The --runtime parsing + help-text edit to gsd-core/bin/gsd-tools.cjs changes
the installed file's content (gsd-tools.cjs is installed and compared by the
golden snapshot, unlike gsd-core/bin/lib/ which is excluded). Regenerated via
UPDATE_GOLDEN=1; every runtime's manifest updates exactly one line (the
gsd-tools.cjs hash).
The legacy _isSkillsRuntime gate was a hardcoded isCodex || isCopilot || ...
roster; zcode was absent so the layout-driven skills-install path was skipped
(commands + agents installed, but no skills). Adding || isZcode would violate
AC#3 (no runtime === 'zcode' branches), so the gate is now derived from the
descriptor: a runtime takes the skills path iff its scoped artifactLayout
declares a skills kind. Behavior-preserving for all 15 existing runtimes
(verified by their install contracts); includes zcode (and any future skills
runtime) with zero per-runtime branches. opencode/kilo keep their specialized
combined path.
Regenerate the golden-install-parity fixtures: the non-zcode runtimes shift only
the shared model-catalog.json hash (zcode was added to the catalog); zcode now
includes its skills tree.
Resolve the zcode test cascade exposed by gsd-test:
- model-catalog.json: add zcode runtimeTierDefaults (null tiers, matching the
other profile-marker-only runtimes) so KNOWN_RUNTIMES stays parity with allRuntimes.
- runtime-name-policy: add zcode to GLOBAL_CONFIG_HOME_FRAGMENTS (~/.zcode) so
getGlobalConfigHomeFragment stops falling through to the .claude default.
- installer-migration-install: add the zcode fresh-install contract (flat-skills,
no settings, no package.json — same shape as trae).
- golden-install-parity: capture the zcode fixture via a standalone generator
script (not node --test — the gate stays gsd-test).
- capability-matrix.md: regenerate so zcode appears as a first-party row.
Fluidify the remaining count-pinned guards so adding a runtime no longer trips a
hand-pinned snapshot: gemini-runtime-removed (flag count derives from the
registry), non-claude-runtimes-registry-derivation (golden list derived from the
registry).
The compiled gsd-core/bin/lib/*.cjs modules are gitignored build artifacts
(ADR-457) shipped prebuilt in the npm tarball. A Claude Code plugin-marketplace
/ git-clone install materializes the repo tree directly and never runs
`npm run build:lib`, so those ~148 files are absent and every CLI command dies
at load with `Cannot find module './lib/cli-exit.cjs'`.
gsd-tools.cjs now calls ensureRuntimeBuild() before its ./lib requires: a new
committed bin helper that compiles the tree once on demand (lock-guarded,
incremental-cache-cleared, portable `node <tsc>` invocation) when the sentinel
cli-exit.cjs is missing, and surfaces an actionable error when TypeScript is
unavailable. The already-built npm path is a single fs.existsSync no-op.
ADR-457 (gitignored, build-at-publish) is preserved; no release-pipeline change.
Verified against a simulated marketplace checkout (148 build outputs removed):
the CLI self-heals and the result is byte-identical to `npm run build:lib`.
Closes#2002
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#2017: grant mcp__plugin_context7_context7__* for plugin-marketplace context7
The 8 context7-using agents granted only mcp__context7__* (standalone server
form). Claude Code's plugin-marketplace context7 install names tools
mcp__plugin_context7_context7__*, so the grant never matched and every
researcher/planner/executor silently lost doc lookup (fell back to WebSearch).
- 8 agents: add mcp__plugin_context7_context7__* alongside mcp__context7__*.
- scripts/research-profiles.cjs: update the researcher profile tools to match.
- tests/context7-plugin-grant-parity.test.cjs: regression guard — no agent
grants the standalone form without the plugin form.
Closes#2017
* docs(#2017): backfill changeset pr 2029
* fix(#2020): remove dead SDK file refs that triggered infinite find.exe on Windows
gsd-executor.md referenced sdk/src/query/QUERY-HANDLERS.md and reapply-patches.md
referenced sdk/dist/cli.js — both retired with the SDK (ADR-0174). AI runtimes
that locate doc refs via filesystem search ran find /, which on Git Bash for
Windows traverses the whole drive (14h+, orphaned find.exe, 4M+ handles).
- agents/gsd-executor.md: drop dead QUERY-HANDLERS.md ref.
- workflows/reapply-patches.md: drop dead sdk/dist/cli.js clause.
- tests/no-dead-sdk-refs.test.cjs: regression guard — no sdk/src|dist|handlers
file refs in agents/workflows/references markdown.
Closes#2020
* docs(#2020): backfill changeset pr 2027
* fix(#2019): planning-config.md global-learnings path ~/.gsd/learnings → ~/.gsd/knowledge
The features.global_learnings row pointed at ~/.gsd/learnings/ but the
implementation (src/learnings.cts, execute-phase.md) uses ~/.gsd/knowledge/.
Docs now match the code.
Closes#2019
* docs(#2019): backfill changeset pr 2026
* fix(#1921): preserve verify-work state across gap-closure + defer follow-ups
Resuming /gsd:verify-work after /gsd:execute-phase --gaps-only lost the
verification state: UAT ## Gaps still read status: failed even after their
fix plans executed, so verify-work re-diagnosed them as fresh blockers,
spawned a new gap plan, and reported only the new plan verified. A new
state contract links each gap to its fix plan so fixed gaps are recognized.
- verify-work.md gap YAML: stable gap_id (G-{phase}-{N}) per gap.
- plan_gap_closure planner prompt: each *-PLAN.md tags the gap_ids it
addresses in its frontmatter (gap_closure: true, gap_ids: [...]).
- new reconcile_gaps step (run at resume_from_file entry): marks a gap
status: resolved when its plan has a matching *-SUMMARY.md — fixed gaps
are not re-diagnosed and do not spawn new gap plans; a re-reported break
is treated as a fresh regression with a new gap_id.
- deferred-follow-up branch: a future-work idea (signals: 'later',
'next version', 'out of scope', ...) is captured to UAT ## Deferred
Follow-Ups instead of becoming a blocking gap/plan.
- workflow-size baseline recaptured (verify-work LARGE tier, 38221/61440).
Closes#1921
* docs(#1921): backfill changeset pr 2025
* fix(#1921): balance <step> tags in verify-work.md via 4-backtick outer fence
The plan_gap_closure step nested a ```yaml example inside a ``` block;
same-length fences made stripFencedCode close the outer block early,
swallowing the step's </step> (14 opens / 13 closes). Widen the outer
fence to 4 backticks so the nested yaml is contained. Regenerate golden
fixtures + size baselines.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Wire the two forward-declared mempalace.memory_mode modes so they actually
route recall/capture instead of silently behaving as `augment`:
- kg_backend: the palace temporal KG is the primary knowledge-graph source;
native .planning/graphs/ is the fallback. Non-KG drawer recall stays additive.
- replace: recall resolves through the palace as the source of truth; native
artifacts are the fallback.
Every mode stays onError:skip and default-resilient — an unreachable palace
degrades to native memory and GSD keeps writing .planning/graphs/, so no memory
is lost. Cross-mode .planning/graphs/ migration remains a documented open
question (PRD/ADR §17), out of scope here.
Surfaces updated (instruction-only contract): recall/capture commands (+ generated
skills), discuss/wave fragments, curator agent, capability.json schema. Docs:
how-to Step 3, CONFIGURATION, FEATURES, CONTEXT glossary. Regenerated
capability-registry, golden install-parity fixtures (mempalace hashes only),
agent-size-baseline. Added a routing-contract + cross-surface parity test.
Incidental (folded per no-defer rule): removed pre-existing unused imports
(spawnSync in capability-registry.test.cjs; fs in issue-498-package-identity.test.cjs)
that eslint flagged in/alongside the touched files.
Closes#2007
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* feat(#1825): configurable graphify graph location (graphify.graph_path)
Add a graphify.graph_path config key (.planning/config.json) that overrides
where /gsd-graphify query|status|diff read the knowledge graph, so one curated
umbrella-level cross-repo graph can serve multiple sibling projects without N
drifting ~5 MB mirror copies. Previously the graph location was hardcoded to
<cwd>/.planning/graphs/.
- src/graphify.cts: resolveGraphLocation(cwd, planningDir) honors the key
(resolved relative to project root; absolute paths honored via path.resolve);
falls back to the historical .planning/graphs/graph.json when unset/blank/
non-string (byte-identical). Wired into graphifyQuery, graphifyStatus,
graphifyDiff (snapshot travels with the configured graph via dirname), and
writeSnapshot. Configured-but-missing -> actionable error naming the path.
Build stays project-scoped (skill hardcodes the cp dest); umbrella graph is
built in the umbrella project, sub-projects only READ it.
- config-schema.manifest.json: register graphify.graph_path in validKeys.
- tests/graphify-graph-path.test.cjs: boundary matrix (unset byte-identical,
set+present reads configured graph not default, set+missing actionable error,
relative resolved vs project root, blank treated as unset, snapshot alongside
configured graph, diff from configured dir, build project-scoped) +
VALID_CONFIG_KEYS registration.
- docs: CONFIGURATION.md row, FEATURES.md REQ-GRAPH-06, CONTEXT.md module note,
.changeset (Added).
Closes#1825
* docs(#1825): backfill changeset pr number 2013
* fix(#1864): balance <step> tags in settings-advanced.md §8 Model Policy
§8 Model Policy ended with </step> but had no matching opening tag (5 opens /
6 closes), leaving it as loose inter-step content. Add the missing
<step name="model_policy"> opener so the section is a proper step.
- gsd-core/workflows/settings-advanced.md: add <step name="model_policy">
- tests/workflow-step-tag-balance.test.cjs: regression guard — every top-level
workflow must have balanced <step>/</step> (fenced code stripped), plus a
focused assertion that §8 is wrapped in model_policy.
- goldens + workflow-size baseline recaptured.
Closes#1864
* docs(#1864): backfill changeset pr 2014
* fix(#1865): runtime launcher honors CLAUDE_CONFIG_DIR
The gsd_run preamble resolved the Claude global install only at
$HOME/.claude/gsd-core/bin/, but the installer honors CLAUDE_CONFIG_DIR —
so a global install redirected via CLAUDE_CONFIG_DIR was invisible to every
gsd_run call (every command failed with 'gsd-tools.cjs not found').
The Claude resolver arm now uses ${CLAUDE_CONFIG_DIR:-$HOME/.claude},
matching the installer + the other runtimes' ${VAR:-default} pattern.
Default $HOME/.claude behavior is unchanged.
- _runtime-launcher.snippet.sh: Claude arm honors CLAUDE_CONFIG_DIR.
- sync-runtime-launcher.cjs re-run: 95 workflows/agents re-synced.
- review.md / discuss-phase.md: trimmed to stay under their byte budgets.
- runtime-launcher-parity.test.cjs: (A) substring updated for the new form
+ explicit #1865 assertion that the snippet honors CLAUDE_CONFIG_DIR.
- goldens + size baselines recaptured.
Closes#1865
* docs(#1865): backfill changeset pr 2024
* fix(#1936): reconstruct OpenCode review from JSON events; diagnosable empty-output stub
On a large review prompt, OpenCode's default `build` agent runs a few read
tool calls then ends its turn with zero output tokens (reason:"stop",
output:0), so `opencode run --format default` emits empty stdout. The reviewer
block redirected stderr to /dev/null and wrote a generic "failed or returned
empty output" stub — so the phase silently lost its second independent reviewer
with no diagnostic and no timeout.
Rewrite the OpenCode reviewer block to invoke `--format json` as the primary
call and reconstruct the review from the assistant `text` parts (jq). Capture
stderr to a `.err` sidecar (mirrors the Codex block). When the agent emits no
text, surface the stop reason, output-token count, and stderr so the failure is
diagnosable. Gate the stub on the extracted CONTENT, not the output file size —
an empty jq extraction still prints a lone newline that a `[ -s file ]` check
would treat as populated. Document the wall-clock timeout as a Bash-tool param
(macOS lacks GNU timeout; opencode has no native timeout flag).
review.md was already at the DEFAULT size-tier ceiling (40956/40960), so the
fix cannot fit without reclassifying it into the LARGE tier (it is a
multi-reviewer orchestration file that outgrew "focused single-purpose"; 43.4 KB
sits well under the LARGE high-water mark). Recapture the 16 golden-install
fixtures — the diff is exactly one review.md hash per runtime. Regression block
folded into review-default-reviewers-workflow.test.cjs (new bug-NNNN test files
are not accepted).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(#1936): add changeset
* test(#1936): property-test the OpenCode review jq reconstruction
Address the re-review's one actionable finding: the jq JSON-event → text
reconstruction had no fast-check property test.
Add tests/opencode-review-reconstruction.property.test.cjs. It extracts the two
shipped jq programs (OPENCODE_REVIEW, OPENCODE_DIAG) verbatim from
gsd-core/workflows/review.md and runs the real jq — not a reimplementation — so
the shipped logic is what gets tested. Properties: the reconstructed review
equals the newline-join of every assistant text part (order preserved); a stream
with no text part reconstructs to empty (drives the #1936 stub); null/absent text
parts are dropped, never rendered as "null". Plus example-based coverage of the
diagnostic edges the reviewer cited: missing .tokens.output and no step_finish
degrade to "?"; non-JSON stdout makes jq fail rather than masquerade as a review.
Verified the invariant has teeth (a comma-join jq fails the property).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(#1936): skip jq reconstruction property test when jq is absent
The property test shells out to `jq`, which GitHub's windows-latest runners do
not ship (macOS/Linux runners do). `execFileSync('jq')` therefore ENOENT-failed
the whole file on `test (windows-latest, *)`. Probe `jq --version` at load and
skip the suite when jq is not on PATH — the reconstruction logic is
platform-independent, so the assertions still run in full on every jq-present
runner (mirrors how golden-install-parity skips on win32).
Verified: jq present → 7 pass; jq removed from PATH → 7 skipped, 0 fail.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(#1936): skip jq reconstruction property test on Windows, not just when jq is absent
The prior guard skipped only when `jq` was absent from PATH — but the
windows-latest runners DO ship jq, so the suite still ran there and failed with
`jq: parse error: Invalid numeric literal` (confirmed from the CI job log). Root
cause is Node's child_process argument quoting mangling the jq program (it embeds
double quotes) on Windows, not the shipped review.md logic — the macOS/Linux legs
pass. Gate the suite on `process.platform === 'win32'` (still also skipping when
jq is absent), mirroring golden-install-parity's win32 skip. Logic is
platform-independent and fully asserted on every macOS/Linux CI leg.
Verified: macOS → 7 pass; simulated win32 → skips.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(#1928): remove sunset gemini cli runtime, redirect to antigravity
Google sunset Gemini CLI on 2026-06-18; Antigravity CLI is its official successor (already a first-class GSD runtime). Remove the gemini runtime from the enum (16->15), aliases, labels, config-home fragment, install path, converters (convertClaudeToGemini{Markdown,Toml,Agent}, convertSlashCommandsToGeminiMentions), capability descriptor, gemini-extension.json, RULESET.GEMINI.*, and the interactive menu (renumbered, no gap).
--gemini now prints an explicit deprecation notice citing the 2026-06-18 sunset and redirects to --antigravity (no silent alias, per the issue's Hyrum's-Law rejection). Antigravity is preserved throughout: its GEMINI.md contextFileName, .gemini/antigravity config home, the shared convertGeminiToolName/claudeToGeminiTools tool vocabulary, and the 'gemini' hookEvents dialect it declares. GEMINI.md retargeted as Antigravity's context file.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(#1928): backfill changeset PR number (#1996)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(#1928): drop Gemini CLI from issue templates (review nit)
Removes the sunset Gemini CLI runtime from the two GitHub issue-template
runtime lists that the removal PR missed, per @davesienkowski's review nit:
- feature_request.yml: 'Applicable runtimes' checkbox (a user could otherwise
request a feature for a runtime GSD no longer supports)
- bug_report.yml: 'Runtime' dropdown + the stale ~/.gemini/settings.json
retrieval-help line
Leaves the post-removal templates fully consistent with the Antigravity redirect.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(#1906): require node-test clean-fixture causation control
The node-test fail-first proof accepted a deceptive content-independent
negative test — one that reds merely because GSD_PROHIB_SUBJECT is set,
ignoring the subject's content — whenever no cleanFixture was supplied,
because #1346's causation control was opt-in. The proof's observed signal
(RED) thus diverged from its target (RED caused by content) by default.
Make the causation control mandatory for the node-test kind: a descriptor
that omits cleanFixture is un-provable (fail-closed), never accepted under
the weaker violation-only proof. When a clean fixture is present, fail-first
is proven exactly as before (RED on violation AND non-vacuous GREEN on clean).
The lint-rule kind is unchanged (its subject IS the linted file; no
GSD_PROHIB_SUBJECT indirection).
Breaking (Hyrum): a previously-green node-test prohibition with no clean
fixture now hard-gates — blast radius is zero in-tree (no node-test
prohibition ships today; only the lint-rule local/no-source-grep dogfood).
Supersedes ADR-1606 Decision 4 / ADR-550 #1346 addendum's opt-in.
Closes#1906
Claude-Session: https://claude.ai/code/session_017vYn26e3nkDNxcpty1ciPJ
* docs(#1906): supersede the #1346 opt-in causation control (mandatory for node-test)
Record the node-test mandatory-causation-control supersede across the
governing surfaces:
- ADR-1606 (the enforcement decision-of-record): addendum + Decision 4
annotated + the "Mandatory causation control — REJECTED" alternative
flipped to accepted (premise no longer holds: zero in-tree node-test
consumers).
- ADR-550: the 2026-06-21 #1346 "Why opt-in, not required" paragraph
marked SUPERSEDED, pointing at ADR-1606.
- spec-phase.md: check_clean_fixture is now REQUIRED for node-test
(was "optional").
- CONTEXT.md: PROHIB.enforce.causation predicate updated.
Regenerated the shipped-artifact cascade from the spec-phase.md edit
(+149 B, well under the 40960 cap): 16 golden-install-parity fixtures
and the workflow size baseline.
Refs #1906
Claude-Session: https://claude.ai/code/session_017vYn26e3nkDNxcpty1ciPJ
* docs: design spec for /gsd smart-entry command
Hybrid approach porting gsd-pi's smart-entry wizard to gsd-core:
deterministic classifier (gsd-tools smart-entry --json) + markdown
command/workflow with AskUserQuestion + --text fallback. Routing-first
('what now?' menu), 10 situations redesigned for gsd-core's phase loop.
* feat: add /gsd-start smart-entry command
State-aware front door adapted from gsd-pi's smart-entry wizard,
redesigned for gsd-core's markdown-first, multi-runtime architecture.
- src/smart-entry.cts: deterministic situation classifier (no-project,
paused, blocked, verify-failed, needs-first-phase, planning, executing,
verify-pending, idle-stranded, complete, unknown). Reads STATE.md,
ROADMAP.md, git, and verify signals; emits JSON the workflow consumes.
- gsd-tools.cjs: wire case + help listing.
- commands/gsd/start.md + gsd-core/workflows/gsd.md: thin markdown
dispatcher presenting an AskUserQuestion menu (with --text fallback for
non-Claude runtimes) and dispatching to existing commands. Falls back
to /gsd:progress if detection is unavailable.
- help.md: document /gsd:start (parity with bug-2954).
- tests: smart-entry.unit.test.cjs (classifier behavior across all
situations + priority + JSON shape) and gsd-workflow.structure.test.cjs
(markdown-layer invariants + every emitted command resolves to a real
slash command).
Spec: docs/superpowers/specs/2026-06-27-gsd-smart-entry-design.md
Note: command-contract (ADR-0002) requires a gsd:* prefix, so the bare
/gsd from the spec surfaces as /gsd-start.
* refactor: rename smart-entry command to /gsd:next
Rename the command from /gsd:start to /gsd:next per feedback. The
command file is now commands/gsd/next.md (name: gsd:next) and the
backing workflow is gsd-core/workflows/smart-entry.md (named for the
smart-entry classifier and gsd-tools smart-entry subcommand; does not
collide with the existing workflows/next.md, which is the progress
--next sub-workflow). help.md and the spec updated to match.
All affected tests (188) pass; lint:ci clean.
* fix: smart-entry reads real STATE.md schema (nested progress YAML + body Phase field)
Codex review found the classifier misread this repo's own STATE.md: it
looked only for scalar current_phase/total_phases frontmatter and body
fields named 'Current Phase'/'Total Phases', but real STATE.md stores
the phase as body 'Phase: N' and total_phases/percent under a nested
'progress:' YAML object. Both came back null, so active projects
(e.g. this repo at Phase 3 / verifying) wrongly classified as
needs-first-phase.
- detectSignals now reads total_phases + percent from nested progress{}
first, then scalar fm, then body; current_phase falls back to the
body 'Phase:' field (parseProsePhaseField lineage).
- Add regression tests against the real schema (nested progress YAML +
body Phase field) covering verify-pending + executing situations.
Verified against this repo: now classifies verify-pending (was
needs-first-phase). Coverage 93.25% lines / 86.99% branches.
* fix(workflow): tiered fallback when gsd-tools is broken (not just smart-entry)
Live test exposed a self-defeating fallback: when smart-entry --json
failed because gsd-tools itself was broken (missing
markdown-sectionizer.cjs), the workflow fell back to /gsd:progress —
which also depends on gsd-tools and would dead-end too.
Replace the single /gsd:progress fallback with a tiered recovery:
1. Probe gsd_run state-snapshot. If it ALSO errors, the whole tool
layer is down — read .planning/STATE.md directly with the Read tool
and synthesize a minimal situation + actions menu so /gsd:next stays
useful. Surface a rebuild hint.
2. Only if smart-entry alone is missing (older gsd-core), fall back to
/gsd:progress as before.
Matches the direct-read resilience the live agent already did by hand.
* docs: add gsd-next skill surface
* chore: trigger no-mistakes validation
* no-mistakes(review): Fix smart-entry phase ordering
* no-mistakes(review): Fix decimal smart-entry phase ordering
* no-mistakes(test): Fix smart-entry next test contracts
* no-mistakes(document): Docs synced for smart entry
* chore: add changeset fragment for #1798 (/gsd:next smart-entry workflow)
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* fix: shorten next.md description and update golden install parity fixtures
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* fix: update /gsd-next refs to /gsd:next in docs and add Smart Entry topic alias
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* chore: trigger no-mistakes validation
* fix: regenerate INVENTORY-MANIFEST.json for new /gsd-next files
Full CI caught that adding commands/gsd/next.md + gsd-core/workflows/smart-entry.md
left docs/INVENTORY-MANIFEST.json stale (not in the affected-test scope that
no-mistakes' test gate runs, so it surfaced in CI). Regenerated via
node scripts/gen-inventory-manifest.cjs --write; inventory-manifest-sync
test now passes.
* fix: add 'next' to core_loop cluster, update INVENTORY-MANIFEST, fix gates.md ref
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* fix: regenerate golden install parity fixtures for /gsd:next
Full CI (shard 3/3) caught that adding commands/gsd/next.md + the
smart-entry workflow/lib made the per-runtime golden install parity
fixtures stale across all 16 runtimes. Regenerated via
UPDATE_GOLDEN=1 node --test tests/golden-install-parity.test.cjs.
All 16 fixtures + inventory-manifest-sync now pass.
* Fix smart-entry verify-failed phase scoping and empty resolve shim step
Scope detectVerifyFailed to STATE.md's current phase so leftover higher
phase directories cannot force verify-failed routing. Move the gsd_run
shim resolver into the workflow resolve step so agents define gsd_run
before the detect step runs smart-entry.
* fix: recapture golden fixtures with updated gates.md hash (/gsd:next)
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* fix: recapture all 16 golden fixtures with updated smart-entry.md hash
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* chore: regenerate fixtures + inventory manifest after rebase onto next
Rebased onto next which adopted #1837 (package-version normalization to
<VERSION> in golden-install-parity hashes). Recaptured the golden fixture
that needed it (hermes), re-sorted INVENTORY-MANIFEST.json, and regenerated
the gsd-next / ns-workflow skill descriptions to match the command surface.
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
* refactor(#1787): delegate /gsd:next in-project advancement to gated /gsd:progress --next
Reconciles the /gsd:next smart-entry front door with the existing
/gsd:progress --next engine (davesienkowski review on PR #1798). The
classifier previously recommended /gsd:execute-phase directly for the
`executing` situation, bypassing workflows/next.md Route 0
(resume-incomplete-phase invariant, #160) and Gates 1-3 — reproducing the
duplication that got the old flat /gsd-next removed (#3054), plus a
correctness hazard (executing the recorded current phase while an earlier
phase is silently incomplete).
Now planning/executing/verify-pending recommend `/gsd:progress --next`
(single gated engine); the specific command stays an explicit secondary.
Off-path states (no-project, paused, blocked, verify-failed,
idle-stranded, complete) keep direct recommendations — smart-entry's
distinct value over --next. Adds docs/adr/1787-gsd-next-smart-entry.md and
a regression test locking the delegation contract.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* docs(#1787): avoid literal /gsd-next token in ADR (bug-3054 guard)
The repo-invariants #3054 guard bans the removed /gsd-next slash form in
docs surfaces. Refer to the removed command as `gsd-next` (prose) — the
historical reference is unchanged, just the banned token is dropped.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore: gitignore compiled host-integration-sdk + handshake-serialized .cjs
Pre-existing gap from #1683: these two src/*.cts modules compile to
gsd-core/bin/lib/*.cjs but were omitted from the per-file ignore list, so
`npm run build`/`npm test` left them as untracked build artifacts (dirty
tree + accidental-commit footgun). Adds them alongside their siblings
(host-integration.cjs, mcp-server.cjs, …). Found while finishing #1798.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test(#1787): lock per-situation action invariants for all 11 situations + ADR typo
Adversarial-review follow-ups:
- Add a test asserting every situation's action set has exactly one
recommended action, 1-4 unique-id /gsd:* actions (previously the
one-recommended/1-4 invariant was only sampled for 6 of 11 situations).
- Fix ADR typo: /gsd-progress → /gsd:progress.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#1798): split oversized test chunks so a slow shard can't trip the per-chunk timeout
Root-cause of the intermittent `full test (windows-latest, 22, shard 1/3)`
failure. It was NOT a leaked handle (the runner's kill message guesses that,
but --test-force-exit already exits leaks cleanly). Diagnosis:
- Ran every shard-1/3 file WITHOUT --test-force-exit + a 45s kill-timer:
zero hangs, zero leaks — every file self-exits. So no leaked handle / hang.
- CI activity profile: output kept flowing (slowly) right up to the 600.0s
kill — a dead hang would go silent. => pure slowness.
- Per-file timing: install-minimal-hooks.test.cjs is a 4987-line / 250-case
consolidation file doing dozens of real installs — 41s even on a fast Mac
(much worse on the slow Windows I/O path), plus an install-heavy cluster.
Mechanism: MAX_FILES_PER_CHUNK=180 packed the whole ~171-file shard into ONE
`node --test` chunk, so the entire shard's wall-clock ran against a single
600s per-chunk backstop. On slow Windows runners that single chunk crossed
600s and was killed mid-run — an intermittent false-negative gate that also
hits `next` directly.
Fix: lower MAX_FILES_PER_CHUNK 180 -> 90 so each shard splits into ~2 chunks,
each with its own fresh 600s budget and a fresh node process (also relieves
per-process memory pressure). Verified locally: shard 1/3 now runs as
chunk 1/2 (90 files) + chunk 2/2 (81 files), 5323 tests, 0 fail. Also made the
timeout kill-message name slowness as a cause instead of asserting a leak, so
the next debugger isn't sent hunting a nonexistent handle leak.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Codesmith <codesmith-bot@users.noreply.github.com>
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* enhance(#1872): document Claude Code advisor inheritance in model profiles
Add an "Advisor Tool (Claude Code)" section to
gsd-core/references/model-profiles.md: session-level advisor is inherited
by all GSD subagents and composes with the per-agent profile/tier system,
candidate executor/advisor pairings per profile (cost/quality/caching
claims attributed to Anthropic's advisor-tool docs, not asserted as GSD
behavior), when it is worth enabling vs not, and the session-level /
no-per-agent-control constraint linking anthropics/claude-code#73072.
Docs-only. Golden-install-parity fixtures recaptured for the edited
reference file (hash-only, one line per runtime).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(#1872): add changeset
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(#1872): use Documentation changeset type for docs-only change
Changeset type was `Changed`, which triggers the docs-required lint
(TRIGGERING_TYPES in scripts/lint-docs-required.cjs). This PR only
touches gsd-core/references/model-profiles.md, so there is no docs/
file to pair with and docs-lint failed. `Documentation` is the correct
type for a docs-only enhancement and is exempt from the trigger.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(#1872): put docs-exempt marker on its own line, revert to type Changed
The prior fix (type: Documentation) was invalid — parse.cjs ALLOWED_TYPES
is {Added, Changed, Deprecated, Removed, Fixed, Security}, so both
changeset-lint and docs-lint failed with invalid_type.
Real root cause of the original docs-lint failure: DOCS_EXEMPT_RE is
anchored to match the `<!-- docs-exempt: ... -->` marker only on its own
line, but the marker was tacked onto the end of the prose line, so it was
never captured (docsExempt: null) and the triggering `Changed` fragment
had no docs/ pairing -> fail_docs_missing.
Fix: keep the valid `type: Changed` and move the marker to its own line.
Verified locally: changeset-lint -> ok_fragment_present,
docs-lint -> ok (own-line marker parses to ok_fragments_exempt).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
* refactor(#1852): lazy-split plan-phase.md into steps/
Extract 3 self-contained, rarely-hit sections into gsd-core/workflows/plan-phase/steps/
via lazy 'Read and execute' pointers (mirrors execute-phase/steps/, ADR-1610 progressive
disclosure — no eager @-import): closed-phase-gate (1.5), prd-express-path (3.5),
windows-troubleshooting. Byte-invariant: plan-phase.md 94459 -> 89775 (-4684), each step
< 32 KiB anchor, resolved instructions unchanged.
Scope note: only 3 sections were extractable. plan-phase.md is guarded by a dense net of
content-presence tests (plan-bounce/enh-3209/phase6-planning-capabilities assert specific
sections/flags inline) that block extracting the larger blocks without also refactoring
those tests — durable low-80s headroom is deferred pending maintainer re-scope (issue #1852).
Cascade: size:baseline regen, 16 golden-install-parity fixtures regen, INVENTORY in sync.
Full plan-phase test surface green (4129/4130, 0 fail); lint:ci exit 0.
* docs(changeset): Changed fragment for #1934 (plan-phase lazy-split)
* docs(changeset): mark #1934 fragment docs-exempt (internal workflow refactor)
* fix(#1852): add gsd_run launcher preamble to prd-express-path step
The extracted prd-express-path.md calls gsd_run but the canonical launcher
preamble lived in the parent plan-phase.md — runtime-launcher-parity (#373)
walks workflows/ recursively and requires every .md using gsd_run to carry
exactly one preamble + the $HOME/.claude fallback arm (same as the existing
execute-phase/steps/ files). Injected via scripts/sync-runtime-launcher.cjs;
golden fixtures + size baseline regenerated. plan-phase.md unchanged (89775).
---------
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
The flattened install layout broke the third-party capability ecosystem in two
ways. Both are fixed at the host-version / installer boundary.
Gap 1 — host version read as 0.0.0. The running GSD version was resolved via
require('../../../package.json') (loader/source) and require('../../package.json')
(the gsd-tools CLI), which in the installed layout is the versionless CommonJS
marker → the fail-closed fallback reported 0.0.0, so `capability install` rejected
any manifest with a real engines.gsd range as "incompatible with GSD 0.0.0". For
runtimes that get no marker, and for local installs, that walked-up package.json
could even be the USER's own project, reporting a wrong version. Fix:
readHostVersion() (capability-loader.cts, capability-source.cts) and capHostVersion()
(gsd-tools.cjs) now prefer the authoritative gsd-core/VERSION the installer already
writes for EVERY runtime (mirrors resolveVersionFrom(), #1383), falling back to the
runtime-root package.json for the dev/source tree, then fail-closing. This fixes
every runtime and the actual `capability install` CLI path without touching the
marker package.json, so uninstall is unchanged (no data-loss surface).
Gap 2 — scripts/gen-capability-registry.cjs (+ its sibling
gen-loop-host-contract.cjs) were never copied by the installer, so the loader's
never-crash invariant discarded every overlay and fell back to the frozen
first-party registry (installed capabilities silently inert). Now copied,
uninstalled, and manifest-tracked exactly like fix-slash-commands.cjs (#1223).
Regenerates the 16 golden-install-parity fixtures to capture the two newly shipped
generator scripts and the gsd-tools.cjs change.
Regression tests (RED→GREEN): readHostVersion VERSION-first / fallback / fail-closed
resolution; an end-to-end `capability install` against a REAL installed layout
proving the engines gate sees the real host version, not 0.0.0; and a real-install
check that both generators are shipped and manifest-tracked.
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* feat(#1933): VS Code IDE reference host binding — completes Phase 5 IDE profile
The reference VS Code host binding composes the Phase-3 engine seams for the
ide profile (host-integration.cts PROFILE_BASELINES): active model via vscode.lm
(createModelAdapter active + sendRequest), engine-owned hook bus
(createHookBus engine — VS Code has no host bus), sandboxed-storage stateIO
(createStateIO sandboxed-storage + host backend — no fs / no child_process),
and the imperative adapter (engine-as-library). Command surface: palette/chat.
VS Code is extension-distributed (Marketplace), not file-projected onto a config
dir, so it intentionally has NO runtime descriptor / --vscode installer entry —
the extension IS the host (architecturally N/A for the descriptor/installer
model, not a deferral). Mock-friendly binding (vscode.lm + hostStorage injected)
so it is behaviorally testable without a live VS Code host.
Tests: IDE profileOf classification; full seam composition (active model routes
to vscode.lm; engine bus pub/sub; sandboxed state routes to backend; imperative
adapter; command surface); fail-closed construction.
* fix(#1933): correct vscode fail-closed test case matrix
Proves the Programmatic-CLI reference binding for pi via the ExtensionAPI
imperative adapter (#1682 AC): createImperativeAdapter({runtime:'pi'}) classifies
as imperative + composes the registry; pi axes (imperative + bun) classify as
'programmatic-cli'; the reference pi host-plugin binds GSD via ExtensionAPI
(registerCommand /gsd + registerTool gsd_invoke + on tool_call).
Reference plugin (tests/fixtures/pi-host-plugin.cjs) is CJS + mock-friendly so
it is behaviorally testable without a live pi runtime. Full --pi installable-
runtime integration (descriptor + installer + golden parity 16→17) is a larger
follow-up — intentionally NOT added to the runtime registry here.
* feat(#1682): OpenCode session.idle + opencode-subset dialect + Claude parity — Slice 1b/c
- Plugin (.opencode/plugins/gsd-core.js): handle session.idle (↔ Claude Stop
lifecycle point; no-op sentinel — state already persisted to .planning/).
Completes the compaction/idle pair (#1914 shipped compaction).
- Declare hookEvents: 'opencode-subset' in the OpenCode descriptor — the
reserved dialect now has a real consumer (no longer zero-consumer).
- host-integration.cts: add HOOK_EVENT_SURFACES + hookEventSurfaceFor() — the
pure consumer that resolves a dialect to its host-fireable event surface.
opencode-subset = session/tool/file subset with NO workflow-phase events
(engine owns phase sequencing; ADR-1239 §OpenCode binding).
- Tests: hookEventSurfaceFor unit tests (claude/gemini/opencode-subset/null);
plugin session.idle no-throw; compaction breadcrumb; opencode-subset surface
parity vs the plugin's handlers (Claude parity).
* fix(#1682): don't declare hookEvents on opencode (hooksSurface:none invariant)
The repo invariant couples runtime.hookEvents to the managed settings.json hook
surface: hooksSurface:'none' runtimes (opencode — plugin owns hooks) must NOT
declare hookEvents. Declaring 'opencode-subset' there violated 3 capability-
registry invariants + 2 install-plan golden masters + opencode golden parity.
The opencode-subset dialect is still IMPLEMENTED — just not via the legacy
descriptor field: hookEventSurfaceFor() (host-integration.cts) is its consumer,
and the OpenCode plugin consumes the subset events at runtime (session.idle
added here; compaction shipped in #1914). Declaring it on the descriptor would
require weakening the hooksSurface:none ⇔ no-hookEvents invariant (flagged for
decision).
* fix(#1682): refresh opencode golden parity for plugins/gsd-core.js (session.idle)
* docs(changeset): OpenCode session.idle + opencode-subset dialect (#1682)
* docs(changeset): backfill PR #1930
* feat(#1914): OpenCode native plugin integration (Option 1 file-copy)
Ship a native OpenCode plugin (.opencode/plugins/gsd-core.js) plus the
installer step that delivers it, so GSD's lifecycle hooks run on OpenCode.
OpenCode declares hooksSurface:'none', so GSD's hook scripts already ship to
<configDir>/hooks/ but nothing invokes them; the plugin bridges OpenCode's
event bus onto those scripts as subprocesses (prompt/read/worktree/workflow
guards, injection scanner, context monitor).
Distribution is Option 1 (file copy) per the #1914 triage decision: no
scripts.build rename, no prepare/prepack removal. package.json gains
main + .opencode in files[] for discovery.
Corrected against OpenCode's docs + loader source (not the reference branch):
- Auto-discovery globs {plugin,plugins}/*.{ts,js} — .cjs is never matched, so
the installed adapter must be .js (config dir carries {"type":"commonjs"}).
- No opencode.json plugin-array patch — that array is npm-only; local files
are auto-discovered.
- REPO_ROOT is resolved by walking up to the dir holding hooks/ + gsd-core/,
correct for package tree, global install, and local install.
- Config-hook registration is gated (IS_PACKAGE_TREE) so it never
double-registers commands/agents/skills already delivered by native copy.
Also fixes an incidental .gitignore drift: 8 ADR-1239 .cts-generated .cjs
artifacts were untracked-and-not-ignored (leak risk) — now ignored.
Tests: tests/opencode-plugin-adapter.test.cjs (14, pure helpers + real
subprocess bridge against stub hooks); golden-install-parity regenerated.
Green on Mac + Linux (gsd-test): 0 failures.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#1914): harden OpenCode plugin export shape + advisory accumulation (adversarial findings)
Address Codex adversarial-review findings:
- HIGH: export `{ id, server }` could trip OpenCode's loader
(`for (entry of Object.values(mod)) getServerPlugin(entry)` throws on a
non-extractable value). Make `id` NON-ENUMERABLE and assign module.exports
from a variable (not a literal) so no string `id` is ever iterated — verified
loader-safe under real import(pathToFileURL) (default + module.exports alias,
both objects with .server; no bare id string).
- MEDIUM: sequential advisory hooks clobbered output.metadata._gsdAdvisory;
now accumulate into an array.
- LOW: resolveRepoRoot fallback returned ".." while the comment said "../.." —
aligned to "../.." (package-tree depth).
Tests: added a faithful loader-loop emulation (raw CJS + ESM namespace views),
advisory-accumulation, and a real bin/install.js copy→manifest→uninstall
integration test. golden regenerated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* docs(#1914): add changeset fragment for OpenCode plugin integration (PR #1923)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#1914): Windows — assert rewritten path with string include, not path-regex
The Read content-rewrite test built a RegExp from `path.join(root,'gsd-core')`.
On Windows the backslashes in the path are interpreted as regex escapes, so the
assertion never matched and `test (windows-latest, 24)` failed — even though the
adapter rewrote the path correctly. Replace the RegExp with a separator-agnostic
`String.includes` check (the repo's no-path-literal-in-assert concern).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#1863): use named flags for state.* calls in executor + workflows
The named-only state-command router (parseNamedArgs) silently drops
positional args, so state.cjs threw its required-arg error and
metrics/decisions/blockers/session continuity were never recorded.
Convert record-metric / add-decision / add-blocker / record-session in
agents/gsd-executor.md to the named-flag form (mirroring execute-plan.md),
and fix the two remaining positional record-session calls in
gsd-core/workflows/milestone-summary.md and forensics.md. Recapture the
golden-install-parity fixtures and size baselines for the edited files.
Also fix a pre-existing detached-rebuild handle leak in
tests/graphify-auto-update.slow.test.cjs: three dispatch tests returned
after observing only the synchronous "running" status without awaiting the
detached rebuild's terminal state. That leak was latent until the new
#1863 regression block's added runtime shifted --test-force-exit timing
and surfaced it as a non-zero chunk exit. The three tests now await
terminal status via the file's existing waitForBuildStatus helper.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(#1863): add changeset
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
* fix(#1716): route resume_from_file to complete_session when no pending tests remain
When a UAT session has status:partial with blocked_count>0 and pending_count==0 (all remaining tests are blocked, none are pending), resume_from_file found no [pending] test and terminated silently — never routing to complete_session. This blocked the issues==0 auto-transition path even when there were zero code defects.
Guard clause added immediately after the find-pending step: if no [pending] test is found, route to complete_session. complete_session then correctly sets status:partial (because blocked_count>0) without presenting further tests.
Closes#1716
* chore(#1716): add changeset fragment and regenerate golden-install-parity fixtures
Changeset fragment for PR #1722 (type: Fixed).
Golden-install-parity fixtures regenerated for all 16 runtimes — the workflow fix shifts verify-work.md's byte-stable hash in the golden manifest. Regenerated via UPDATE_GOLDEN=1 node --test tests/golden-install-parity.test.cjs.
Follow-up to #1919 (archive-then-remove core). Closes the remaining #1871
acceptance criteria so phase history is preserved across the full milestone
lifecycle, not just at phases.clear:
- #2 src/milestone.cts + src/phases-command-router.cts: extract shared
archivePhaseDirectories() helper; add cmdPhasesArchive (the previously
half-wired phases.archive alias now routes instead of erroring Unknown).
- #4 gsd-tools.cjs + src/milestone.cts: milestone complete archives phase
dirs by default (--no-archive-phases opts out; --archive-phases is now a
harmless no-op). complete-milestone.md updated to drop the redundant manual
Yes/Skip archive prompt.
- #3 gsd-core/workflows/new-milestone.md: §6 stages the archive move + source
removal (git add .planning/milestones/ .planning/phases/) in the same commit
as the milestone start, so the archive lands atomically — no orphaned
uncommitted deletions, no un-archived dirs inherited.
- docs/CLI-TOOLS.md (+ ja/zh/ko/pt) + help/modes/full.md: flag accuracy.
- tests: phases archive command (#2) + milestone complete default archive /
--no-archive-phases opt-out (#4). Goldens + workflow size baseline refreshed.
Closes#1871
Each of the 22 consumer agents now self-loads its configured agent_skills
in its mandatory init step, so .planning/config.json agent_skills.<type>
reaches the agent on every runtime — including Cursor and /gsd-autonomous,
where Skill()-delegated workflow bash init did not reliably execute.
- gsd-core/references/agent-skills-bootstrap.md: shared contract
(query + Read + dedup guard that skips when <agent_skills> is already
in the prompt, so Claude's orchestrator-side injection never doubles)
- 22 agents/gsd-*.md: one self-load line naming the agent's own type
- gsd-core/workflows/autonomous.md: note that delegated agents self-load
- tests/agent-skills-bootstrap.test.cjs: regression + parity (CONSUMER_AGENTS
bijection + fast-check property) — Generative-Fix-Divergence guard
- docs: ADR-1866, CONFIGURATION dual-injection How It Works, INVENTORY
row, Changed changeset
Closes#1866
execute-phase.md and gsd-ai-researcher.md are installed artifacts; their edits
shift install hashes and file sizes. Diff is scoped to those two files' hashes.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The sonnet-5 catalog change alters model-catalog.json and settings-advanced.md,
so the per-runtime install-hash fixtures are recaptured (UPDATE_GOLDEN=1). Only
those two file hashes change per runtime.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#1528): drop next-phase guidance from security-blocked verify-work presentation
When security enforcement blocks phase advancement (no SECURITY.md produced),
the verify-work presentation told the user advancement was blocked but still
offered `/gsd:plan-phase {next}` and `/gsd:execute-phase {next}`, competing
with the current-phase fix. Remove those two next-phase lines so the blocked
state routes only to the current-phase resolution (secure-phase, ui-review).
The post-transition presentation — reached only after the completion contract
passes — still offers next-phase planning, which is the correct place for it.
Regression coverage added to tests/ui-review-next-guidance.test.cjs: the
security-blocked block must not offer next-phase actions, and the
post-completion block must still offer them. Regenerated workflow size baseline.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(#1528): add changeset for security-blocked next-phase fix
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(#1528): recapture golden-install-parity fixtures for verify-work.md change
Rebased onto next; verify-work.md's installed hash changed across all 16
runtime fixtures. Diff confined to the single gsd-core/workflows/verify-work.md
key per runtime. Assert mode 16/16 green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
CHANGELOG.md contains historical version strings from prior releases.
The PKG_VERSION normalization applied to all files only replaces the
*current* package version, so locally (PKG_VERSION=1.6.0) the normalization
mutates CHANGELOG.md content (1.6.0 appears in old entries), producing a
different hash than in CI (PKG_VERSION=1.7.0-rc.1, which doesn't appear
in CHANGELOG.md). The hash can never match across build contexts.
Add gsd-core/CHANGELOG.md to VOLATILE_FILES so it is excluded from the
parity manifest. It's release documentation — not a functional install
artifact — and changes with every release anyway.
Regenerate all 16 golden fixtures to remove the stale CHANGELOG.md entry
and establish the new baseline.
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The rc release step runs `npm version X.Y.Z-rc.N` before tests, which
rebakes the current version string into hook files and `gsd-core/VERSION`.
Without normalization, every golden parity hash differed post-bump and the
entire test suite failed with 16 golden failures — even though no files
actually changed in a semantically meaningful way.
Add `PKG_VERSION` normalization (`.split(PKG_VERSION).join('<VERSION>')`)
alongside the existing `<HOME>` root normalization so the goldens are
stable across version bumps. Regenerate all 16 fixtures with the new
normalization to establish the new baseline.
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(verify-phase): honest verifier — abstain (insufficient_spec) on non-inferable backstop truths (#1154)
Carry the edge-probe's existing `backstop` (non-inferable) tier through the
plan-phase projection as a structured flat-scalar marker instead of a prose
parenthetical, and make verify-phase abstain -> human_needed (never silent-pass)
on a backstop truth it cannot confirm with explicit evidence. Truth-axis mirror
of #644's prohibition judgment-tier (ADR-550 D4).
Engine (deterministic, CI-tested per ADR-550 D5 — never the LLM verdict):
- src/probe-core.cts: truthStatement/truthVerification normalizers, projectTruths
(conservative serializer), dispositionForUnverifiableTruth (backstop+no-evidence
-> unverified/flagged/insufficient_spec; backstop+evidence -> green; inferable
-> green, the over-abstention guard).
- src/roadmap.cts: coerceTruthToString now reads `statement` first so an object-form
backstop truth is surfaced, not dropped (Hyrum backward-compat for truth-readers).
Workflow/agent/docs: plan-phase emits the structured marker (flat scalar, ADR-550
#1278); verify-phase + gsd-verifier add the abstain arm; new references/honest-verifier.md;
FEATURES/COMMANDS document insufficient_spec; ADR-550 amended (truth-axis D4 mirror).
Decisions adopted (trek-e review): insufficient_spec feeds existing human_needed with a
distinguishable reason (no new VERIFIER_STATUS); changeset Changed; round-trip parity
test; abstain-on-unconfirmed-backstop regression test red-first.
Implementation notes (deviations from the issue's proposed file list, verified live):
- frontmatter.cts needs no change — its flat parser already round-trips object-form truths.
- verify.cts needs no change — it grades artifacts/key_links structurally; truths are
LLM-graded at the workflow layer, so consumption lives there + the deterministic helper.
- No CJS<->SDK hand-sync — the SDK seam was retired (ADR-0174); src/*.cts is sole source.
Regenerated artifacts: golden-install-parity fixtures, INVENTORY-MANIFEST, size baselines.
* chore(#1154): add changeset (Changed) for honest verifier
User-facing changelog fragment for #1738. Typed `Changed` (not `Added`) per
trek-e review condition 3 — the verify behavior shifts for backstop-bearing specs
(a confident silent `passed` becomes `human_needed`), which is user-visible even
though the schema marker is additive.
* docs(#1154): score-formula also excludes abstained insufficient_spec truths (review nit-1)
trek-e review nit: the verify-phase score sentence said PRESENT_BEHAVIOR_UNVERIFIED
truths were "the only ones excluded" from verified_truths. Post-#1154 an abstained
`insufficient_spec` backstop truth is also excluded (it is not ✓ VERIFIED and routes
to human_needed). Behavior was already correct; this tightens the wording.
Regenerated golden-install-parity fixtures + workflow-size baseline for the touched
verify-phase.md. (Nit-2 — a dedicated insufficient_spec_items frontmatter list — is
intentionally not taken: the current design is ADR-550-D4-conformant, the abstain
cause rides as a distinguishable report reason, and adding it would exceed the
approved scope.)
---------
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
* fix(#1778): use 1.6 named-flag frontmatter.set form in thread workflow
The thread workflow's CLOSE and RESUME branches called frontmatter.set with
the pre-1.6 fully-positional shape (frontmatter.set <file> <field> <value>).
Since 1.6 the dispatcher (gsd-tools.cjs) parses the file positionally and
reads field/value from the named flags --field/--value via parseNamedArgs;
the positional form leaves field/value undefined, cmdFrontmatterSet errors
'file, field, and value required', and the status/updated writes are
silently skipped. Closing a thread never marked it status: resolved and
resuming never marked it status: in_progress.
Switch all four sites (CLOSE status+updated, RESUME status+updated) to the
1.6 hybrid form that verify-work.md already uses:
frontmatter.set <file> --field <field> --value <value>
Add a regression test with three guards: (1) behavioral — the named-flag
form writes the field while the positional form errors with the documented
message and does not mutate the file; (2) workflow parity — no workflow
under gsd-core/workflows/ emits the positional form, so a future edit that
reintroduces it anywhere fails CI; (3) thread-specific — CLOSE writes
status: resolved and RESUME writes status: in_progress via the named flags.
* docs(#1778): add changeset fragment for thread workflow frontmatter fix
* docs(#1778): fix unclosed inline-code backtick in changeset fragment
* fix(#1778): move regression into owning test + regen baselines
lint-regression-test-names rejects new bug-NNNN-*.test.cjs files; move the
#1778 regression (behavioral named-vs-positional + workflow-parity scan +
thread CLOSE/RESUME assertions) into tests/frontmatter-cli.test.cjs, the
canonical home for frontmatter CLI regressions, and delete the standalone
file. frontmatter-cli.test.cjs already carries the allow-test-rule exemption
for workflow .md content tests.
gsd-core/workflows/thread.md ships to every runtime and is size-tracked, so
recapture the 16 golden-install-parity fixtures (thread.md hash) and the
per-file workflow size baseline (thread.md 12400 -> 12464) via UPDATE_GOLDEN=1
and npm run size:baseline.
* fix(#1747): register four search-provider keys in the config schema
buildNewProjectConfig emits seven search-provider availability flags and
research-provider.cts providerAvailability() consumes all seven, but only
three were registered in VALID_CONFIG_KEYS (config-schema.manifest.json).
config-loader.cts then printed an 'unknown config key(s)' warning for the
four unregistered keys (tavily_search, ref_search, perplexity, jina) on
every freshly generated .planning/config.json.
Register the four missing keys in the schema manifest and document them
alongside brave/exa/firecrawl in CONFIGURATION.md. Add a regression test
plus a structural drift guard that requires every config-driven
research-provider flag to be in VALID_CONFIG_KEYS, so a future provider
addition cannot silently reintroduce the drift.
* fix(#1747): move regression into owning test file + add changeset
lint-regression-test-names rejects new bug-NNNN-*.test.cjs files; move the
#1747 regression (four provider keys in VALID_CONFIG_KEYS + provider-flag
drift guard) into tests/bug-2530-valid-config-keys.test.cjs, the canonical
home for VALID_CONFIG_KEYS regressions, and delete the standalone file.
Add the missing .changeset fragment — config-schema.manifest.json lives
under gsd-core/ (user-facing), so changeset-lint requires a fragment.
* test(#1747): regenerate golden-install-parity fixtures for schema change
Adding four provider keys to config-schema.manifest.json shifts its shipped
content hash (65dea848 -> 7d398e94); recapture all 16 runtime fixtures via
UPDATE_GOLDEN=1. Each fixture changes exactly one line — the manifest hash.
* fix(#1772): read full multi-line command in graphify-update hook Gate 2
The PostToolUse hook joined tool_name + newline + tool_input.command and
extracted the command with sed -n '2p' — line 2 only. Agent runtimes
(Claude Code's Bash tool among them) routinely emit HEAD-advancing commits
as multi-line scripts ('cd /path', then 'git add', then 'git commit …'), so
line 2 is the 'cd', Gate 2's *"git commit"* match failed, and the rebuild
silently no-op'd on real commits despite graphify.auto_update: true.
Capture line 2 through EOF (sed -n '2,$p') so the case glob sees the full
multi-line command string. Single-line behavior is unchanged (the match
only widens); non-HEAD-advancing multi-line commands still no-op cleanly.
Regression tests cover multi-line commit/merge/pull dispatch plus a
multi-line no-op no-regression guard.
* docs(#1772): add changeset fragment for graphify-update multi-line fix
* test(#1772): regenerate golden-install-parity fixtures for hook change
gsd-graphify-update.sh ships to 9 graphify-aware runtimes; widening the
sed range (2p -> 2,$p) shifts its shipped hash. Recapture the 9 affected
fixtures via UPDATE_GOLDEN=1 — each changes exactly one line (the hook hash).
Rebased onto next; review.md's installed hash changed across all 16
runtime fixtures. Diff confined to the single gsd-core/workflows/review.md
key per runtime. Assert mode 16/16 green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(#1517): support custom reviewer instances for /gsd:review
Add a bounded review.reviewer_instances config surface so one model-capable
adapter (e.g. opencode) can run as several independent reviewer identities in a
single /gsd:review pass. Instances participate only via review.default_reviewers,
expand before built-in slugs, are available iff their cli is detected, and a
non-matching entry is a hard error (typo must be loud). >=2 same-cli instances
emit a shared-adapter caveat in REVIEWS.md. Default path with no instances is
byte-for-byte unchanged.
Single-source instance->cli resolution lives in resolveReviewerSelection /
normalizeReviewerInstances (parity-locked in
tests/review-reviewer-instances.test.cjs). cli validated against
KNOWN_REVIEWER_SLUGS only (never arbitrary shell); model/agent opaque, never
shell-interpolated.
Closes#1517
* chore(#1517): backfill changeset pr:1766
---------
Co-authored-by: review-bot <review-bot@gsd>
* feat(#1754): CLI version-skew detection — warn when a global install shadows project-local GSD
Addresses #1754 (approved-enhancement). Detects when the running gsd-tools.cjs
is outside the project root while a project-local install exists — the shadowing
scenario from #1748 where a stale global canary CLI (retired @gsd-build/sdk)
silently overrides project-local GSD.
Implementation (Node CLI entry-point, not shell snippet — avoids bloating 93
workflow files past their size caps):
- src/cli-skew-check.cts: pure function checkCliSkew({resolvedPath, projectRoot,
projectLocalExists}) → string|null. Compares paths via path.relative; returns
a warning when the resolved CLI is outside the project root AND a project-local
install exists. Includes @gsd-build/sdk removal hint when the path matches.
No I/O (pure), no gsd-sdk literal (avoids bug-2801 lint).
- gsd-core/bin/gsd-tools.cjs: wired at startup via the existing findProjectRoot
resolver. Non-blocking (try/catch; advisory stderr warning, never gates).
- eslint.config.mjs: registers the new ADR-457 generated artifact in the ignores.
- tests: 6-case suite (skew/no-skew/legacy/normalization); all green.
- Golden fixtures regenerated (UPDATE_GOLDEN=1) for the new compiled artifact.
- docs/how-to/update-gsd.md: Diátaxis reference note for the skew warning.
Full suite: 3354 pass, 0 regressions (1 pre-existing local AGENTS.md failure).
lint:ci green.
Closes#1754
* chore(#1754): backfill changeset pr placeholder
* chore(#1754): regenerate INVENTORY-MANIFEST for the new cli-skew-check source module
---------
Co-authored-by: review-bot <review-bot@gsd>
* refactor(#1734): extract install engine from bin/install.js (ADR-1239 Phase B deep move)
Relocate the runtime-artifact install cluster out of the 12,490-line
bin/install.js into a dedicated src/install-engine.cts -> install-engine.cjs:
installRuntimeArtifacts, uninstallRuntimeArtifacts, installOpencodeFamilySkills,
and their cluster helpers (_copyStaged, snapshot/restore, legacy migration,
GSD-entry pruning, preserve/restoreUserArtifacts, OpenCode-family converters,
USER_OWNED_ARTIFACTS).
- bin/install.js imports the engine and re-exports the moved symbols for
back-compat; getCommitAttribution STAYS in install.js (impure config I/O +
argv explicitConfigDir global) and is injected via a resolveAttribution param.
- 17 test files migrated to import the moved symbols from the engine.
- Bookkeeping: eslint built-artifact ignore, .gitignore, INVENTORY manifest+row,
CONTEXT.md Install Engine Module glossary seam.
Behaviour-preserving: install output is byte-identical for all 16 runtimes
(golden-parity harness #1730) — the only delta is the new install-engine.cjs
file shipping in the installed gsd-core/bin/lib/ tree.
Closes#1734
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore(#1734): backfill changeset PR number (#1735)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: review-bot <review-bot@gsd>
ADR-1239 Phase B (parent #1679). Safety net for the upcoming engine deep-move:
captures the COMPLETE emitted install output of all 16 runtimes as a golden
baseline so the move PR can prove byte-identical parity.
tests/golden-install-parity.test.cjs spawns the real installer per runtime into
a temp HOME, normalizes the temp path to <HOME>, excludes the two volatile
metadata files (gsd-file-manifest.json, gsd-install-state.json — the only
run-to-run variance after normalization, empirically), SHA-256s every remaining
file, and asserts the manifest matches tests/fixtures/golden-install-parity/<rt>.json
(8,957 hashes total). UPDATE_GOLDEN=1 regenerates; mismatches list
added/removed/changed paths. Non-vacuous (corrupting a hash fails the run).
Closes#1730
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* fix: recurse test discovery so subdir test suites actually run
scripts/run-tests.cjs discovered tests with a flat readdirSync(testDir),
silently excluding tests/observability/ (4 files), tests/dispatch/ (1) and
tests/installer-migrations/ (1) — 94 passing tests — from `npm test` and all
CI lanes. Walk the tree recursively (relative subpaths preserved), classify
suites by basename, and add a fail-on-zero-executed guard for suite/default
runs (escape hatch GSD_ALLOW_EMPTY_SUITE=1) while preserving the empty
--files/--files-from path the CI inert lane relies on.
Unit suite 735 -> 741 files; surfaces ADR-227's observability/dispatch seam.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: retire 5 verified-worthless tests
Adversarial verification confirmed these 5 prove nothing — their coverage is
provided more strictly elsewhere:
- enh-2790 'has a name: field' spot-checks (command-contract enforces /^gsd[:-]/)
- command-routing-hub duplicate construct + duplicate ERROR_KINDS assertions
- no-cjs-sdk-handsync-tooling (guarded files that never existed on main; bug-190
covers the real retired SDK artifacts)
- runtime-artifact-layout cline edge case (subsumed by the explicit-global test
and bug-782-cline-skills-emission)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: add ADR-218 release version-validation coverage
ADR-218 (reject leading-zero versions like 1.01.0; npm duplicate pre-check) had
zero tests — the logic lived only in release.yml bash. Add a test that extracts
the actual rejection regexes from the workflow and exercises them against a
boundary table (leading-zero/malformed rejected, valid accepted) plus structural
wiring assertions. Goes red if the regex is reverted to [0-9]+.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: redesign weak tests into behavioral, deterministic assertions
Per the ADR test audit, rewrite 27 weak test files (test-only, no source
changes) so each can go red for the defect it guards:
- kill pass-always assert.ok(true) placeholders (research-cli, worktree-baseref,
bug-260 security guard, eslint-rules x24, clusters '|| true')
- replace source-text grep with behavioral calls (install Kilo, sh-hook-paths,
plan-review-convergence) and add a repo-layout governance test
- de-flake real-clock/Math.random coupling (phase last_updated, bug-3707 mtime,
context-utilization property, feat-3594)
- fix independence/shared-state violations (bug-492 singleton, issue-844 tmpRoot,
core reapStaleTempFiles, active-workstream TTY, feat-488 GSD_HOME)
- strengthen property/shape-only tests (research-provider/store classification +
collision) and unconditional plugin.json schema validation (issue-766)
Verified: all 28 files run together 1220 pass / 0 fail / 1 skip.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore: add no-tautological-assert lint rule, error in test suite
New custom ESLint rule (eslint-rules/no-tautological-assert.cjs) bans asserts
that can never fail: assert(true)/assert.ok(<always-truthy literal>),
'cond || true' inside an assert, and equality asserts comparing two identical
literals. Wired as error on tests/**; full sweep confirmed zero existing
violations so the suite stays green. Prevents the placeholder-assert regressions
the audit redesigns just removed. RuleTester coverage added (6 valid, 8 invalid).
Note: no-only-tests was already enforced via eslint-plugin-no-only-tests, so no
duplicate rule was added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore: gate new allow-test-rule exemptions to require an issue ref
ADR-456 requires any allow-test-rule exemption added after the ADR to carry a
tracking issue number, but nothing enforced it. New ratchet gate
(scripts/lint-allow-test-rule-refs.cjs, wired into lint:ci) fails when a NEW
allow-test-rule comment lacks a #NNN/URL reference; the 323 existing untracked
exemptions are grandfathered in an allowlist that ratchets down as they gain
refs. Red-green verified (novel untracked offender fails; compliant passes).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* docs: add ADR test-audit evidence report (#1192)
Full risk-first qa-test-architect audit of the ADR portfolio (37 ADRs + 4
platform lenses, adversarial verification of retire verdicts) that drove the
P0 discovery fix, ADR-218 coverage, 5 retires, 27 redesigns, and the two new
lint gates. Filed as point-in-time evidence under docs/issueevidence/, named
for tracking issue #1192.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: replace pre-existing raw NUL byte with escape in feat-3594 fixture
feat-3594's null-byte parser fixture contained a literal NUL byte (pre-existing
on next at b10e5681 — confirmed: base blob has 1 NUL, this fix has 0), which
made git treat the file as binary and would break grep/editors. Switch to the
\x00 escape; the runtime string value (a real NUL in the parser input) is
unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: address adversarial-review findings
Codex adversarial pass over the branch:
- capability-registry drift test no longer mutates the committed generated
capability-registry.cjs in place (concurrency hazard) — uses in-memory
checkPipeline comparison instead.
- allow-test-rule ratchet now detects exemptions in ALL comment forms (block
/* */ too, matching no-source-grep) so a block comment can't bypass it;
one newly-surfaced pre-existing offender grandfathered (323->324).
- install.test Kilo case asserts on what install(false,'kilo') actually writes
rather than manually calling configureKiloPermissions (masked the call site).
- issue-766 drops the undeclared transitive ajv dep for explicit structural
assertions from the schema fixture.
- adr-218 test notes the hotfix leading-zero gap is tracked in #1186.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix: address code-review findings (subdir discovery, rule + test gaps)
xhigh code review surfaced 15 confirmed issues, all fixed:
- run-tests.cjs --files now resolves subdir tests by bare basename + handles
Windows backslash paths (ambiguous basenames error clearly).
- affected-tests-lib.cjs listTestFiles made recursive — the targeted CI lane was
silently dropping changed subdir tests (same false-green class the audit fixed).
- no-tautological-assert now catches 'true || cond' and empty []/{} equality.
- verify-test-quality: restore provenance-classification coverage, tighten the
writeFile circular-detection check, guard the module-level file read.
- sh-hook-paths: cover the global-install .sh delegation branch (#2045 guard).
- active-workstream null-guard runs deterministically (no longer skipped on TTY).
- adr-218 structural guards tightened (major/minor leading-zero; needs: membership).
- repo-layout AGENTS.md guard no longer false-alarms on equivalent refactors.
- cross-ai ordering guard fails red when the step is missing.
- issue-766 parses required fields from the schema fixture (auto-enforced).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: stub USERPROFILE alongside HOME in feat-488 (Windows parity)
The feat-488 redesign stubbed process.env.HOME but not USERPROFILE; os.homedir()
resolves from USERPROFILE on Windows, so the home stub was not hermetic there —
caught by windows-test-parity-guard (stubsHomeNoUserProfile). Save/set/restore
USERPROFILE symmetrically with HOME (delete-if-originally-undefined).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore: reconcile allow-test-rule allowlist after rebase onto next
Rebasing onto current next pulled in merged PR #1170, which added
inventory-headings-countfree.test.cjs (a baseline allow-test-rule exemption) and
deleted inventory-counts.test.cjs. Grandfather the former and prune the latter so
the ratchet matches the merged tree. No new debt from this PR.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* fix(#1012): invoke fallow with its real CLI and wire the report normalizer
The /gsd-code-review structural pre-pass invoked fallow with flags no published
fallow version accepts (--json, --profile, --stdin-files), so it failed on every
run and degraded silently per REQ-FALLOW-02 — the feature never delivered on any
fallow version. Three compounding defects:
1. Invalid flags. Real fallow audit uses --format json (not --json), -q/--quiet,
--changed-since/--base for changed-files scoping (no file-list input), and
--max-crap for thresholds. There is no --profile or --stdin-files.
2. Exit-code handling. fallow audit exits 1 when it FINDS issues (verdict=fail),
0 when clean. The pre-pass treated any non-zero exit as a crash and discarded
the output — i.e. it threw away exactly the findings it exists to surface.
Success is now decided by whether a valid fallow JSON report was produced,
not by the exit code.
3. Schema mismatch. normalizeFallowReport parsed a fictional top-level schema
(unusedExports/duplicates/circularDependencies) fallow never shipped, and was
dead code (the workflow embedded raw JSON; its tests asserted the fictional
schema, one even calling a non-existent runFallowAudit and passing vacuously).
Fixes: align the invocation to fallow's documented agent-facing pattern; map the
profile preset (minimal/standard/strict) to --max-crap (50/30/15); scope phase
runs via --changed-since with a repo-scope fallback; rewrite the normalizer to
fallow's real schema (dead_code.unused_exports/unused_files/circular_dependencies
+ duplication.clone_groups) and wire it into the workflow so the reviewer
receives normalized findings; replace the fictional-schema fixtures and tests
with real-schema ones and delete the vacuous runFallowAudit test.
Closes#1012
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore(#1012): backfill changeset PR number to 1044
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Renames (git mv) with all references updated (ci-test-scope RULES,
windows-parity allowlist, test-file-count allowlist, docs in 6 locales):
- 5 scanner tests -> *.security.test.cjs — the 'Run security tests' CI step
ran zero files since the suite taxonomy landed; it is now honest.
- graphify-auto-update -> *.slow.test.cjs (36s, slowest file in the suite;
e2e gsd-tools spawns) — runs on full-matrix lanes and push to next.
- installer-migration-install-integration -> *.integration.test.cjs
(13s; an integration test by its own name).
Coverage gate measured after retags: 88.55% lines (gate 70%).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Editing the body of an already-released installer migration drifts its computed
checksum (it hashes plan.toString()). The integrity guard then hard-aborted
every prior install on upgrade with "applied migration checksum changed" — a
100% reproducible blocker (v1.3.0, all platforms).
Already-applied migrations are filtered out of `pending` and never re-run, so
a drifted checksum is functionally inert. ADR-0008 anticipates checksum-mismatch
state as something the install-state layer must handle gracefully (plan -> apply
-> recover/report), not abort on.
This supersedes the published-checksum allowlist merged in #674 (per-release
maintenance debt — every historical checksum hand-pinned, still throws for any
unregistered value) with a general, self-healing recovery:
- Replace the throwing guard with non-fatal `collectAppliedChecksumDrift`,
surfaced on `plan.checksumDrift`.
- Reconcile drifted stored checksums durably on the next state write
(`reconcileDriftedChecksums`), idempotently (no perpetual writes).
- Relocate the "shipped migration bodies are immutable" rule to a CI baseline
test that locks every shipped migration's checksum and fails on body drift —
where #615 should have been caught, instead of blocking users.
Removes #674's legacyChecksums field, per-migration checksum pins,
published-checksums.json fixture, and compat test.
Fixes#670
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
