',
'- [ ] **INSP-04**: Inspection works',
'',
'## Traceability',
'',
'| Requirement | Phase | Status |',
'|-------------|-------|--------|',
'| AUTH-01 | Phase 1 | Complete |',
'| INSP-04 | Phase 1 | Pending |',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
const ids = payload.requirements.map((r) => r.id).sort();
assert.deepStrictEqual(ids, ['AUTH-01', 'INSP-04']);
const auth = payload.requirements.find((r) => r.id === 'AUTH-01');
assert.strictEqual(auth.complete, true);
const insp = payload.requirements.find((r) => r.id === 'INSP-04');
assert.strictEqual(insp.complete, false);
});
test('doesNotTreatTableSeparatorAsARequirementRow', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, [
'# Requirements',
'',
'## Traceability',
'',
'| Requirement | Phase | Status |',
'|-------------|-------|--------|',
'| REQ-06 | Phase 1 | Pending |',
'',
].join('\n'));
const payload = parseInspect(tmpDir);
// The header row and the `|---|---|` separator row must never themselves
// be parsed as requirement rows — exactly one requirement, REQ-06.
assert.strictEqual(payload.requirements.length, 1);
assert.strictEqual(payload.requirements[0].id, 'REQ-06');
});
});
// ─── 4b. Phase completion and plan liveness (§7.4/§7.5) ───────────────────────
describe('planning inspect — phase completion and plan liveness', () => {
test('treatsZeroPlanPhaseWithPassingVerificationAsComplete', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
// Row 34: no plans at all, but a passing VERIFICATION.md — §7.4 gates
// completion on verification alone; plan count is not a precondition.
writeVerification(phaseDir, '01', 'passed');
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.phases[0].complete, true);
assert.strictEqual(payload.phases[0].plan_count, 0);
});
test('treatsAbsentVerificationAsNotComplete', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
// Row 35: plans exist and are summarized, but no *-VERIFICATION.md was
// ever written — completion still requires the verification record.
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks></tasks>',
]);
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', [], [
'# Summary', '', '## Files Created/Modified',
]);
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.phases[0].complete, false);
assert.strictEqual(payload.phases[0].verification.status, 'missing');
});
test('excludesSupersededPlanFromLiveCounts', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'live one', '</objective>', '<tasks></tasks>',
]);
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', [], [
'# Summary', '', '## Files Created/Modified', '- `src/a.ts` - a',
]);
// Row 37: a plan whose frontmatter declares `status: superseded` — still
// LISTED in `plans`, but excluded from the live plan_count/summary_count.
writePlanDoc(phaseDir, '1-02-PLAN.md', ['status: superseded'], [
'<objective>', 'old one', '</objective>', '<tasks></tasks>',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.plan_count, 1, 'superseded plan must not inflate the live plan count');
const supersededRow = phase.plans.find((p) => p.id === '1-02');
assert.ok(supersededRow, 'the superseded plan is still listed');
assert.strictEqual(supersededRow.superseded, true);
const liveRow = phase.plans.find((p) => p.id === '1-01');
assert.strictEqual(liveRow.superseded, false);
});
test('countsProseRetiredPlanAsLive', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
// Row 38 (§7.5 GAP, deliberately characterized): a plan whose PROSE says
// "RETIRED" but carries no `status:` frontmatter key is still counted
// LIVE — the parser never compensates for a prose-only retirement claim.
writePlanDoc(phaseDir, '1-01-PLAN.md', [], [
'<objective>',
'RETIRED — this plan was abandoned, see prose below.',
'</objective>',
'<tasks></tasks>',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.plan_count, 1, 'a prose-only retirement claim must not remove the plan from the live count');
assert.strictEqual(phase.plans[0].superseded, false);
});
test('blockedSummaryIsNotACompletionRecord', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>',
'<task type="auto"><name>Task 1</name><files>src/a.ts</files><action>a</action><done>done</done></task>',
'</tasks>',
]);
// Row 39: SUMMARY declares `status: blocked` — a failure record, not a
// completion record. The plan/SUMMARY filename pairing still resolves
// (hasSummary true, provenance still parsed), but the phase-level
// summary_count must NOT count it.
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', ['status: blocked'], [
'# Summary', '', '## Files Created/Modified', '- `src/a.ts` - a',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.summary_count, 0, 'a blocked SUMMARY must not count as a completion record');
assert.strictEqual(phase.plans[0].hasSummary, true, 'filename pairing itself is unaffected by status');
});
test('haltedSummaryIsStillACompletionRecord', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>',
'<task type="auto"><name>Task 1</name><files>src/a.ts</files><action>a</action><done>done</done></task>',
'</tasks>',
]);
// Row 40: `status: halted` (#2830) — a DESIGNED stop still writes a
// completion record, unlike `status: blocked` above.
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', ['status: halted'], [
'# Summary', '', '## Files Created/Modified', '- `src/a.ts` - a',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.plan_count, 1);
assert.strictEqual(phase.summary_count, 1, 'a halted SUMMARY still counts as a completion record');
});
});
// ─── 5. Never-infer boundary ──────────────────────────────────────────────────
describe('planning inspect — never infers task-level file provenance', () => {
test('reportsAbsentProvenanceWhenATaskHasFilesButNoSummaryExists', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>', '',
'<task type="auto">',
' <name>Task 1: Do it</name>',
' <files>src/a.ts</files>',
' <action>Do it</action>',
' <done>Done</done>',
'</task>',
'',
'</tasks>',
]);
// No SUMMARY.md written.
const payload = parseInspect(tmpDir);
const task = payload.phases[0].plans[0].tasks[0];
assert.strictEqual(task.changedFiles, null);
assert.strictEqual(task.provenance, 'absent');
});
test('neverAttributesPlanScopedSummaryFilesToAnIndividualTask', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>', '',
'<task type="auto">',
' <name>Task 1: Build</name>',
' <files>src/a.ts</files>',
' <action>Build it</action>',
' <done>Done</done>',
'</task>',
'',
'</tasks>',
]);
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', ['status: complete'], [
'# Summary',
'',
'## Files Created/Modified',
'- `src/a.ts` - built',
]);
const payload = parseInspect(tmpDir);
const plan = payload.phases[0].plans[0];
const task = plan.tasks[0];
// Task-level half of the contract.
assert.strictEqual(task.changedFiles, null);
assert.strictEqual(task.provenance, 'plan_scoped');
assert.ok(payload.diagnostics.some((d) => d.code === 'task_changed_files_plan_scoped'));
// Plan-level half of the contract — the plan-scoped list still surfaces,
// just never spread across tasks.
assert.deepStrictEqual(plan.changedFiles, ['src/a.ts']);
});
test('attributesOnlyTheTaskTheSummaryDeviationBlockNames', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>', '',
'<task type="auto">',
' <name>Task 1: Build</name>',
' <files>src/a.ts</files>',
' <action>Build it</action>',
' <done>Done</done>',
'</task>',
'',
'<task type="auto">',
' <name>Task 2: Fix</name>',
' <files>src/b.ts, src/c.ts</files>',
' <action>Fix it</action>',
' <done>Done</done>',
'</task>',
'',
'</tasks>',
]);
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', ['status: complete'], [
'# Summary',
'',
'## Files Created/Modified',
'- `src/a.ts` - built',
'',
'## Deviations from Plan',
'',
'### Auto-fixed Issues',
'',
'**1. Some fix**',
'- **Found during:** Task 2 (Fix)',
'- **Issue:** something',
'- **Files modified:** src/b.ts, src/c.ts',
'- **Verification:** tests pass',
]);
const payload = parseInspect(tmpDir);
const tasks = payload.phases[0].plans[0].tasks;
assert.deepStrictEqual(tasks[1].changedFiles, ['src/b.ts', 'src/c.ts']);
assert.strictEqual(tasks[1].provenance, 'task_scoped');
// Task 1 is untouched by the deviation block naming Task 2.
assert.strictEqual(tasks[0].changedFiles, null);
assert.strictEqual(tasks[0].provenance, 'plan_scoped');
});
test('emitsConflictingProvenanceWithoutReconcilingPlannedAndChangedFiles', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>', '',
'<task type="auto">',
' <name>Task 1: Build</name>',
' <files>src/a.ts</files>',
' <action>Build it</action>',
' <done>Done</done>',
'</task>',
'',
'</tasks>',
]);
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', ['status: complete'], [
'# Summary',
'',
'## Deviations from Plan',
'',
'### Auto-fixed Issues',
'',
'**1. Scope change**',
'- **Found during:** Task 1 (Build)',
'- **Issue:** plan undershot',
'- **Files modified:** src/b.ts',
'- **Verification:** tests pass',
]);
const payload = parseInspect(tmpDir);
const task = payload.phases[0].plans[0].tasks[0];
assert.strictEqual(task.agreement, 'conflicting');
assert.deepStrictEqual(task.plannedFiles, ['src/a.ts']);
assert.deepStrictEqual(task.changedFiles, ['src/b.ts']);
assert.ok(payload.diagnostics.some((d) => d.code === 'task_changed_files_conflicting'));
});
});
// ─── 6. Evidence kept separate ─────────────────────────────────────────────────
describe('planning inspect — evidence kept separate, never folded', () => {
test('uatAbsenceDoesNotAffectAcceptedPhases', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 50: no UAT.md at all — `uat: []` plus a `uat_absent` diagnostic,
// and — the load-bearing half of the row — phase acceptance is entirely
// unaffected by UAT's absence (UAT never gates `accepted_phases`).
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '01', 'passed');
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.deepStrictEqual(phase.uat.unresolved, []);
assert.ok(payload.diagnostics.some((d) => d.code === 'uat_absent'));
assert.strictEqual(phase.complete, true);
assert.strictEqual(payload.progress.accepted_phases.percent, 100);
});
test('keepsUnresolvedUatAndPassingVerificationSeparateWithNoCombinedVerdict', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDoc(phaseDir, '1', [
'### 1. Check something',
'expected: it works',
'result: pending',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.complete, true);
assert.ok(phase.uat.unresolved.length > 0);
// No combined verdict field exists alongside the raw evidence sources.
assert.deepStrictEqual(sortedKeys(phase), EXPECTED_PHASE_ROW_KEYS);
});
test('uatParseGapNeverClaimsCompleteScopeWithEmptyUnresolved', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Security review finding 1 (#3707 second surface): a `### N.` test block
// with no `result:` line is a genuine parse gap — the audit-uat side
// (`cmdAuditUat`) already flags this file as `parse_gap: true`. Before
// the fix, planning-inspect's `buildUatRows` called `parseUatItems`
// (which silently drops a gap-only heading from BOTH items and any
// scope signal), so this exact file was reported as `uat: { unresolved:
// [], scope: 'complete' }` — an affirmative completeness claim over rows
// it never actually derived.
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDoc(phaseDir, '1', [
'### 1. Check something',
'expected: it works',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.deepStrictEqual(phase.uat.unresolved, []);
assert.notStrictEqual(phase.uat.scope, 'complete');
assert.strictEqual(phase.uat.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'uat_unreadable' && d.subject.includes('1-UAT.md')));
});
// ─── #3078 round-8: no frontmatter kill switch over the parse-gap detector ──
//
// `buildUatRows` used to carry `&& status !== 'complete'` alongside its
// `headingsSeen > 0` check, long after `cmdAuditUat` dropped the identical
// guard. Nothing pinned this arm: the fixture above writes no frontmatter
// `status` at all, so the clause was unreachable from the test suite and one
// word of frontmatter silently switched off the detector on the real CLI.
// These four rows are that pin. Each asserts the SCOPE VALUE and the
// DIAGNOSTIC explicitly — "something came back" is what let this survive.
/** A UAT document with `status:` frontmatter, for the kill-switch rows below. */
function writeUatDocWithStatus(phaseDir, phaseToken, status, bodyLines) {
writeUatDoc(phaseDir, phaseToken, ['---', `status: ${status}`, '---', '', ...bodyLines]);
}
/**
* A closed fence that OPENS after test 1 and CLOSES after test 2, hiding
* test 2's `result: blocked` from the heading tokenizer entirely
* (`src/uat.cts`'s fence-straddle case — the row is absent from the token
* stream, not merely unparseable).
*/
const FENCE_STRADDLED_BLOCKED_BODY = [
'### 1. Alpha',
'expected: a',
'result: pass',
'',
'```',
'### 2. Beta',
'expected: b',
'result: blocked',
'```',
'',
];
for (const status of ['complete', 'in_progress']) {
test(`reportsAFenceStraddledBlockedRowOnAUatFileMarkedStatus_${status}`, (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDocWithStatus(phaseDir, '1', status, FENCE_STRADDLED_BLOCKED_BODY);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
// The hidden row yields no item — it was never in the token stream — so
// the DIAGNOSTIC is the only channel that reports it. Under the old
// guard, `status: complete` produced `scope: "complete"` with ZERO
// diagnostics: an affirmative completeness claim over a `blocked` row
// the tool never read.
assert.strictEqual(phase.uat.scope, 'truncated');
assert.deepStrictEqual(phase.uat.unresolved, []);
assert.ok(payload.diagnostics.some((d) => d.code === 'uat_unreadable' && d.subject.includes('1-UAT.md')));
});
test(`claimsCompleteUatScopeWithNoDiagnosticWhenEveryRowParsesAndPassesAtStatus_${status}`, (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDocWithStatus(phaseDir, '1', status, [
'### 1. Alpha',
'expected: a',
'result: pass',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
// The other half of the pin: removing the status guard must NOT make
// every file noisy. A file with nothing unread still claims `complete`
// and raises nothing — otherwise the detector would be uninformative.
assert.strictEqual(phase.uat.scope, 'complete');
assert.deepStrictEqual(phase.uat.unresolved, []);
assert.strictEqual(payload.diagnostics.filter((d) => d.code === 'uat_unreadable').length, 0);
});
}
// ─── #3078 round-8 REVERTED (fix/3707-fold-shield-revert): the shield is
// GONE — a fence-suppression shortfall now degrades the fold exactly like
// every other UAT parse gap. `phase.uat.scope` (this phase's own reported
// evidence quality) and `phase.scope` (the `worstScope` fold that gates
// `phase_scope_degraded` and — via `progress.*` — the milestone's
// percentages) remain two SEPARATE fields with two separate meanings, but
// as of this revert they no longer disagree on a shortfall-only gap: both
// report degraded evidence. `src/uat.cts` still counts `shortfallBlocks` as
// its own documented ACCEPTED-OVER-REPORT subset of `headingsSeen`, but
// `buildUatRows` no longer reads that subset when deciding `foldScope` — see
// `src/planning-inspect.cts`'s `buildUatRows` for the current (unconditional)
// rule: `headingsSeen > 0` alone sets BOTH `scope` and `foldScope` to
// TRUNCATED, with no `headingsSeen > shortfallBlocks` comparison left to
// exempt anything.
//
// Boundary matrix (`headingsSeen`, `shortfallBlocks`):
// (0, 0) -> no diagnostic, fold COMPLETE — unchanged
// (1, 1) -> fold TRUNCATED — THE CHANGE (below)
// (1, 0) -> fold TRUNCATED — unchanged
// (2, 1) -> fold TRUNCATED — unchanged
test('shortfallAloneDegradesTheFoldAndWithholdsThePercentage', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
// A `## Notes` section documenting the row format inside a CLOSED fence —
// literal digits, so `TEST_HEADING_LINE_RE` counts it and the shortfall
// scan fires: headingsSeen === 1, shortfallBlocks === 1 (the boundary
// case the removed `headingsSeen > shortfallBlocks` comparison used to
// treat specially — post-revert there is no comparison left, so this is
// ordinary `headingsSeen > 0`).
writeUatDocWithStatus(phaseDir, '1', 'complete', [
'# UAT',
'',
'## Notes',
'',
'How to write a row:',
'',
'```',
'### 1. Example Row',
'expected: x',
'result: pass',
'```',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
// Post-revert: the fold now degrades on a shortfall-only gap, and the
// milestone percentage is withheld — the exact behavior #3707's shield
// used to suppress.
assert.strictEqual(phase.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'phase_scope_degraded' && d.subject === phase.dir));
assert.strictEqual(payload.progress.accepted_phases.percent, null);
assert.ok(payload.diagnostics.some((d) => d.code === 'percent_withheld'));
});
test('shortfallAloneStillReportsTruncatedUatScopeAndTheUnreadableDiagnostic', (t) => {
// CONTROL: the per-phase `uat.scope` reporting and the `uat_unreadable`
// diagnostic are untouched by the shield revert — they already went
// TRUNCATED for every gap, shortfall included. Only the FOLD (asserted in
// the row above) changed.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDocWithStatus(phaseDir, '1', 'complete', [
'# UAT',
'',
'## Notes',
'',
'How to write a row:',
'',
'```',
'### 1. Example Row',
'expected: x',
'result: pass',
'```',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.uat.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'uat_unreadable' && d.subject.includes('1-UAT.md')));
});
test('aNonShortfallParseGapStillDegradesThePhaseAndWithholdsThePercentage', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
// A column-0 `### N.` block with no `result:` line — visible to the
// tokenizer, so it is NOT a shortfall. No accepted-over-report story
// exists for it, so the teeth stay on.
writeUatDocWithStatus(phaseDir, '1', 'complete', [
'### 1. Alpha',
'expected: a',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.uat.scope, 'truncated');
assert.strictEqual(phase.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'phase_scope_degraded'));
assert.strictEqual(payload.progress.accepted_phases.percent, null);
});
test('anUnreadableUatFileStillDegradesThePhaseAndWithholdsThePercentage', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
// The pre-existing genuinely-truncated derivation: the UAT path exists and
// is selected, but reading it yields nothing. A DIRECTORY at the file path
// makes the read fail deterministically on every OS and as root — no mode
// bits, which root bypasses (CLAUDE.md, IO-failure injection).
fs.mkdirSync(path.join(phaseDir, '1-UAT.md'));
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.uat.scope, 'truncated');
assert.strictEqual(phase.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'uat_unreadable' && d.subject.includes('1-UAT.md')));
assert.ok(payload.diagnostics.some((d) => d.code === 'phase_scope_degraded'));
assert.strictEqual(payload.progress.accepted_phases.percent, null);
});
test('aPhaseWithNoUatGapAtAllStillPublishesThePercentage', (t) => {
// CONTROL for the catastrophic-revert failure mode: a revert that sets
// `foldScope = SCOPE.TRUNCATED` unconditionally, OUTSIDE the
// `headingsSeen > 0` branch, would withhold every percentage in the
// project — including this phase, which has no gap whatsoever.
// Boundary case: headingsSeen === 0, shortfallBlocks === 0.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDocWithStatus(phaseDir, '1', 'complete', [
'### 1. Alpha',
'expected: a',
'result: pass',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.uat.scope, 'complete');
assert.strictEqual(phase.scope, 'complete');
assert.strictEqual(payload.diagnostics.filter((d) => d.code === 'phase_scope_degraded').length, 0);
assert.strictEqual(payload.diagnostics.filter((d) => d.code === 'uat_unreadable').length, 0);
assert.strictEqual(payload.progress.accepted_phases.percent, 100);
});
test('aMixedShortfallAndGenuineParseGapPhaseDegradesTheFold', (t) => {
// CONTROL: a file carrying BOTH a shortfall block and a genuine
// (non-shortfall) parse gap in the same document — headingsSeen === 2,
// shortfallBlocks === 1. `headingsSeen > shortfallBlocks` was already true
// under the OLD shielded rule (2 > 1), so this row degraded the fold
// before the revert too; it stays green throughout.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDocWithStatus(phaseDir, '1', 'complete', [
// A genuine gap: a column-0 `### N.` block with no `result:` line.
'### 1. Alpha',
'expected: a',
'',
// A shortfall: a closed-fence documentation sample with literal digits.
'## Notes',
'',
'```',
'### 2. Example Row',
'expected: x',
'result: pass',
'```',
'',
]);
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.uat.scope, 'truncated');
assert.strictEqual(phase.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'phase_scope_degraded' && d.subject === phase.dir));
assert.strictEqual(payload.progress.accepted_phases.percent, null);
});
// ─── #3707-CR security review MEDIUM ────────────────────────────────────────
//
// A lone CR (`String.fromCharCode(13)`, no paired LF) is a CommonMark line
// ending — a document using it renders as separate lines to a human reader.
// `src/uat.cts` now normalizes every line ending at parse ingress
// (`normalizeLineEndings`), so the row is no longer hidden: it is parsed and
// surfaces as a visible outstanding `uat.unresolved` item, exactly like its
// LF twin. See `tests/uat.test.cjs`'s "#3707-CR" describe block for the
// parser-level pin this end-to-end case is downstream of.
//
// This test does NOT assert `percent === null` (an earlier version of this
// test did, and was wrong — reasoning from the PRE-fix symptom instead of
// the post-fix behavior). A surfaced `result: blocked` row is VISIBLE
// outstanding work, not unreadable evidence, and this module's pinned
// invariant is that visible outstanding UAT work deliberately does NOT
// withhold percentages or degrade scope — only genuinely UNREADABLE
// evidence does (`keepsUnresolvedUatAndPassingVerificationSeparateWithNoCombinedVerdict`,
// `uatAbsenceDoesNotAffectAcceptedPhases`, "evidence kept separate, never
// folded"). Do not "fix" this test back to `percent === null` — the
// asymmetry it once demanded would be the bug, not the fix.
//
// The contract that actually matters, and is much harder to satisfy
// accidentally, is PARITY: a line-ending convention must not change what
// the audit reports. Both documents below are derived from ONE source
// string, differing only in which separator carries the line break, so the
// two fixtures cannot drift apart under later editing.
test('lineEndingConventionDoesNotChangeUatAuditOutputLoneCrMatchesLf', () => {
const LF = '\n';
const CR = String.fromCharCode(13);
const sourceLines = [
'---',
'status: complete',
'---',
'',
'### 1. Alpha',
'expected: a',
'result: pass',
'',
'Notes.',
'### 2. Beta',
'expected: the export works',
'result: blocked',
'',
];
function runWithEol(eol) {
const tmpDir = createTempProject();
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeUatDoc(phaseDir, '1', sourceLines, eol);
const payload = parseInspect(tmpDir);
cleanup(tmpDir);
return payload;
}
const lfPayload = runWithEol(LF);
const crPayload = runWithEol(CR);
const lfPhase = lfPayload.phases[0];
const crPhase = crPayload.phases[0];
const describeAll = () => JSON.stringify({ lf: lfPhase, cr: crPhase }, null, 2);
const rowIdentity = (row) => ({ test: row.test, name: row.name, result: row.result });
const diagnosticCodes = (payload) => [...new Set(payload.diagnostics.map((d) => d.code))].sort();
assert.strictEqual(crPhase.uat.scope, lfPhase.uat.scope, describeAll());
assert.strictEqual(crPhase.scope, lfPhase.scope, describeAll());
assert.strictEqual(
crPayload.progress.accepted_phases.percent,
lfPayload.progress.accepted_phases.percent,
describeAll(),
);
assert.deepStrictEqual(diagnosticCodes(crPayload), diagnosticCodes(lfPayload), describeAll());
// LOAD-BEARING (#3707-CR MINOR 2): with normalization stripped from
// src/uat.cts, `uat.scope`, `phase.scope`, `accepted_phases.percent`, and
// `diagnosticCodes` above are ALL identical between the lone-CR and LF
// payloads even while the bug is present — a lone-CR document degrades
// scope to 'truncated' on BOTH sides identically (the CR document simply
// fails to parse either row, LF parses both), so those four assertions
// pass regardless of whether the CR fix exists. The ONLY assertion below
// that actually discriminates the fix from the bug is the
// `uat.unresolved` row-identity `deepStrictEqual`: pre-fix, `crPhase.uat.
// unresolved` is `[]` while `lfPhase.uat.unresolved` contains the "Beta"
// row, so this is the one comparison that fails without the fix. Do NOT
// remove this assertion as "redundant" with the four above — removing it
// makes this whole test vacuously green under the pre-fix behavior.
assert.deepStrictEqual(
crPhase.uat.unresolved.map(rowIdentity),
lfPhase.uat.unresolved.map(rowIdentity),
describeAll(),
);
// Sanity: the row is genuinely surfaced on both sides, not vacuously
// absent from both (which would make the equality checks above trivially
// pass without proving anything).
assert.strictEqual(lfPhase.uat.scope, 'complete', describeAll());
assert.strictEqual(lfPayload.progress.accepted_phases.percent, 100, describeAll());
assert.ok(lfPhase.uat.unresolved.some((r) => r.name === 'Beta' && r.result === 'blocked'), describeAll());
});
// ─── #3707-CR follow-up MINOR 1 ─────────────────────────────────────────────
//
// A lone-CR VERIFICATION.md with `status: passed` was read by
// `readVerificationStatus` (src/verification.cts) as `status: "missing"` —
// `extractFrontmatter`'s byte-0 `---\n` / `---\r\n` fence check never
// matches a lone-CR `---\r`, so the frontmatter block was invisible and the
// completed verification was reported as though the step never ran.
// Under-reports rather than over-reports (fail-safe direction), but the
// same root cause as the false-clean class fixed above: a line-ending
// convention must not change what `planning.inspect` reports.
test('loneCrVerificationStatusPassedIsNotReportedAsMissing', () => {
const tmpDir = createTempProject();
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
const CR = String.fromCharCode(13);
writeVerification(phaseDir, '1', 'passed', CR);
const payload = parseInspect(tmpDir);
cleanup(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.verification.status, 'passed',
`lone-CR VERIFICATION.md with status: passed must not report as missing: ${JSON.stringify(phase.verification)}`);
});
// ─── Multi-file degrade ─────────────────────────────────────────────────────
//
// Two files scope to the SAME phase: one carries a shortfall-only gap, one
// is entirely clean. The fold must degrade when EITHER file order is used —
// `fs.readdirSync` order is deterministically controlled via method
// monkeypatching (never mode bits; real directory order is OS/filesystem-
// dependent and would make this a flaky race), per CLAUDE.md's
// cross-platform IO-failure-injection rule.
//
// These two variants do NOT test file-order independence as a guarantee:
// `foldScope` in `buildUatRows` is monotonic (it is only ever set to
// `SCOPE.TRUNCATED`, never reset back to `SCOPE.COMPLETE`), so which file is
// visited first is structurally irrelevant to the current implementation,
// not something this test asserts. What the `shortfallFileFirst` variant
// DOES incidentally guard is a future regression that adds a reset path
// (e.g. code that sets `foldScope` back to COMPLETE upon encountering a
// later clean file) — running the shortfall file first and the clean file
// second is exactly the ordering such a bug would need to slip through.
function writeCustomUatFile(phaseDir, fileName, status, bodyLines) {
writeAbs(path.join(phaseDir, fileName), ['---', `status: ${status}`, '---', '', ...bodyLines].join('\n'));
}
const CLEAN_UAT_BODY = ['### 1. Alpha', 'expected: a', 'result: pass', ''];
const SHORTFALL_UAT_BODY = [
'# UAT', '', '## Notes', '', 'How to write a row:', '',
'```', '### 1. Example Row', 'expected: x', 'result: pass', '```', '',
];
for (const [label, order] of [
['cleanFileFirst', ['1-UAT-clean.md', '1-UAT-shortfall.md']],
['shortfallFileFirst', ['1-UAT-shortfall.md', '1-UAT-clean.md']],
]) {
test(`multiFileUatDegradesFoldWhenAnyFileHasShortfallOnlyGap_${label}`, (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '1', 'passed');
writeCustomUatFile(phaseDir, '1-UAT-clean.md', 'complete', CLEAN_UAT_BODY);
writeCustomUatFile(phaseDir, '1-UAT-shortfall.md', 'complete', SHORTFALL_UAT_BODY);
const planningInspectLib = require('../gsd-core/bin/lib/planning-inspect.cjs');
const originalReaddirSync = fs.readdirSync;
t.mock.method(fs, 'readdirSync', function mockedReaddirSync(target, ...rest) {
const result = originalReaddirSync.call(this, target, ...rest);
if (target === phaseDir) {
const others = result.filter((f) => f !== '1-UAT-clean.md' && f !== '1-UAT-shortfall.md');
return [...order, ...others];
}
return result;
});
const payload = planningInspectLib.buildPlanningInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'phase_scope_degraded' && d.subject === phase.dir));
assert.strictEqual(payload.progress.accepted_phases.percent, null);
});
}
test('roadmapAcceptanceIsNeverAuthoritativeOnAnyPhaseRow', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const payload = parseInspect(tmpDir);
assert.ok(payload.phases.length > 0);
for (const phase of payload.phases) {
assert.strictEqual(phase.roadmap_acceptance.authoritative, false);
}
});
test('roadmapCheckboxNeverOverridesDiskCompletion', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Checkbox ticked in ROADMAP, but no passing VERIFICATION on disk.
declarePhase(tmpDir, '1', 'Foo', { checkedInPhaseList: true });
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.complete, false);
assert.strictEqual(phase.roadmap_acceptance.checkbox, true);
});
test('reports a ticked ROADMAP checkbox for a slugged phase directory', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// ROADMAP prose carries the BARE numeric token ("Phase 1"/"Phase 2"/
// "Phase 3"), while the phase directories are the SLUGGED on-disk
// convention ("01-auth" etc) — the real-world mismatch that made
// `roadmap_acceptance.checkbox` always null: comparing
// `checkboxes["1"]` against `phase.dir === "01-auth"` by raw string
// equality never matches. Three phases distinguish all three checkbox
// states so none of them collapses into another: ticked (true), unticked
// (false), and no checkbox bullet at all (null, NOT false).
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
'- [x] **Phase 1: Auth** - stub',
'- [ ] **Phase 2: Billing** - stub',
'',
'### Phase 1: Auth',
'',
'### Phase 2: Billing',
'',
'### Phase 3: Reports',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Auth')), { recursive: true });
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('2', 'Billing')), { recursive: true });
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('3', 'Reports')), { recursive: true });
const payload = parseInspect(tmpDir);
const auth = payload.phases.find((p) => p.dir === '01-auth');
const billing = payload.phases.find((p) => p.dir === '02-billing');
const reports = payload.phases.find((p) => p.dir === '03-reports');
assert.ok(auth, 'slugged phase directory 01-auth must be present as a phase row');
assert.ok(billing, 'slugged phase directory 02-billing must be present as a phase row');
assert.ok(reports, 'slugged phase directory 03-reports must be present as a phase row');
assert.strictEqual(auth.roadmap_acceptance.checkbox, true);
assert.strictEqual(auth.roadmap_acceptance.authoritative, false);
assert.strictEqual(billing.roadmap_acceptance.checkbox, false);
assert.strictEqual(billing.roadmap_acceptance.authoritative, false);
// Phase 3 has no checkbox bullet under `## Phases` at all — `null` (no
// evidence), never collapsed into `false` (unticked evidence).
assert.strictEqual(reports.roadmap_acceptance.checkbox, null);
assert.strictEqual(reports.roadmap_acceptance.authoritative, false);
});
});
// ─── 6b. Per-phase goal / dependency evidence (#2790) ─────────────────────────
describe('planning inspect — per-phase goal and dependency evidence', () => {
test('reportsThePhaseHeadingProseAsGoalWithCompleteScope', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
'- [ ] **Phase 1: Auth** - stub',
'',
'### Phase 1: Auth',
'',
'Build authentication.',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Auth')), { recursive: true });
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.goal.value, 'Build authentication.');
assert.strictEqual(phase.goal.scope, 'complete');
});
test('reportsDependsOnPhaseTokensAsAStringArray', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
'- [ ] **Phase 1: Auth** - stub',
'- [ ] **Phase 2: Billing** - stub',
'',
'### Phase 1: Auth',
'',
'### Phase 2: Billing',
'',
'Charge the customer.',
'',
'**Depends on:** Phase 1',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Auth')), { recursive: true });
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('2', 'Billing')), { recursive: true });
const payload = parseInspect(tmpDir);
const billing = payload.phases.find((p) => p.dir === '02-billing');
assert.ok(billing, '02-billing phase row must be present');
assert.ok(billing.dependencies.value.includes('1'));
assert.strictEqual(billing.dependencies.scope, 'complete');
});
test('reportsNoDependsOnLineAsAnEmptyArrayNotADegradedScope', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
// Absent is a real answer, not a failure — same scope as a found section
// that simply carries no dependency line.
assert.deepStrictEqual(phase.dependencies.value, []);
assert.strictEqual(phase.dependencies.scope, 'complete');
});
test('excludesTheDependsOnAnnotationFromGoalEvenThoughItIsSurfacedSeparately', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
'- [ ] **Phase 1: Auth** - stub',
'',
'### Phase 1: Auth',
'',
'Build authentication.',
'',
'**Depends on:** Phase 1',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Auth')), { recursive: true });
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
// The annotation is data this payload already surfaces via
// `dependencies` — asserted together so it appears in exactly one place.
assert.strictEqual(phase.goal.value, 'Build authentication.');
assert.ok(!phase.goal.value.includes('Depends on'));
assert.ok(phase.dependencies.value.includes('1'));
});
test('excludesThePlansChecklistFromGoalProse', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
'- [ ] **Phase 1: Auth** - stub',
'',
'### Phase 1: Auth',
'',
'Build authentication.',
'',
'Plans:',
'- [ ] 01-PLAN.md',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Auth')), { recursive: true });
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
assert.strictEqual(phase.goal.value, 'Build authentication.');
assert.ok(!phase.goal.value.includes('Plans:'));
assert.ok(!phase.goal.value.includes('- [ ]'));
});
test('reportsNullGoalWithCompleteScopeWhenTheSectionIsPureMetadata', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
'- [ ] **Phase 1: Auth** - stub',
'',
'### Phase 1: Auth',
'',
'**Depends on:** Phase 1',
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', 'Auth')), { recursive: true });
const payload = parseInspect(tmpDir);
const phase = payload.phases[0];
// A section that is pure metadata genuinely has no goal prose — a real
// answer, not a failed read: `complete` still means "the section was
// found", never "prose was found".
assert.strictEqual(phase.goal.value, null);
assert.strictEqual(phase.goal.scope, 'complete');
});
});
// ─── 7. Percent withholding ───────────────────────────────────────────────────
describe('planning inspect — percent withholding', () => {
test('percentIsAnIntegerZeroToHundredForAHealthyProject', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const payload = parseInspect(tmpDir);
const percent = payload.progress.accepted_phases.percent;
assert.ok(Number.isInteger(percent) && percent >= 0 && percent <= 100, `got ${percent}`);
});
test('emitsZeroPercentNotNullNotHundredForAZeroPhaseButReadableProject', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, ['## v1.0 Current 🚧', '']);
// No phase directories at all.
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.progress.accepted_phases.percent, 0);
});
test('withholdsPercentAndFlagsItWhenThePhasesDirectoryIsUnreadable', (t) => {
// Fault-injection level (CONTRIBUTING QA matrix "Integration + mock.method"):
// mock.method cannot reach a spawned CLI subprocess, so this row calls the
// built module directly — the same pattern tests/planning-snapshot.test.cjs
// uses for its own readdirSync fault-injection rows.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const planningInspectLib = require('../gsd-core/bin/lib/planning-inspect.cjs');
const phasesDir = phasesDirOf(tmpDir);
const originalReaddirSync = fs.readdirSync;
t.mock.method(fs, 'readdirSync', function mockedReaddirSync(target, ...rest) {
if (target === phasesDir) {
const err = new Error(`EACCES: permission denied, scandir '${phasesDir}'`);
err.code = 'EACCES';
throw err;
}
return originalReaddirSync.call(this, target, ...rest);
});
const payload = planningInspectLib.buildPlanningInspect(tmpDir);
assert.strictEqual(payload.progress.accepted_phases.percent, null);
assert.ok(payload.diagnostics.some((d) => d.code === 'percent_withheld'));
});
});
// ─── 7b. Per-document/per-phase fault isolation (D33/D34) and combinations ───
describe('planning inspect — fault isolation and combinations', () => {
test('reportsTruncatedScopeForUnreadableNestedPlansDir', (t) => {
// Row 53: the phase directory itself is readable, but its NESTED
// `plans/` subdirectory is not — scanPhasePlans (plan-scan.cts) reports
// this as TRUNCATED, never COMPLETE-with-zero, and that folds into the
// phase row's own `scope`.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
const nestedDir = path.join(phaseDir, 'plans');
fs.mkdirSync(nestedDir, { recursive: true });
writeAbs(path.join(nestedDir, 'PLAN-01-x.md'), ['<objective>', 'a', '</objective>', '<tasks></tasks>'].join('\n'));
const planningInspectLib = require('../gsd-core/bin/lib/planning-inspect.cjs');
const originalReaddirSync = fs.readdirSync;
t.mock.method(fs, 'readdirSync', function mockedReaddirSync(target, ...rest) {
if (target === nestedDir) {
const err = new Error(`EACCES: permission denied, scandir '${nestedDir}'`);
err.code = 'EACCES';
throw err;
}
return originalReaddirSync.call(this, target, ...rest);
});
const payload = planningInspectLib.buildPlanningInspect(tmpDir);
assert.strictEqual(payload.phases[0].scope, 'truncated');
assert.ok(payload.diagnostics.some((d) => d.code === 'phase_scope_degraded'));
});
test('isolatesAnUnreadablePlanFromItsSiblings', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', [], ['<objective>', 'good one', '</objective>', '<tasks></tasks>']);
// Row 54: a SECOND plan that cannot be read (directory-in-file-position —
// no chmod, root-proof). Its own row must degrade; the sibling plan's
// row must be entirely unaffected.
fs.mkdirSync(path.join(phaseDir, '1-02-PLAN.md'), { recursive: true });
const payload = parseInspect(tmpDir);
const rows = payload.phases[0].plans;
const good = rows.find((p) => p.id === '1-01');
const bad = rows.find((p) => p.id === '1-02');
assert.strictEqual(good.scope, 'complete');
assert.strictEqual(good.objective, 'good one');
assert.strictEqual(bad.scope, 'unreadable');
assert.strictEqual(bad.objective, null);
assert.ok(payload.diagnostics.some((d) => d.code === 'plan_unreadable' && d.subject === '1-02-PLAN.md'));
});
test('treatsDirectoryInPlanPositionAsUnreadable', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', [], ['<objective>', 'good one', '</objective>', '<tasks></tasks>']);
// Row 55: the SAME technique as row 54, named explicitly for the
// "PLAN.md is a directory" input class — a directory sitting exactly
// where a plan file is expected. `readDocument`'s `statSync().isFile()`
// guard rejects it before any `readFileSync` on THIS path, so it degrades
// to `unreadable` rather than crashing on EISDIR.
fs.mkdirSync(path.join(phaseDir, '2-01-PLAN.md'), { recursive: true });
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected exit 0 even with a directory in plan position: ${result.error}`);
const payload = JSON.parse(result.output);
const bad = payload.phases[0].plans.find((p) => p.id === '2-01');
assert.ok(bad, 'the directory-shaped plan entry must still be listed as a row');
assert.strictEqual(bad.scope, 'unreadable');
assert.ok(payload.diagnostics.some((d) => d.code === 'plan_unreadable' && d.subject === '2-01-PLAN.md'));
});
test('rejectsPlanSymlinkEscapingThePlanningRoot', (t) => {
// Row 56 — SECURITY FIX (#2790 follow-up). Previously CHARACTERIZED as
// leaking: `readDocument` (planning-inspect.cts) opened a document via
// `fs.statSync`/`fs.readFileSync`, both of which FOLLOW symlinks, so a
// `*-PLAN.md` that was actually a symlink pointing outside `.planning/`
// had its target's content surfaced into the schema-v1 payload — an
// exfiltration path, since these documents are UNTRUSTED input (a clone,
// a PR branch, a teammate's working tree) and the payload is handed to
// downstream tooling verbatim. `readDocument` now resolves both the
// document and the planning root via `fs.realpathSync` and refuses to
// read a target that resolves outside the root; the escaping plan must
// degrade exactly like any other unreadable plan, and a sibling readable
// plan in the SAME phase must be entirely unaffected.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const secretDir = createTempDir('gsd-2790-secret-');
t.after(() => cleanup(secretDir));
const secretFile = path.join(secretDir, 'secret.txt');
const sentinel = 'TOP_SECRET_OBJECTIVE_VALUE_9f3c1a';
fs.writeFileSync(secretFile, ['<objective>', sentinel, '</objective>', ''].join('\n'));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
// A sibling, legitimately-readable plan in the SAME phase — proves the
// escaping symlink degrades ALONE and does not drag its sibling down.
writePlanDoc(phaseDir, '1-01-PLAN.md', [], ['<objective>', 'good one', '</objective>', '<tasks></tasks>']);
try {
fs.symlinkSync(secretFile, path.join(phaseDir, '1-02-PLAN.md'));
} catch (_symlinkErr) {
t.skip('symlink creation unsupported on this platform/privilege');
return;
}
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected exit 0 even with an escaping symlink: ${result.error}`);
// Raw-string ABSENCE proof is the sanctioned exception to the no-raw-text
// rule: the entire point of this assertion is that the sentinel is NEVER
// emitted anywhere in stdout, so only a substring-absence check — not a
// value-level assertion — can express that.
assert.ok(!result.output.includes(sentinel), 'the secret sentinel must never appear anywhere in the payload');
const payload = JSON.parse(result.output);
const good = payload.phases[0].plans.find((p) => p.id === '1-01');
const escaped = payload.phases[0].plans.find((p) => p.id === '1-02');
assert.strictEqual(good.scope, 'complete');
assert.strictEqual(good.objective, 'good one');
assert.strictEqual(escaped.scope, 'unreadable');
assert.strictEqual(escaped.objective, null);
assert.ok(payload.diagnostics.some((d) => d.code === 'plan_unreadable' && d.subject === '1-02-PLAN.md'));
});
test('stillReadsPlansWhenTheWholePlanningRootIsALegitimateSymlinkElsewhere', (t) => {
// Companion to `rejectsPlanSymlinkEscapingThePlanningRoot`: containment
// must not over-reject the case it exists to preserve — "so a legitimately
// symlinked `.planning/` directory ... still works" (per brief). Both
// `filePath` and `root` are resolved with `fs.realpathSync` before the
// boundary check, so a project that relocates its ENTIRE `.planning/`
// directory to elsewhere on disk (a synced folder, a monorepo shared
// location, etc.) and symlinks it back in keeps reading normally — the
// resolved plan target still nests under the resolved root, it just does
// so via the relocated location rather than the literal `cwd/.planning`
// path. Without this test, the fix above could silently regress into
// rejecting every document in such a project.
const bareDir = createTempDir('gsd-2790-bare-');
t.after(() => cleanup(bareDir));
const realPlanningDir = createTempDir('gsd-2790-real-planning-');
t.after(() => cleanup(realPlanningDir));
try {
fs.symlinkSync(realPlanningDir, path.join(bareDir, '.planning'));
} catch (_symlinkErr) {
t.skip('symlink creation unsupported on this platform/privilege');
return;
}
const phaseDir = declarePhase(bareDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', [], ['<objective>', 'relocated plan', '</objective>', '<tasks></tasks>']);
const payload = parseInspect(bareDir);
const plan = payload.phases[0].plans.find((p) => p.id === '1-01');
assert.ok(plan, 'plan row must exist through the symlinked planning root');
assert.strictEqual(plan.scope, 'complete');
assert.strictEqual(plan.objective, 'relocated plan');
});
test('oneDocumentsFaultDoesNotDowngradeAnother', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 57: REQUIREMENTS.md is unreadable (directory-in-file-position);
// ROADMAP.md is clean. Requirements degrades ALONE — milestone identity
// and phase completion must stay unaffected.
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writeVerification(phaseDir, '01', 'passed');
fs.mkdirSync(path.join(tmpDir, '.planning', 'REQUIREMENTS.md'), { recursive: true });
const payload = parseInspect(tmpDir);
assert.strictEqual(payload.milestone.scope, 'complete');
assert.strictEqual(payload.phases[0].complete, true);
assert.deepStrictEqual(payload.requirements, []);
assert.ok(payload.diagnostics.some((d) => d.code === 'requirements_unreadable'));
});
test('foldsWorstScopeAcrossPhasesAndWithholdsPercent', (t) => {
// Row 58: phase A is unreadable (its own directory listing fails), phase
// B is clean and complete. The top-level `accepted_phases.percent` must
// withhold (fold via worstScope) rather than silently computing a
// fraction from only the phases that happened to be readable.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧', '', '### Phase 1: A', '', '### Phase 2: B', '',
]);
const phaseADir = phaseDirOf(tmpDir, slugPhaseDirName('1', 'A'));
const phaseBDir = phaseDirOf(tmpDir, slugPhaseDirName('2', 'B'));
fs.mkdirSync(phaseADir, { recursive: true });
writePlanDoc(phaseBDir, '2-01-PLAN.md', [], ['<objective>', 'b', '</objective>', '<tasks></tasks>']);
writeSummaryDoc(phaseBDir, '2-01-SUMMARY.md', [], ['# s', '', '## Files Created/Modified']);
writeVerification(phaseBDir, '02', 'passed');
const planningInspectLib = require('../gsd-core/bin/lib/planning-inspect.cjs');
const originalReaddirSync = fs.readdirSync;
t.mock.method(fs, 'readdirSync', function mockedReaddirSync(target, ...rest) {
if (target === phaseADir) {
const err = new Error(`EACCES: permission denied, scandir '${phaseADir}'`);
err.code = 'EACCES';
throw err;
}
return originalReaddirSync.call(this, target, ...rest);
});
const payload = planningInspectLib.buildPlanningInspect(tmpDir);
const a = payload.phases.find((p) => p.dir === '01-a');
const b = payload.phases.find((p) => p.dir === '02-b');
assert.strictEqual(a.scope, 'unreadable');
assert.strictEqual(a.complete, false);
assert.strictEqual(b.scope, 'complete');
assert.strictEqual(b.complete, true);
assert.strictEqual(payload.progress.accepted_phases.percent, null);
});
});
// ─── 7c. The 50 KB @file: spill boundary (io.cjs's own `> 50000` predicate) ──
const ioLib = require('../gsd-core/bin/lib/io.cjs');
/**
* Build a JSON payload whose `serializeForOutput` byte length is EXACTLY
* `targetLen` — calibrated once against the real overhead of `{"padding":
* "..."}` at 2-space indent, then padded with ASCII (`'a'`, no JSON escaping)
* so one added character is exactly one added byte. Asserts the calibration
* landed exactly, so a future change to `serializeForOutput`'s formatting
* fails this helper loudly rather than silently testing the wrong boundary.
*/
function paddedPayloadOfSerializedLength(targetLen) {
const overhead = ioLib.serializeForOutput({ padding: '' }).length;
const padLen = Math.max(targetLen - overhead, 0);
const payload = { padding: 'a'.repeat(padLen) };
const actualLen = ioLib.serializeForOutput(payload).length;
assert.strictEqual(actualLen, targetLen, `padding calibration failed: expected length ${targetLen}, got ${actualLen}`);
return payload;
}
describe('planning inspect — the 50 KB @file: spill boundary', () => {
test('emitsInlineJsonJustBelowTheSpillThreshold', () => {
// Row 60 (limit-1): io.cjs's own predicate is `json.length > 50000` —
// 49999 bytes must NOT spill.
const payload = paddedPayloadOfSerializedLength(49999);
assert.strictEqual(ioLib.serializeForOutput(payload).length > 50000, false);
});
test('matchesIoSpillThresholdAtTheBoundary', () => {
// Row 61 (limit): exactly 50000 bytes — the predicate is strictly
// GREATER THAN, so the boundary value itself does NOT spill.
const payload = paddedPayloadOfSerializedLength(50000);
assert.strictEqual(ioLib.serializeForOutput(payload).length > 50000, false);
});
test('spillsOverThresholdAndResolvesBackToJsonOnStdout', (t) => {
// Row 62 (limit+1), part A: the exact boundary+1 byte value against the
// serializer's own predicate.
const payload = paddedPayloadOfSerializedLength(50001);
assert.strictEqual(ioLib.serializeForOutput(payload).length > 50000, true);
// Row 62, part B: a genuinely oversized END-TO-END fixture through the
// real CLI — proving `resolveAtFileOutput` (gsd-tools.cjs) transparently
// resolves the `@file:` spill back into full JSON on stdout, so the
// caller never has to know the spill happened.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
const lines = ['# Requirements', '', '## v1 Requirements', ''];
const padding = 'x'.repeat(400);
for (let i = 1; i <= 300; i += 1) {
lines.push(`- [ ] **REQ-${String(i).padStart(3, '0')}**: padded description ${padding}`);
}
writeRequirements(tmpDir, lines.join('\n'));
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected exit 0 for an oversized payload: ${result.error}`);
assert.ok(result.output.length > 50000, 'the raw fixture text alone should already exceed the spill threshold');
const parsedPayload = JSON.parse(result.output);
assert.deepStrictEqual(sortedKeys(parsedPayload), EXPECTED_TOP_LEVEL_KEYS);
assert.strictEqual(parsedPayload.requirements.length, 300);
});
});
// ─── 8. Task grammar ──────────────────────────────────────────────────────────
describe('planning inspect — task grammar', () => {
test('emitsOneRowPerXmlTaskBlockInDocumentOrder', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>', '',
'<task type="auto"><name>Task A</name><files>src/a.ts</files><action>a</action><done>done</done></task>',
'<task type="auto"><name>Task B</name><files>src/b.ts</files><action>b</action><done>done</done></task>',
'</tasks>',
]);
const payload = parseInspect(tmpDir);
const tasks = payload.phases[0].plans[0].tasks;
assert.strictEqual(tasks.length, 2);
assert.strictEqual(tasks[0].name, 'Task A');
assert.strictEqual(tasks[1].name, 'Task B');
assert.deepStrictEqual(tasks.map((task) => task.index), [1, 2]);
});
test('fallsBackToMarkdownTaskHeadingsWhenNoXmlTaskBlocksExist', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'## Task 1: First',
'',
'## Task 2: Second',
'',
]);
const payload = parseInspect(tmpDir);
const tasks = payload.phases[0].plans[0].tasks;
assert.strictEqual(tasks.length, 2);
assert.strictEqual(tasks[0].kind, 'auto');
assert.strictEqual(tasks[0].name, 'Task 1: First');
assert.strictEqual(tasks[1].name, 'Task 2: Second');
});
test('xmlTaskBlocksWinOverMarkdownHeadingsWhenBothArePresent', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>',
'<task type="auto"><name>Only XML task</name><files>src/a.ts</files><action>a</action><done>done</done></task>',
'</tasks>',
'',
'## Task 1: Legacy heading one',
'## Task 2: Legacy heading two',
'## Task 3: Legacy heading three',
]);
const payload = parseInspect(tmpDir);
const tasks = payload.phases[0].plans[0].tasks;
assert.strictEqual(tasks.length, 1);
assert.strictEqual(tasks[0].name, 'Only XML task');
});
test('emitsCheckpointTaskAsItsOwnKindNotAsMalformed', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>',
'<task type="checkpoint:decision" gate="blocking">',
' <decision>Pick one</decision>',
' <context>Because reasons</context>',
' <options>',
' <option id="a"><name>A</name><pros>x</pros><cons>y</cons></option>',
' </options>',
' <resume-signal>Select: a</resume-signal>',
'</task>',
'</tasks>',
]);
const payload = parseInspect(tmpDir);
const task = payload.phases[0].plans[0].tasks[0];
assert.strictEqual(task.kind, 'checkpoint');
assert.strictEqual(task.name, null);
assert.ok(payload.diagnostics.some((d) => d.code === 'task_shape_checkpoint'));
});
test('characterizesFenceBlindMarkdownTaskFallback', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
// Row 44 (deliberately characterized, not endorsed — see
// plan-document.cts's own header comment): a `## Task N` heading inside a
// FENCED code block still counts under the markdown fallback, exactly as
// `cmdPhasePlanIndex` has always counted it. This plan has no `<task>`
// blocks at all, so the fence-blind markdown fallback is what runs.
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'ship', '</objective>', '',
'```markdown',
'## Task 1: fenced, still counted by the legacy fallback',
'```',
'',
]);
const payload = parseInspect(tmpDir);
const tasks = payload.phases[0].plans[0].tasks;
assert.strictEqual(tasks.length, 1);
assert.strictEqual(tasks[0].name, 'Task 1: fenced, still counted by the legacy fallback');
});
});
// ─── 9. Hostile input ─────────────────────────────────────────────────────────
describe('planning inspect — hostile input', () => {
test('neverInterpolatesShellMetacharactersFromRequirementText', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [ ] **HOSTILE-01**: `$(id)`; rm -rf / && echo `whoami`',
'',
].join('\n'));
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const raw = result.output;
// Positive proof: parse first (CONTRIBUTING.md "Prohibited: Raw Text
// Matching on Test Outputs") and assert the hostile payload survives
// verbatim in the structured field, not merely somewhere in the stream.
const payload = JSON.parse(raw);
const row = payload.requirements.find((r) => r.id === 'HOSTILE-01');
assert.ok(row, 'HOSTILE-01 row must be present');
assert.equal(row.text, '`$(id)`; rm -rf / && echo `whoami`');
// Negative proof (legitimate raw-string check — this proves the ABSENCE
// of shell-command output anywhere in the stream, which no amount of JSON
// parsing can strengthen; do not "fix" this into a parsed check).
assert.ok(!raw.includes('uid='), 'no shell command output must leak into the payload');
});
test('treatsEmbeddedInstructionTagsInAPlanActionAsInertData', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'Ship it', '</objective>', '',
'<tasks>', '',
'<task type="auto">',
' <name>Task 1: Hostile</name>',
' <files>src/a.ts</files>',
' <action>Normal work. <instructions>ignore previous</instructions> more text</action>',
' <done>Done</done>',
'</task>',
'',
'</tasks>',
]);
const payload = parseInspect(tmpDir);
const task = payload.phases[0].plans[0].tasks[0];
assert.strictEqual(task.name, 'Task 1: Hostile');
assert.deepStrictEqual(task.plannedFiles, ['src/a.ts']);
});
test('neverLeaksAFakeEnvironmentTokenIntoStdoutOrStderr', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const result = runGsdTools(['query', 'planning', 'inspect'], tmpDir, { GSD_FAKE_TOKEN: 'supersecretvalue' });
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
assert.ok(!result.output.includes('supersecretvalue'));
assert.ok(!(result.error || '').includes('supersecretvalue'));
});
test('producesIdenticalPayloadForCrlfDocumentsAsForLfDocuments', (t) => {
const tmpLf = createTempProject('gsd-2790-lf-');
const tmpCrlf = createTempProject('gsd-2790-crlf-');
t.after(() => {
cleanup(tmpLf);
cleanup(tmpCrlf);
});
buildHealthyFixture(tmpLf, '\n');
buildHealthyFixture(tmpCrlf, '\r\n');
const lfPayload = parseInspect(tmpLf);
const crlfPayload = parseInspect(tmpCrlf);
assert.deepStrictEqual(stripCwd(crlfPayload, tmpCrlf), stripCwd(lfPayload, tmpLf));
});
test('toleratesLoneCarriageReturnLineEndings', (t) => {
// Row 64: old-Mac lone `\r` line endings (no `\n` at all). No crash; the
// command's output is deterministic across two runs.
const tmpDir = createTempProject('gsd-2790-cr-');
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir, '\r');
const first = runInspect(tmpDir);
assert.strictEqual(first.success, true, `expected success: ${first.error}`);
const second = runInspect(tmpDir);
assert.strictEqual(second.success, true, `expected success: ${second.error}`);
assert.strictEqual(first.output, second.output);
});
test('handlesNullByteAndReplacementCharInDocumentText', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
// Row 65: a NUL byte and a U+FFFD replacement character embedded in a
// requirement description. Both are valid UTF-8 payload bytes/codepoints
// — the value must be carried verbatim, never silently truncated at the
// NUL.
const nul = String.fromCharCode(0);
const replacementChar = '<27>';
const description = `has a${nul}nul and a ${replacementChar} replacement char inline`;
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
`- [ ] **REQ-10**: ${description}`,
'',
].join('\n'));
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const payload = JSON.parse(result.output);
const row = payload.requirements.find((r) => r.id === 'REQ-10');
assert.ok(row, 'REQ-10 row must be present');
assert.ok(row.text.includes(nul), 'the NUL byte must be carried verbatim, not silently dropped');
assert.ok(row.text.includes(replacementChar), 'U+FFFD must be carried verbatim');
});
test('neverResolvesTraversalShapedPhaseTokenToAPath', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
// Row 66: a `../../x`-shaped token in a traceability row's Phase cell.
// The phase-token extractor only pulls DIGIT tokens out of that cell
// (`parseTraceability`'s `\d+(?:\.\d+)*` scan) — a traversal shape simply
// yields no numeric token, so it is never resolved to a filesystem path.
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
'- [ ] **REQ-09**: something',
'',
'## Traceability',
'',
'| Requirement | Phase | Status |',
'|-------------|-------|--------|',
'| REQ-09 | ../../../etc | Pending |',
'',
].join('\n'));
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const raw = result.output;
const payload = JSON.parse(raw);
const row = payload.requirements.find((r) => r.id === 'REQ-09');
assert.ok(row, 'REQ-09 row must be present');
assert.deepStrictEqual(row.mappedPhases, []);
// Negative proof: no path-resolution artifact (a resolved root path, or
// the traversal string itself surviving into a filesystem-shaped field)
// leaks anywhere in the stream.
assert.ok(!raw.includes('../../../etc'));
});
test('boundsAPathologicallyLongDocumentValue', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
declarePhase(tmpDir, '1', 'Foo');
// Row 70: a single requirement description ~1 MB long. The command must
// still complete and exit 0 — no unbounded read, no hang.
const longValue = 'x'.repeat(1024 * 1024);
writeRequirements(tmpDir, [
'# Requirements',
'',
'## v1 Requirements',
'',
`- [ ] **REQ-11**: ${longValue}`,
'',
].join('\n'));
// Assert via `--pick schema_version` rather than the full JSON payload.
// gsd-tools handles this correctly end-to-end: output() spills the >50KB
// JSON to a tmpfile and resolves the @file: reference back to stdout, so
// the full 1 MB document IS read and parsed — but resolved stdout is then
// well over runGsdTools's subprocess maxBuffer, which makes the HELPER
// report BUFFER_OVERFLOW/ENOBUFS. That failure is the test harness's
// buffer ceiling, not the product. `--pick` makes gsd-tools extract a
// single small field and print only that, so stdout stays tiny while the
// full 1 MB document is still read and parsed end to end — proving the
// command completes successfully on a pathologically long value without
// measuring runGsdTools's maxBuffer instead of the product.
const result = runInspect(tmpDir, ['--pick', 'schema_version']);
assert.strictEqual(result.success, true, `expected success for a 1MB value: ${result.error}`);
assert.strictEqual(result.output, '1', 'schema_version must round-trip as 1 through --pick');
});
test('preservesUnicodeAndRtlDocumentText', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
// Row 71: Unicode + RTL headings and names must round-trip byte-identical
// into the payload — no width/parse corruption.
const rtlName = 'المصادقة'; // Arabic: "Authentication"
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'## Phases',
'',
`- [ ] **Phase 1: ${rtlName}** - stub`,
'',
`### Phase 1: ${rtlName}`,
'',
rtlName,
'',
]);
fs.mkdirSync(phaseDirOf(tmpDir, slugPhaseDirName('1', rtlName)), { recursive: true });
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const payload = JSON.parse(result.output);
const phase = payload.phases[0];
assert.strictEqual(phase.goal.value, rtlName);
});
});
// ─── 10. Cross-consumer parity ────────────────────────────────────────────────
describe('planning inspect — cross-consumer parity with phase-plan-index', () => {
test('phasePlanIndexAndPlanningInspectAgreeOnPlanObjectiveAndTaskCount', (t) => {
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Parity');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>',
'Ship the parity check',
'</objective>',
'',
'<tasks>',
'<task type="auto"><name>Task 1: A</name><files>src/a.ts</files><action>a</action><done>done</done></task>',
'<task type="auto"><name>Task 2: B</name><files>src/b.ts</files><action>b</action><done>done</done></task>',
'</tasks>',
]);
const inspectPayload = parseInspect(tmpDir);
const planIndexResult = runGsdTools(['phase-plan-index', '1', '--raw'], tmpDir);
assert.strictEqual(planIndexResult.success, true, `phase-plan-index should succeed: ${planIndexResult.error}`);
const planIndexPayload = JSON.parse(planIndexResult.output);
const inspectPlan = inspectPayload.phases[0].plans[0];
const indexPlan = planIndexPayload.plans.find((p) => p.id === inspectPlan.id);
assert.ok(indexPlan, 'phase-plan-index must report the same plan id');
assert.strictEqual(inspectPlan.objective, indexPlan.objective);
assert.strictEqual(inspectPlan.tasks.length, indexPlan.task_count);
});
test('phasePlanIndexBehaviorUnchangedByPlanDocumentExtraction', (t) => {
// Row 73: `phase-plan-index`'s wave/depends_on/incomplete/runnable shape
// — the live regression surface named by the matrix's own risk section —
// exercised directly, independent of `planning.inspect`, to prove the
// `plan-document.cts` extraction did not change this command's output.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const phaseDir = declarePhase(tmpDir, '1', 'Regress');
writePlanDoc(phaseDir, '1-01-PLAN.md', ['wave: 1'], [
'<objective>', 'First', '</objective>', '',
'<tasks>',
'<task type="auto"><name>Task 1</name><files>src/a.ts</files><action>a</action><done>done</done></task>',
'</tasks>',
]);
writeSummaryDoc(phaseDir, '1-01-SUMMARY.md', ['status: complete'], [
'# Summary', '', '## Files Created/Modified', '- `src/a.ts` - a',
]);
writePlanDoc(phaseDir, '1-02-PLAN.md', ['wave: 2', 'depends_on: 1-01'], [
'<objective>', 'Second', '</objective>', '',
'<tasks>',
'<task type="auto"><name>Task 1</name><files>src/b.ts</files><action>b</action><done>done</done></task>',
'</tasks>',
]);
// No SUMMARY for 1-02 — it stays incomplete/runnable.
const result = runGsdTools(['phase-plan-index', '1', '--raw'], tmpDir);
assert.strictEqual(result.success, true, `expected success: ${result.error}`);
const payload = JSON.parse(result.output);
const first = payload.plans.find((p) => p.id === '1-01');
const second = payload.plans.find((p) => p.id === '1-02');
assert.ok(first, '1-01 must be present');
assert.ok(second, '1-02 must be present');
assert.strictEqual(first.wave, 1);
assert.strictEqual(second.wave, 2);
assert.deepStrictEqual(second.depends_on, ['1-01']);
assert.deepStrictEqual(payload.incomplete, ['1-02']);
assert.deepStrictEqual(payload.runnable, ['1-02']);
assert.ok(Array.isArray(payload.waves['1']));
assert.ok(payload.waves['1'].includes('1-01'));
assert.ok(Array.isArray(payload.waves['2']));
assert.ok(payload.waves['2'].includes('1-02'));
});
test('producesDeterministicOutputAcrossRuns', (t) => {
// Row 75: two runs over the same fixture must produce byte-identical
// stdout — array ordering (task rows, plan rows, diagnostics) is
// deterministic, not incidentally stable.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
buildHealthyFixture(tmpDir);
const first = runInspect(tmpDir);
const second = runInspect(tmpDir);
assert.strictEqual(first.success, true, `expected success: ${first.error}`);
assert.strictEqual(second.success, true, `expected success: ${second.error}`);
assert.strictEqual(first.output, second.output);
});
});
// ─── 11. Property tests ───────────────────────────────────────────────────────
const EOL_ARB = fc.constantFrom('\n', '\r\n');
/**
* Document-shaped: presence/absence of each document, 0..3 phases, 0..3
* requirements, CRLF vs LF, empty vs populated — NOT seeded from
* `planning-inspect.cts`'s own parsing model (see file-header provenance note).
*/
const PLANNING_PROJECT_SHAPE_ARB = fc.record({
hasState: fc.boolean(),
hasRoadmap: fc.boolean(),
hasRequirements: fc.boolean(),
requirementsEmpty: fc.boolean(),
phaseCount: fc.integer({ min: 0, max: 3 }),
requirementCount: fc.integer({ min: 0, max: 3 }),
eol: EOL_ARB,
});
function buildDocumentShapedProject(cwd, cfg) {
if (cfg.hasState) {
writeState(cwd, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0'], [], cfg.eol);
}
if (cfg.hasRoadmap) {
const lines = ['## v1.0 Current 🚧', ''];
for (let i = 1; i <= cfg.phaseCount; i += 1) {
lines.push(`### Phase ${i}: Phase${i}`, '');
}
writeRoadmap(cwd, lines, cfg.eol);
}
for (let i = 1; i <= cfg.phaseCount; i += 1) {
const token = String(i);
const phaseDir = phaseDirOf(cwd, slugPhaseDirName(token, `Phase${i}`));
writePlanDoc(phaseDir, `${token}-01-PLAN.md`, ['wave: 1'], [
'<objective>', `Ship phase ${i}`, '</objective>', '',
'<tasks>',
'<task type="auto"><name>Task 1</name><files>src/x.ts</files><action>do it</action><done>done</done></task>',
'</tasks>',
], cfg.eol);
writeSummaryDoc(phaseDir, `${token}-01-SUMMARY.md`, ['status: complete'], [
'# Summary', '', '## Files Created/Modified', '- `src/x.ts` - x',
], cfg.eol);
writeVerification(phaseDir, token, 'passed', cfg.eol);
}
if (cfg.hasRequirements) {
if (cfg.requirementsEmpty) {
writeRequirements(cwd, '');
} else {
const lines = ['# Requirements', '', '## v1 Requirements', ''];
for (let i = 1; i <= cfg.requirementCount; i += 1) {
lines.push(`- [ ] **REQ-0${i}**: Requirement ${i}`);
}
writeRequirements(cwd, lines.join(cfg.eol));
}
}
}
describe('planning inspect — property tests', () => {
test('propertySchemaIsTotalOverDocumentShapedInputs', (t) => {
const createdDirs = [];
t.after(() => {
for (const d of createdDirs) cleanup(d);
});
fc.assert(
fc.property(PLANNING_PROJECT_SHAPE_ARB, (cfg) => {
const tmpDir = createTempProject('gsd-2790-prop-');
createdDirs.push(tmpDir);
buildDocumentShapedProject(tmpDir, cfg);
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `command must exit 0 for cfg=${JSON.stringify(cfg)}: ${result.error}`);
const payload = JSON.parse(result.output);
assert.deepStrictEqual(sortedKeys(payload), EXPECTED_TOP_LEVEL_KEYS);
const percent = payload.progress.accepted_phases.percent;
assert.ok(
percent === null || (Number.isInteger(percent) && percent >= 0 && percent <= 100),
`percent must be null or an integer in [0,100], got ${percent} for cfg=${JSON.stringify(cfg)}`,
);
}),
{ numRuns: 20, seed: 279040, verbose: true },
);
});
});
// ─── 12. plan-document task-count parity (property, direct require) ──────────
const { parsePlanDocument } = require('../gsd-core/bin/lib/plan-document.cjs');
const TASK_COUNT_ARB = fc.record({
xmlCount: fc.integer({ min: 0, max: 5 }),
mdCount: fc.integer({ min: 0, max: 5 }),
});
describe('plan-document — task count parity (property)', () => {
test('parsesWithTheArgumentShapeProductionActuallyPasses', () => {
// Row 74: `planning-inspect.cjs`'s OWN call site (`buildPlanRows`) is
// `parsePlanDocument(doc.text)` — `planPath` OMITTED. Proven directly
// against that exact shape, not a hand-passed `(content, path)` pair.
const content = [
'---',
'wave: 2',
'depends_on: 1-01',
'files_modified: src/a.ts, src/b.ts',
'---',
'',
'<objective>',
'Ship the omitted-argument case',
'</objective>',
'',
'<tasks>',
'<task type="auto"><name>Task 1</name><files>src/a.ts</files><action>do it</action><done>done</done></task>',
'</tasks>',
].join('\n');
const parsed = parsePlanDocument(content);
assert.strictEqual(parsed.objective, 'Ship the omitted-argument case');
assert.strictEqual(parsed.declaredWave, 2);
assert.deepStrictEqual(parsed.dependsOn, ['1-01']);
// A single scalar frontmatter value is NOT comma-split — only an actual
// YAML array is mapped element-wise (see parsePlanDocument's `fmFiles`
// handling). One frontmatter line yields one array element verbatim.
assert.deepStrictEqual(parsed.filesModified, ['src/a.ts, src/b.ts']);
assert.strictEqual(parsed.tasks.length, 1);
assert.strictEqual(parsed.taskCount, 1);
});
test('propertyTaskCountMatchesLegacyFallbackRule', () => {
fc.assert(
fc.property(TASK_COUNT_ARB, ({ xmlCount, mdCount }) => {
const xmlBlocks = [];
for (let i = 0; i < xmlCount; i += 1) {
xmlBlocks.push(`<task type="auto"><name>Task ${i + 1}</name></task>`);
}
const mdBlocks = [];
for (let i = 0; i < mdCount; i += 1) {
mdBlocks.push(`## Task ${i + 1}`);
}
const content = [
'<objective>',
'Objective',
'</objective>',
'<tasks>',
...xmlBlocks,
'</tasks>',
...mdBlocks,
].join('\n');
// Default-argument caller shape — `parsePlanDocument(content)` with
// `planPath` omitted, matching how `planning-inspect.cts` calls it.
const parsed = parsePlanDocument(content);
const expected = xmlCount > 0 ? xmlCount : mdCount;
assert.strictEqual(parsed.tasks.length, expected);
}),
{ numRuns: 40, seed: 279041, verbose: true },
);
});
});
// ─── 9. Containment: escaped phase directories / verification files (#2790 follow-up) ─
describe('planning inspect — phase-directory and verification-file containment', () => {
test('escapedPhaseDirectorySymlinkLeaksNoFilenamesOrContent', (t) => {
// GAP 1: `readdirSync` follows a directory symlink, so a phase directory
// that is itself a symlink escaping the planning root must contribute NO
// filenames and NO content. Empirically, a symlinked entry directly under
// `.planning/phases/` never even reaches `buildPlanRows`/`buildUatRows`
// in the first place: `listMilestonePhaseDirs`/`buildAllPhaseDirNamesField`
// (`src/phase-locator.cts` / `src/planning-snapshot.cts` — both OUT OF
// SCOPE for this fix, and unrelated to it) enumerate phase directories via
// `readdirSync(..., { withFileTypes: true }).filter((e) => e.isDirectory())`,
// and `Dirent#isDirectory()` reports a directory SYMLINK as `false` (it is
// typed from the directory entry itself, never `stat`-resolved) — so the
// escaped entry is excluded from BOTH the windowed `phases` array and
// `orphan_phase_dirs` before `planning-inspect.cts` ever sees it. The
// `isPathContained` guard added to `buildPlanRows`/`buildUatRows` is
// still correct defense-in-depth (a direct call, a future refactor of the
// upstream filter, or a platform where a directory reparse point reports
// as a directory could all reach it) — this test proves the OUTCOME the
// security review actually cares about: end to end, nothing about the
// escaped directory or its contents is ever observable, and a healthy
// sibling phase is entirely unaffected.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const externalDir = createTempDir('gsd-2790-external-phase-');
t.after(() => cleanup(externalDir));
const sentinelFileName = 'TOP-SECRET-FILENAME-8f21ac.md';
const sentinelContent = 'TOP_SECRET_PHASE_CONTENT_4d81af';
fs.writeFileSync(path.join(externalDir, sentinelFileName), sentinelContent);
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'### Phase 1: Foo',
'',
'### Phase 2: Bar',
'',
]);
const phase1Dir = phaseDirOf(tmpDir, slugPhaseDirName('1', 'Foo'));
fs.mkdirSync(path.dirname(phase1Dir), { recursive: true });
try {
fs.symlinkSync(externalDir, phase1Dir);
} catch (_symlinkErr) {
t.skip('symlink creation unsupported on this platform/privilege');
return;
}
const phase2Dir = phaseDirOf(tmpDir, slugPhaseDirName('2', 'Bar'));
fs.mkdirSync(phase2Dir, { recursive: true });
writeVerification(phase2Dir, '02', 'passed');
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected exit 0 with an escaping phase directory: ${result.error}`);
// Raw-string ABSENCE proof — the sanctioned exception to the no-raw-text
// rule (see `rejectsPlanSymlinkEscapingThePlanningRoot` above).
assert.ok(!result.output.includes(sentinelFileName), 'the external filename must never appear anywhere in the payload');
assert.ok(!result.output.includes(sentinelContent), 'the external content must never appear anywhere in the payload');
const payload = JSON.parse(result.output);
const escaped = payload.phases.find((p) => p.dir === slugPhaseDirName('1', 'Foo'));
const healthy = payload.phases.find((p) => p.dir === slugPhaseDirName('2', 'Bar'));
// The escaped directory is invisible end to end — neither a phase row nor
// an orphan entry names it (see the upstream Dirent-filtering note above).
assert.strictEqual(escaped, undefined, 'the escaped phase directory must not surface as a phase row');
assert.ok(!payload.orphan_phase_dirs.includes(slugPhaseDirName('1', 'Foo')));
assert.ok(healthy, 'the sibling phase must be entirely unaffected by the escape');
assert.strictEqual(healthy.scope, 'complete');
assert.strictEqual(healthy.complete, true);
});
test('escapedVerificationFileSymlinkLeaksNoFrontmatterValue', (t) => {
// GAP 2: `readVerificationStatus` (src/verification.cts) is a shared
// owner with its own unguarded `readFileSync` — an unrecognized `status:`
// value is copied verbatim into `next_action`. A `*-VERIFICATION.md`
// symlinked outside the planning root must never surface that value.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
const externalDir = createTempDir('gsd-2790-external-verification-');
t.after(() => cleanup(externalDir));
const sentinel = 'TOP_SECRET_STATUS_VALUE_c93af1';
const secretFile = path.join(externalDir, 'secret-status.md');
fs.writeFileSync(secretFile, ['---', `status: ${sentinel}`, '---', ''].join('\n'));
const phaseDir = declarePhase(tmpDir, '1', 'Foo');
try {
fs.symlinkSync(secretFile, path.join(phaseDir, '01-VERIFICATION.md'));
} catch (_symlinkErr) {
t.skip('symlink creation unsupported on this platform/privilege');
return;
}
const result = runInspect(tmpDir);
assert.strictEqual(result.success, true, `expected exit 0 with an escaping verification symlink: ${result.error}`);
assert.ok(!result.output.includes(sentinel), 'the sentinel status value must never appear anywhere in the payload');
const payload = JSON.parse(result.output);
const phase = payload.phases[0];
// Degrades exactly like an unreadable/absent verification report already
// does — never like a recognized-but-unknown status carrying the raw value.
assert.strictEqual(phase.verification.status, 'missing');
assert.ok(!phase.verification.next_action.includes(sentinel), 'next_action specifically must never carry the sentinel');
});
test('negativeControlRelocatedPhasesDirAndPlainVerificationFileBothStillWork', (t) => {
// Companion to both tests above: containment must not over-reject the
// cases it exists to preserve. Per the Dirent-filtering note in
// `escapedPhaseDirectorySymlinkLeaksNoFilenamesOrContent` above, symlinking
// an INDIVIDUAL phase directory is never recognized as a phase by the
// upstream enumerator regardless of where it points — that is a pre-existing,
// out-of-scope limitation of `listMilestonePhaseDirs`, not a containment
// question. The reachable, meaningful "contained relocation" case is
// instead the whole `.planning/phases/` PARENT being a symlink to another
// directory INSIDE the planning root, with ORDINARY (non-symlink) phase
// subdirectories nested inside it — `readdirSync` resolves the symlinked
// parent path once and then lists genuinely-typed directory entries
// within it, so those phase rows DO reach `buildPlanRows`/`buildUatRows`/
// the verification `fs` seam with a `phaseDir` whose resolved realpath
// sits under the relocated-but-contained real target. This is the same
// "legitimately relocated planning tree" shape
// `stillReadsPlansWhenTheWholePlanningRootIsALegitimateSymlinkElsewhere`
// proves for the WHOLE `.planning/` root, one level down at `phases/`.
const tmpDir = createTempProject();
t.after(() => cleanup(tmpDir));
writeState(tmpDir, ["gsd_state_version: '1.0'", 'status: planning', 'milestone: v1.0']);
writeRoadmap(tmpDir, [
'## v1.0 Current 🚧',
'',
'### Phase 1: Foo',
'',
'### Phase 2: Bar',
'',
]);
const realPhasesDir = path.join(planningDirOf(tmpDir), '_actual-phases');
const phase1RealDir = path.join(realPhasesDir, slugPhaseDirName('1', 'Foo'));
fs.mkdirSync(phase1RealDir, { recursive: true });
writePlanDoc(phase1RealDir, '1-01-PLAN.md', [], [
'<objective>', 'relocated-parent plan', '</objective>', '<tasks></tasks>',
]);
// (b) Phase 2 gets a plain, non-symlink verification report, proving the
// common case is unaffected by this fix.
const phase2RealDir = path.join(realPhasesDir, slugPhaseDirName('2', 'Bar'));
fs.mkdirSync(phase2RealDir, { recursive: true });
writeVerification(phase2RealDir, '02', 'passed');
// `createTempProject` already creates an empty `.planning/phases/` — it
// must be removed before a symlink can take its place.
fs.rmdirSync(phasesDirOf(tmpDir));
try {
fs.symlinkSync(realPhasesDir, phasesDirOf(tmpDir));
} catch (_symlinkErr) {
t.skip('symlink creation unsupported on this platform/privilege');
return;
}
const payload = parseInspect(tmpDir);
const phase1 = payload.phases.find((p) => p.dir === slugPhaseDirName('1', 'Foo'));
assert.ok(phase1, 'a phase reached through a relocated-but-contained phases/ parent must still be listed');
assert.strictEqual(phase1.scope, 'complete');
const plan = phase1.plans.find((p) => p.id === '1-01');
assert.ok(plan, 'a plan inside the relocated-but-contained directory must still be read');
assert.strictEqual(plan.objective, 'relocated-parent plan');
const phase2 = payload.phases.find((p) => p.dir === slugPhaseDirName('2', 'Bar'));
assert.ok(phase2, 'the sibling phase reached through the same relocated parent must still be listed');
assert.strictEqual(phase2.verification.status, 'passed', 'a normal verification file must still surface its status');
});
});