Files
msd-core/tests/workflow-fragments-emission.install.test.cjs
Tom Boucher ff4a57b78c chore(#1671): migrate the remaining 13 LARGE/XL workflows to the fragment model — Phase 6.3 (#3030)
* chore(#2994): fragmentize progress.md forensic audit onto the fragment model

Extract the --forensic-gated forensic_audit step to
workflows/progress/steps/forensic-audit.md behind a section marker, and
repair progress.md's init line to forward --forensic so the atom is
actually true in production rather than only under direct CLI tests.

progress.md shrinks 32630 -> 27207 bytes.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): fragmentize the four manifest-wired workflows

new-project, quick, new-milestone and progress each already had a
dedicated cmdInit* entry point but zero marked sections. Extract nine
gated bodies to workflows/<wf>/steps/ behind section markers and repair
each init line to forward its flags.

Fold --full into the discuss/research/validate facts inside cmdInitQuick
so the when= grammar never sees an OR, per the chunked-mode precedent.

Fixes found while working, per the no-defer rule:
- cmdInitProgress passed no phase info to buildSectionManifestField, so
  state:phase-mvp-mode was permanently false — an atom in the vocabulary
  whose fact could never be computed.
- the quick init router folded flag tokens into the free-text
  description, which the new forwarding would have corrupted.
- a #2508 dispatch note was nested inside quick.md's Agent(prompt=)
  fence, leaking orchestrator guidance into the subagent prompt.
- progress.md had a 3-vs-4 backtick outer-fence imbalance.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): fragmentize verify-work.md and admit state:ui-phase-active

Wire cmdInitVerifyWork to buildSectionManifestField — it was a dedicated
entry point that never emitted a manifest — and mark two sections.

state:ui-phase-active folds (plan:pre hooks include an active ui step) OR
(the phase dir holds a *-UI-SPEC.md) into one boolean in init.cts, so the
grammar still sees a single operator-free atom. The inner Playwright-MCP
check stays as prose inside the fragment: it is live session state and no
init seam can precompute it.

The MVP false-branch note is a real fallback, not redundant prose, so it
sits outside the marker — gating it away would delete the text needed
precisely when MVP mode is off.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* test(#2994): follow moved workflow content in drift guards

Retarget every guard that asserted on content this branch moved into
workflows/<wf>/steps/, mirroring 815b3d897. Each retargeted assertion was
verified to still fail when its step file is blanked, so none was
weakened into vacuity.

Three assertions in verify-mvp-uat were genuinely red. Three more were
worse than red — passing for the wrong reason:
- quick-commit-boundary and worktree-cleanup anchored on indexOf('Step
  5.6'), which matched a later cross-reference and sliced 16069 chars
  that coincidentally held the asserted substrings. Replaced with an
  expandWorkflowSections helper that splices step content back in place.
- phase6-review-capabilities lost its end boundary and widened to EOF.
- playwright-ui-verify matched 'UI' in an unrelated bullet and 'fall
  back' in a subagent-dispatch line after the real content moved.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): fragmentize code-review and complete-milestone, admit three atoms

Add dedicated cmdInitCodeReview and cmdInitCompleteMilestone entry points
alongside the shared generic ones rather than modifying them — init.phase-op
and init.manager carry a CRITICAL blast radius (179 dependents, 24
processes) and stay byte-identical for their other callers.

Admit flag:--fix, state:fallow-enabled and state:git-create-tag, each with
a consuming section and a fact its own entry point computes.

Both sections had the resolver-in-body hazard: the fallow config-gate and
the git.create_tag check each sat inside the very block being gated, so
gating would have disabled the resolver that decides the gate. Both are
hoisted into init and the bodies now consume the resolved fact.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* test(#2994): retarget code-review and milestone drift guards, fix two red tests

Retarget guards that asserted on content moved into steps/, proving
non-vacuity by blanking each step file and confirming failure.

Also fixes two genuinely red tests found while working, per the no-defer
rule:
- workflow-fragments' frozen-vocabulary lock was missing
  state:ui-phase-active, so commit 7ef7f8336 shipped red. Lint and build
  both passed over it, which is why neither is sufficient verification.
- code-review's quick.md capability-hook assertion carried a stale
  delimiter after the 18ff35d20 extraction.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): fragmentize autonomous.md and admit state:plan-strategy-converge

Five sections share one atom, the pattern plan-phase already uses for
flag:--research-phase. The atom folds --converge OR --cross-ai into a
single boolean in cmdInitAutonomous so the grammar stays operator-free.

cmdInitAutonomous is additive; init.milestone-op, init.manager and
init.phase-op are untouched and still consumed. The $PLAN_STRATEGY bash
resolver is deliberately retained — ungated local-planning bullets still
read it, so the init-side fact supplements it rather than replacing it.

converge-fail-fast required splitting one bash fence so the always-run
CONVERGENCE_ARGS construction stays outside the marker. All three
flag-absent fallbacks were left outside their markers.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): fragmentize review and discuss-phase-assumptions

Admit state:reviewer-instances-configured (two peripheral notes share it;
the core reviewer-lane dispatch stays unmarked — it is the workflow's
primary always-evaluated logic, not an optional branch) and
state:auto-advance-active, which folds --auto OR two config keys into one
boolean so the grammar stays operator-free.

discuss-phase-assumptions was the highest-risk edit in this PR. Its
auto_advance step is a full if/elif/else; gating it whole would have
deleted the flag-absent fallback needed exactly when --auto is off. Split
verified exact: resolvers 636-651 and the 'End here' fallback 668-669 both
stay outside the marker; only 653-667 is gated.

Adds emitted-drift acks for the two files that grew — review.md (+55 B)
and autonomous.md (+737 B from 80799211c, which had none and would have
red-gated the push.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): fragmentize docs-update, update, transition and new-milestone Part A

Completes the 13-workflow rollout. Three of these had no init call at all
and gained a dedicated entry point plus their first gsd_run query line.

Admits state:is-monorepo and adds state:next-channel, state:workstream-active
and state:flat-mode. Vocabulary 26 -> 30 atoms.

Part A of new-milestone applies when NO workstream is active — the negation
of state:workstream-active. Rather than teach the grammar negation, which is
the Greenspun drift the frozen list exists to prevent, it gets a separate
positively-phrased atom whose fact is the inverse. Part B, which always runs,
stays outside the marker.

flag:--verify-only is deliberately NOT admitted: docs-update has no
contiguous purely-additive region for it, and an atom without a consuming
section is dead vocabulary. Evidence recorded in the slice report.

update.md reuses its existing resolved $GSD_TOOLS rather than prepending the
canonical preamble, which would have clobbered it.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(#2994): stop automated-ui-verification re-resolving its own gate, retire dead vocabulary

Two defects the new tests caught.

The automated-ui-verification step re-ran gsd_run loop render-hooks and
recomputed UI_PHASE_ACTIVE inside a body that is only read when that fact
is already true — the circular self-disabling pattern this design forbids,
introduced by 3c654b168. cmdInitVerifyWork now exposes ui_phase_active and
the step consumes it. Its launcher preamble goes too: no gsd_run remains.
The Playwright-MCP check stays as prose — that is live session state.

Dead vocabulary predating this PR: flag:--full and state:needs-codebase-map
were admitted with a gate-1 claim that never materialized. flag:--full is
removed, redundant once quick folds it into discuss/research/validate.
state:needs-codebase-map gets the real consumer it always lacked, gating
new-project's codebase-map offer. Vocabulary 30 -> 29, and no atom is now
without a consuming section.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* test(#2994): add the atom-admission, inversion and resolver-hoist gates

The two existing parity guards prove vocabulary/predicate symmetry but
never that a fact is computed — an atom no cmdInit* assembles evaluates
false forever. These close that hole:

- per-atom satisfiability for all 29 atoms, plus an anti-vacuity assertion
  so the loop cannot silently cover zero atoms
- dead-vocabulary check against the shipped manifest
- inversion guard: the flag-absent fallbacks in discuss-phase-assumptions
  and verify-work must stay outside their markers
- data-driven resolver-hoist guard over the shipped manifest, so a future
  extraction cannot reintroduce the circular class
- compound-fold coverage (--full, --cross-ai, --rc, config-only --auto)
- null-vs-[] degraded/computed distinction, and flag value shapes

Also repairs the frozen-vocabulary lock, which was stale and red for the
seven atoms earlier commits on this branch shipped.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* docs(#2994): add changeset for the fragment-model rollout

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* test(#2994): cite the issue on the two new allow-test-rule exemptions

ADR-456 requires an issue ref on the same line as the annotation.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* docs(#2994): correct the atom-count claims after retiring flag:--full

The vocabulary doc comments still said 30 entries; it is 29 since
flag:--full was removed as dead vocabulary.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(#2994): dedupe the phase-fallback block and harden --ws parsing

Review findings.

MAJOR: the three new init entry points each pasted a verbatim copy of the
guardedFindPhase/guardedGetRoadmapPhase fallback, taking the repo from four
copies to seven — DEFECT.GENERATIVE-FIX. Extracted applyRoadmapFallback and
folded six of the seven; each call site keeps its own field-set via a
closure. Duplication removed rather than papered over with a parity test.
cmdInitPhaseOp stays out: its fallback omits has_reviews, so it is not a
byte-identical copy, and it is CRITICAL-radius.

LOW, pre-existing: GSD_WS captured [^[:space:]]+ and expands unquoted, so a
workstream name holding glob metacharacters would expand against the
filesystem. Narrowed to [A-Za-z0-9._-]+. The unquoted expansion is kept —
it must word-split into two args and vanish when empty.

Also restores the vocabulary ordering convention, and fixes a masked test
bug the mandated run surfaced: the flag-forwarding guard checked only the
first init line per workflow, but new-milestone has two, so a real failure
was reporting exit 0.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(#2994): drop the stale new-milestone emitted-drift ack

new-milestone.md was acked for a +406 B growth measured against an
intermediate commit. Net against origin/next it SHRANK by 8 bytes, so
nothing needed the ack and it explained nothing — which the differential
attribution check reports as a stale acknowledgment, not a pass.

update.md's entry stays: it genuinely grew +703 B.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(#2994): resolve the 15 failures from the full matrix run

All 15 were real and identical on both lanes.

REAL REGRESSION: autonomous.md hit 41479 chars against the #2196 guard's
40960 cap — a CHARS cap distinct from the LARGE tier byte cap, which the
five section stubs pushed it over. Extracted the 3a.5 UI Design Contract
body to references/; now 39968 chars, and the file nets -795 B vs base, so
its growth ack is deleted rather than left stale.

REAL DEFECT: docs referenced /gsd-transition, which is not a live
registered command. Reworded.

STALE FIXTURE: the emission byte-identity test hardcoded two marked
workflows; this branch legitimately marks fifteen. Fixture corrected — the
source was right.

The rest were drift guards over the eight workflows the earlier sweep did
not cover, retargeted at where the content now lives with non-vacuity
proven by blanking each step file and confirming failure. The GSD_WS
forwarding guard was checked as a possible real break and is not one: the
charclass narrowing is intact and forwarding works end to end.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(#2994): drop the ack for a newly-added reference file

A new file's emitted ripple is attributable to the diff that adds it, so
the acknowledgment explained nothing and the differential check reports it
as stale. Removing the last entry removes the fragment — an empty one
signals nothing.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(#2994): retarget the UI-contract guards and clear two transitive advisories

The §3a.5 extraction that brought autonomous.md under the #2196 char cap
moved its body to references/autonomous-ui-design-contract.md, so ten
guards in autonomous-ui-steps and check-ui-safety-gate were asserting it
against the host. Retargeted via a combined read, each proven non-vacuous
by blanking the reference file and confirming failure.

This class had already bitten twice on this branch because each sweep was
scoped to the workflows touched at that moment, so this one was
exhaustive: ~70 test files across all 13 workflows, zero further broken or
vacuous assertions found.

Also clears two high transitive advisories the matrix flagged on one lane
— fast-uri GHSA-7p8r-x3mc-p8w7 and three ip-address SSRF/trust-boundary
issues. Both pre-date this branch: package-lock.json was untouched until
now, so the production tree was byte-identical to the base. Lockfile-only,
package.json unchanged, verified against a real npm ci install.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* chore(#2994): backfill changeset pr number to 3030

---------

Co-authored-by: sim <sim@local>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-03 19:59:58 -04:00

551 lines
25 KiB
JavaScript

'use strict';
// allow-test-rule: source-text-is-the-product — noSectionMarkerLeaksIntoEmittedArtifacts (#2930)
// asserts on the literal bytes of an EMITTED install artifact, which IS the
// deployed contract (a leaked `gsd:section` marker byte would ship to every user). This
// mirrors the test-matrix's own row-34/35 exemption from the "no source-grep" rule
// (50-test-matrix.md "No source-grep" note) — the ESLint rule itself only fires on
// readFileSync of a .cjs/.js/.ts SOURCE path, never on an installed .md artifact, so this
// annotation is documentation of intent, not a required suppression.
/**
* workflow-fragments-emission.install.test.cjs — 50-test-matrix.md rows 32-36
* (issue #2930, epic #1671 Phase 3).
*
* Real spawn-install coverage for `composeWorkflow`'s wiring into
* `bin/install.js`'s `copyWithPathReplacement` (ADR-1671 "Architecture and
* contracts": an engine-direct assertion is false-green for install
* behavior — only a real spawned installer proves bytes actually reach
* disk). The pure parser/composer itself is covered by
* tests/workflow-fragments.test.cjs (unit, rows 1-29/37) and
* tests/workflow-fragments.property.test.cjs (prop, rows 30-31).
*
* Each test builds and tears down its own tmp fixture(s) inline (no shared
* `before()` install cache) — independence per matrix row 38.
*
* ── The overlay technique (rows 33/36) ───────────────────────────────────
*
* Rows 33 and 36 need a spawned `bin/install.js` that reads a DIFFERENT
* `gsd-core/workflows/execute-phase.md` (malformed, row 36) or a different
* `gsd-core/bin/lib/workflow-fragments.cjs` (stubbed to identity, row 33)
* than this checkout's real files, without paying to copy the ~400 MB
* repository (mostly node_modules) for every run. `buildOverlayRepo` mirrors
* the repo tree with real directories (so `copyWithPathReplacement`'s own
* `entry.isDirectory()` / `entry.isFile()` Dirent checks — which do NOT
* follow symlinks — see the correct type) and HARD-LINKS every unmodified
* leaf file (not symlinks: a symlinked leaf file also fails an `isFile()`
* Dirent check elsewhere in the installer, verified empirically — "Failed
* to install agents: directory is empty" against a symlink-leaf overlay).
* Only `node_modules` and `.git` are symlinked at the top level (install.js
* never walks into either), which is what keeps the overlay build fast.
* Every overlay-spawned installer runs with `--preserve-symlinks
* --preserve-symlinks-main` as a defensive belt: with an all-hardlink leaf
* layout this checkout does not currently NEED symlink-preservation for
* correctness, but the flag is free insurance against a future install.js
* change that resolves a node_modules package by real path.
*/
const { test } = require('node:test');
const assert = require('node:assert/strict');
const fs = require('node:fs');
const os = require('node:os');
const path = require('node:path');
const crypto = require('node:crypto');
const { spawnSync } = require('node:child_process');
const { cleanup } = require('./helpers.cjs');
const { RUNTIME_META, runMinimalInstall, installerEnv } = require('./helpers/install-shared.cjs');
const { executionContextRefs } = require('../scripts/command-contract-helpers.cjs');
const { composeWorkflow } = require('../gsd-core/bin/lib/workflow-fragments.cjs');
const REPO_ROOT = path.join(__dirname, '..');
const PILOT_REL = path.join('gsd-core', 'workflows', 'execute-phase.md');
const PILOT_PATH = path.join(REPO_ROOT, PILOT_REL);
// plan-phase.md was the original #2930 pilot but was reverted to unmarked
// (chore/2930 retarget: it sits 36 B under the ADR-857 Phase-6 PRE_PHASE6
// gate and cannot absorb marker overhead) — it was a genuinely unmarked
// file again, so row 33 used it instead of plan-phase.md. #2993 (epic #1671
// Phase 6.2) now marks plan-phase.md itself (6 sections, the fragmentization
// this change ships), so it is no longer a valid "genuinely unmarked" fixture
// either — retargeted a second time to discuss-phase.md, which carries no
// gsd:section markers as of this change.
const UNMARKED_REL = path.join('gsd-core', 'workflows', 'discuss-phase.md');
const RUNTIMES = Object.keys(RUNTIME_META);
// ─── Overlay-repo builder (rows 33/36) ─────────────────────────────────────
const OVERLAY_SKIP_TOP = new Set(['node_modules', '.git']);
/** Hard-link a file, falling back to a real copy only if the two paths sit on
* different filesystems/devices (EXDEV) or linking is denied (EPERM) — both
* cross-platform-legitimate, unlike a symlink's Dirent type-detection gap. */
function linkOrCopyFile(src, dest) {
try {
fs.linkSync(src, dest);
} catch (err) {
if (err.code === 'EXDEV' || err.code === 'EPERM') {
fs.copyFileSync(src, dest);
} else {
throw err;
}
}
}
/**
* Build a throwaway mirror of REPO_ROOT with real directories throughout and
* every unmodified leaf file hard-linked, except the paths named in
* `fileOverrides` (POSIX-relative-path -> content string), which are written
* as real files. Returns the mirror's absolute path; caller must
* `fs.rmSync(..., {recursive:true, force:true})` it away.
*
* @param {{[relPath: string]: string}} fileOverrides
*/
function buildOverlayRepo(fileOverrides) {
const tmpRepo = fs.mkdtempSync(path.join(os.tmpdir(), 'gsd-2930-overlay-'));
const entries = Object.entries(fileOverrides).map(([relPath, content]) => ({
parts: relPath.split('/'),
content,
}));
function place(srcDir, destDir, pending, isTop) {
fs.mkdirSync(destDir, { recursive: true });
const grouped = new Map();
for (const e of pending) {
const [head, ...rest] = e.parts;
if (!grouped.has(head)) grouped.set(head, []);
grouped.get(head).push({ parts: rest, content: e.content });
}
for (const de of fs.readdirSync(srcDir, { withFileTypes: true })) {
if (isTop && OVERLAY_SKIP_TOP.has(de.name)) {
fs.symlinkSync(path.join(srcDir, de.name), path.join(destDir, de.name));
continue;
}
const srcPath = path.join(srcDir, de.name);
const destPath = path.join(destDir, de.name);
const overridden = grouped.get(de.name);
const leaf = overridden && overridden.find((s) => s.parts.length === 0);
if (leaf) {
fs.writeFileSync(destPath, leaf.content);
continue;
}
// fs.statSync follows symlinks (unlike Dirent.isDirectory()), so a
// symlinked source directory is still recursed as a REAL directory in
// the overlay — the property copyWithPathReplacement itself needs.
if (fs.statSync(srcPath).isDirectory()) {
place(srcPath, destPath, overridden || [], false);
} else {
linkOrCopyFile(srcPath, destPath);
}
}
}
place(REPO_ROOT, tmpRepo, entries, true);
return tmpRepo;
}
/** Spawn a (possibly overlaid) installScript at global scope. Does NOT
* assert success — callers decide (row 36 expects failure). */
function spawnGlobalInstall(installScript, runtime, extraArgs = []) {
const root = fs.mkdtempSync(path.join(os.tmpdir(), `gsd-2930-dest-${runtime}-`));
const args = [
'--preserve-symlinks',
'--preserve-symlinks-main',
installScript,
`--${runtime}`,
'--global',
'--config-dir',
root,
...extraArgs,
];
const result = spawnSync(process.execPath, args, {
cwd: root,
encoding: 'utf8',
env: installerEnv({ HOME: root, USERPROFILE: root }),
});
return { result, configDir: root, root };
}
/** Convert native path separators to POSIX forward slashes, unconditionally
* (never gate on `path.sep` — CONTEXT.md's path-separator-normalization
* rule). Windows installs embed the SAME root in more than one spelling:
* the `@`-ref / pathPrefix rewrites always emit posix-normalized
* forward-slash paths, while other embedded content can still carry the
* native backslash spelling. `root` itself (from `fs.mkdtempSync`) is a
* native-separator string, so comparing it against text verbatim only
* matches ONE of those spellings. */
function toPosixSlashes(value) {
return value.replace(/\\/g, '/');
}
/** Strip an install's own absolute root out of emitted text so two installs
* under DIFFERENT temp roots (different lengths, different runtime-name
* prefixes) can be compared byte-for-byte. Normalizes BOTH the text and the
* root to forward-slash spelling first, so every embedded spelling of the
* root collapses onto the SAME placeholder — leaving the comparison
* measuring only composeWorkflow's own contribution. */
function stripRoot(text, root) {
return toPosixSlashes(text).split(toPosixSlashes(root)).join('<ROOT>');
}
// ─── Row 32: emitted execute-phase.md shrinks by exactly the marker bytes ────
//
// Defect found and fixed inline while verifying (chore/2930 review; not one
// of the five assigned findings, but discovered incidentally): this test
// previously compared the RAW `composeWorkflow(source)` byte length directly
// against `fs.statSync(emittedPath).size`. That equality only holds when the
// OTHER rewrites `copyWithPathReplacement` also runs (the `~/.claude/` ->
// `pathPrefix` substitution, attribution stamping, per-runtime converters)
// happen to be byte-neutral — which they are NOT here: `runMinimalInstall`
// passes `--config-dir root` with `HOME=root` (root IS the target, not
// `root/.claude`), so `computePathPrefix` degenerates to the short literal
// `$HOME/` instead of the real-world `$HOME/.claude/`, shrinking the
// installed `~/.claude/` references by additional bytes unrelated to
// composeWorkflow. Proven with a real spawned install and reverted to
// confirm this reproduces on the UNMODIFIED tree, before this change. Fixed
// by isolating composeWorkflow's OWN contribution the same way row 33
// already does: compare two REAL installs of the pilot workflow, one via
// this checkout's real composeWorkflow and one via an identity-stubbed
// composeWorkflow, and assert the size DELTA equals exactly the marker
// bytes stripped — never an absolute emitted byte count, which conflates
// unrelated rewrites this module does not own.
//
// A second, independent contaminant surfaced fixing the first one: opencode
// embeds the install's own absolute configDir path into execute-phase.md
// content (same fact row 33/atRefContractStillResolvesAfterComposition
// documents for SKILL.md), and `runMinimalInstall`'s temp-dir prefix
// (`gsd-<runtime>-<scope>-`) is a DIFFERENT length than
// `spawnGlobalInstall`'s (`gsd-2930-dest-<runtime>-`) — so comparing RAW
// file sizes between the two installs bakes in a root-path-length delta
// that has nothing to do with composeWorkflow. Normalize each side's own
// root out of the text before measuring, exactly as row 33 already does.
test('emittedWorkflowShrinksByMarkerBytesForEveryRuntime', () => {
const source = fs.readFileSync(PILOT_PATH, 'utf8');
const composed = composeWorkflow(source, { sourcePath: PILOT_PATH });
const sourceBytes = Buffer.byteLength(source, 'utf8');
const composedBytes = Buffer.byteLength(composed, 'utf8');
const expectedMarkerBytes = sourceBytes - composedBytes;
assert.ok(
expectedMarkerBytes > 0,
'sanity: the pilot workflow must actually carry gsd:section markers to strip',
);
const identityStubRepo = buildOverlayRepo({
'gsd-core/bin/lib/workflow-fragments.cjs': 'module.exports = { composeWorkflow: (c) => c };\n',
});
try {
for (const runtime of RUNTIMES) {
const real = runMinimalInstall({ runtime, scope: 'global' });
const stub = spawnGlobalInstall(path.join(identityStubRepo, 'bin', 'install.js'), runtime);
try {
assert.equal(
stub.result.status,
0,
`${runtime}: identity-stub install must succeed\nstderr: ${stub.result.stderr}`,
);
const realPath = path.join(real.configDir, PILOT_REL);
const stubPath = path.join(stub.configDir, PILOT_REL);
assert.ok(fs.existsSync(realPath), `${runtime}: real install is missing execute-phase.md`);
assert.ok(fs.existsSync(stubPath), `${runtime}: identity-stub install is missing execute-phase.md`);
const realText = stripRoot(fs.readFileSync(realPath, 'utf8'), real.root);
const stubText = stripRoot(fs.readFileSync(stubPath, 'utf8'), stub.root);
const realBytes = Buffer.byteLength(realText, 'utf8');
const stubBytes = Buffer.byteLength(stubText, 'utf8');
assert.equal(
stubBytes - realBytes,
expectedMarkerBytes,
`${runtime}: emitted size delta (stub ${stubBytes} - real ${realBytes}, root-normalized) must equal exactly the marker bytes stripped (${expectedMarkerBytes})`,
);
} finally {
cleanup(real.root);
cleanup(stub.root);
}
}
} finally {
cleanup(identityStubRepo);
}
});
// ─── Row 33: an unmarked workflow emits byte-identical for every runtime ──
//
// "Byte-identical" here means identical to what the SAME runtime's install
// pipeline would emit WITHOUT the #2930 composeWorkflow wiring — not
// necessarily identical to the raw repo source, since path-prefix rewrites,
// attribution stamping, and per-runtime .md converters already ran before
// this change and still run today. Proven empirically per runtime by
// comparing two REAL installs of the SAME unmarked file: one through this
// checkout's real composeWorkflow, one through an overlay whose
// gsd-core/bin/lib/workflow-fragments.cjs is stubbed to plain identity — any
// difference is attributable ONLY to the compose wiring, never to an
// unrelated converter (which fires identically on both sides).
test('unmarkedWorkflowEmitsByteIdenticalForEveryRuntime', () => {
const identityStubRepo = buildOverlayRepo({
'gsd-core/bin/lib/workflow-fragments.cjs': 'module.exports = { composeWorkflow: (c) => c };\n',
});
try {
for (const runtime of RUNTIMES) {
const real = runMinimalInstall({ runtime, scope: 'global' });
const stub = spawnGlobalInstall(path.join(identityStubRepo, 'bin', 'install.js'), runtime);
try {
assert.equal(
stub.result.status,
0,
`${runtime}: identity-stub install must succeed\nstderr: ${stub.result.stderr}`,
);
const realPath = path.join(real.configDir, UNMARKED_REL);
const stubPath = path.join(stub.configDir, UNMARKED_REL);
assert.ok(fs.existsSync(realPath), `${runtime}: real install is missing discuss-phase.md`);
assert.ok(fs.existsSync(stubPath), `${runtime}: stub install is missing discuss-phase.md`);
// Normalize each side's own randomly-generated temp root out of the
// content before hashing: some runtimes (opencode) embed the
// install's own absolute configDir path in execution_context refs,
// and the two installs necessarily used DIFFERENT temp roots — an
// unnormalized compare would report a spurious mismatch driven by
// temp-path length, not by anything composeWorkflow's wiring did.
const realText = stripRoot(fs.readFileSync(realPath, 'utf8'), real.root);
const stubText = stripRoot(fs.readFileSync(stubPath, 'utf8'), stub.root);
assert.equal(
Buffer.byteLength(realText, 'utf8'),
Buffer.byteLength(stubText, 'utf8'),
`${runtime}: discuss-phase.md byte size drifted between real compose and identity-stub compose`,
);
const realHash = crypto.createHash('sha256').update(realText).digest('hex');
const stubHash = crypto.createHash('sha256').update(stubText).digest('hex');
assert.equal(
realHash,
stubHash,
`${runtime}: discuss-phase.md content drifted between real compose and identity-stub compose`,
);
} finally {
cleanup(real.root);
cleanup(stub.root);
}
}
} finally {
cleanup(identityStubRepo);
}
});
// ─── Row 34: no gsd:section marker survives into any emitted artifact ─────
test('noSectionMarkerLeaksIntoEmittedArtifacts', () => {
for (const runtime of RUNTIMES) {
const { configDir, root } = runMinimalInstall({ runtime, scope: 'global' });
try {
const emittedPath = path.join(configDir, PILOT_REL);
assert.ok(fs.existsSync(emittedPath), `${runtime}: emitted execute-phase.md is missing`);
const emittedText = fs.readFileSync(emittedPath, 'utf8');
assert.equal(
emittedText.includes('gsd:section'),
false,
`${runtime}: emitted execute-phase.md still contains a gsd:section marker token`,
);
} finally {
cleanup(root);
}
}
});
// ─── Row 35: ADR-0002 @-ref contract still resolves after composition ─────
//
// Two representative runtimes chosen to cover BOTH observed @-ref forms
// (empirically confirmed, #2930 dispatch): claude/cursor/codex rewrite to
// `@$HOME/...`, while opencode rewrites to a bare `@<absolute-path>/...`.
// Both installed SKILL.md files themselves pass through composeWorkflow too
// (as a no-op, being unmarked) — this proves that pass never corrupts or
// relocates the referenced workflow file.
/** Detect absoluteness from the token's own shape only — never from
* `process.platform` — so the same logic runs identically on every OS.
* Covers POSIX (`/...`), Windows drive-letter (`C:/...` or `C:\...`), and
* UNC (`\\server\share`) forms. */
function isAbsoluteRefTarget(candidate) {
return (
candidate.startsWith('/') ||
/^[a-zA-Z]:[\\/]/.test(candidate) ||
candidate.startsWith('\\\\')
);
}
function resolveExecutionContextRefTarget(token, root) {
const withoutAt = token.replace(/^@/, '');
if (isAbsoluteRefTarget(withoutAt)) return withoutAt; // already absolute (opencode form)
const stripped = withoutAt.replace(/^(?:~|\$HOME)\//, '');
return path.join(root, stripped);
}
test('atRefContractStillResolvesAfterComposition', () => {
for (const runtime of ['claude', 'opencode']) {
const { configDir, root } = runMinimalInstall({ runtime, scope: 'global' });
try {
const skillPath = path.join(configDir, 'skills', 'gsd-plan-phase', 'SKILL.md');
assert.ok(fs.existsSync(skillPath), `${runtime}: installed gsd-plan-phase SKILL.md is missing`);
const skillContent = fs.readFileSync(skillPath, 'utf8');
const refs = executionContextRefs(skillContent);
assert.ok(refs.length > 0, `${runtime}: SKILL.md has no execution_context @-refs to check`);
for (const { token } of refs) {
const target = resolveExecutionContextRefTarget(token, root);
assert.ok(
fs.existsSync(target),
`${runtime}: execution_context @-ref "${token}" resolved to "${target}", which does not exist on disk`,
);
}
} finally {
cleanup(root);
}
}
});
// ─── FIX 1 (chore/2930 review): composeWorkflow is scoped to gsd-core/workflows/ ──
//
// copyWithPathReplacement is the emit path for the ENTIRE gsd-core/ tree
// (skillSrc = path.join(src, 'gsd-core'), bin/install.js:10806-10809), not
// just gsd-core/workflows/. A non-workflow .md elsewhere under gsd-core/
// (e.g. gsd-core/references/) that merely DOCUMENTS the marker syntax with
// an unfenced, structurally-invalid example line must never be run through
// composeWorkflow — doing so would either throw (breaking install for an
// unrelated file class) or silently strip/mis-parse the documentation line.
// Proven here by overlaying BOTH a marked workflow (must still compose) and
// an EXISTING non-workflow reference doc (buildOverlayRepo can only replace
// the content of a real leaf file, not graft in a net-new path — see the
// module doc comment's overlay-technique note) rewritten to carry an
// intentionally-UNCLOSED marker-shaped line (would throw if composeWorkflow
// ever touched it) in the SAME install run.
test('nonWorkflowMarkdownWithMarkerShapedLineIsNotComposed', () => {
const markedWorkflow = '<!-- gsd:section id="a" when="always" -->\nbody\n<!-- /gsd:section -->\n';
const nonWorkflowDoc =
'# Marker syntax\n\nExample (deliberately unfenced and unclosed to prove non-composition):\n\n<!-- gsd:section id="x" when="always" -->\nnever closed on purpose\n';
const NON_WORKFLOW_DOC_REL = path.join('gsd-core', 'references', 'context-budget.md');
const overlayRepo = buildOverlayRepo({
'gsd-core/workflows/execute-phase.md': markedWorkflow,
[NON_WORKFLOW_DOC_REL.split(path.sep).join('/')]: nonWorkflowDoc,
});
let dest;
try {
dest = spawnGlobalInstall(path.join(overlayRepo, 'bin', 'install.js'), 'claude');
assert.equal(
dest.result.status,
0,
`install must succeed: a non-workflow doc's marker-shaped line must never reach composeWorkflow\nstderr: ${dest.result.stderr}`,
);
const emittedWorkflowPath = path.join(dest.configDir, PILOT_REL);
assert.ok(fs.existsSync(emittedWorkflowPath), 'emitted execute-phase.md is missing');
assert.equal(
fs.readFileSync(emittedWorkflowPath, 'utf8'),
'body\n',
'gsd-core/workflows/execute-phase.md must still compose (markers stripped)',
);
const emittedDocPath = path.join(dest.configDir, NON_WORKFLOW_DOC_REL);
assert.ok(fs.existsSync(emittedDocPath), 'emitted context-budget.md is missing');
assert.equal(
fs.readFileSync(emittedDocPath, 'utf8'),
nonWorkflowDoc,
'a non-workflow .md must pass through composeWorkflow untouched, byte-identical, including its marker-shaped line',
);
} finally {
cleanup(overlayRepo);
if (dest) cleanup(dest.root);
}
});
// ─── Row 36: a malformed marker fails install loudly, with no partial emit ─
// ─── Row 63 (50-test-matrix.md, issue #2932 Phase 5): extracting
// execute-phase.md's 3 sections must not perturb any OTHER workflow's
// emission — independence guard for the CRITICAL blast radius Phase 5's own
// design doc calls out. Pure-function check (no spawn needed): composeWorkflow
// is a documented no-op for every unmarked file, so any file other than the
// one Phase 5 migrates must still compose to itself, byte-identical. ────────
test('leavesUnmarkedWorkflowEmissionByteIdentical', () => {
const workflowsDir = path.join(REPO_ROOT, 'gsd-core', 'workflows');
// execute-phase.md (#2932 Phase 5), plan-phase.md (#2993 Phase 6.2), and
// the thirteen workflows #2994 (epic #1671 Phase 6.3) fragmentizes onto the
// marker grammar are the files this repo has fragmentized with gsd:section
// markers — all excluded here since composeWorkflow is deliberately NOT a
// no-op for them. This set is DELIBERATELY hardcoded rather than derived
// from "does the file contain a marker": deriving it would make the guard
// tautological — a marker that LEAKED into an unrelated file via a bad
// extraction would just get silently excluded instead of failing the
// independence check this test exists to enforce. Update this list by hand
// whenever a workflow is legitimately marked.
const MARKED_WORKFLOWS = new Set([
'autonomous.md',
'code-review.md',
'complete-milestone.md',
'discuss-phase-assumptions.md',
'docs-update.md',
'execute-phase.md',
'new-milestone.md',
'new-project.md',
'plan-phase.md',
'progress.md',
'quick.md',
'review.md',
'transition.md',
'update.md',
'verify-work.md',
]);
const workflowFiles = fs
.readdirSync(workflowsDir, { withFileTypes: true })
.filter((d) => d.isFile() && d.name.endsWith('.md'))
.map((d) => d.name);
assert.ok(workflowFiles.length > MARKED_WORKFLOWS.size, 'sanity: there must be more than the marked workflows on disk');
let checkedCount = 0;
for (const fileName of workflowFiles) {
if (MARKED_WORKFLOWS.has(fileName)) continue;
const filePath = path.join(workflowsDir, fileName);
const source = fs.readFileSync(filePath, 'utf8');
const composed = composeWorkflow(source, { sourcePath: filePath });
assert.equal(
composed,
source,
`${fileName}: emission drifted — a marked workflow's extraction must not touch any other workflow`,
);
checkedCount += 1;
}
assert.equal(
checkedCount,
workflowFiles.length - MARKED_WORKFLOWS.size,
'every workflow file except the marked ones must have been checked',
);
});
test('malformedMarkersFailInstallWithoutPartialEmit', () => {
const malformed = '<!-- gsd:section id="broken" when="always" -->\nnever closed\n';
const overlayRepo = buildOverlayRepo({ 'gsd-core/workflows/execute-phase.md': malformed });
let dest;
try {
dest = spawnGlobalInstall(path.join(overlayRepo, 'bin', 'install.js'), 'claude');
// stderr text is a child process's rendered prose, not a typed value
// this test can assert on across the process boundary (CONTRIBUTING.md
// "Prohibited: Raw Text Matching on Test Outputs" — err.reason is only
// reachable in-process; see tests/workflow-fragments.test.cjs's REASON
// assertions for the in-process equivalent of this same failure mode).
// Assert typed, observable facts instead: the install process exits
// non-zero, and no output file is written for the file that failed to
// compose.
assert.notEqual(
dest.result.status,
0,
`install must fail loudly on a malformed marker, got exit 0\nstdout: ${dest.result.stdout}`,
);
const emittedPath = path.join(dest.configDir, PILOT_REL);
assert.equal(
fs.existsSync(emittedPath),
false,
'a half-composed execute-phase.md must never be written when composition throws',
);
} finally {
cleanup(overlayRepo);
if (dest) cleanup(dest.root);
}
});