[[agents]] sequence format (introduced in #2645) is rejected by
codex-cli 0.124.0 with "invalid type: sequence, expected struct AgentsToml".
Revert to [agents.<name>] struct format which is correct for 0.120.0+.
stripCodexGsdAgentSections already handles both formats for self-healing
configs written by previous GSD versions using [[agents]].
Closes#2727
Fixes#2698 — The two separate LF/CRLF .replace() calls in the Codex hooks
migration could not handle mixed line endings (e.g. header in LF, body in
CRLF), leaving stale gsd-update-check blocks after reinstall. Consolidated to
a single \r?\n-aware regex with gm flags that handles LF, CRLF, and mixed
content in one pass.
Fixes#2678 — installSdkIfNeeded() called process.exit(1) unconditionally when
sdk/dist/cli.js was missing, even during --local installs where users cannot
write to global node_modules. Added isLocal option: when true, prints a warning
and returns instead of exiting.
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(#2645): emit [[agents]] array-of-tables in Codex config.toml
Codex ≥0.116 rejects `[agents.<name>]` map tables with `invalid type:
map, expected a sequence`. Switch generateCodexConfigBlock to emit
`[[agents]]` array-of-tables with an explicit `name` field per entry.
Strip + merge paths now self-heal on reinstall — both the legacy
`[agents.gsd-*]` map shape (pre-#2645 configs) and the new
`[[agents]]` with `name = "gsd-*"` shape are recognized and replaced,
while user-authored `[[agents]]` entries are preserved.
Fixes#2645
* fix(#2645): use TOML-aware parser to strip managed [[agents]] sections
CodeRabbit flagged that the prior regex-based stripper for [[agents]]
array-of-tables only matched headers at column 0 and stopped at any line
beginning with `[`. An indented [[agents]] header would not terminate the
preceding match, so a managed `gsd-*` block could absorb a following
user-authored agent and silently delete it.
Replace the ad-hoc regex with the existing TOML-aware section parser
(getTomlTableSections + removeContentRanges) so section boundaries are
authoritative regardless of indentation. Same logic applies to legacy
[agents.gsd-*] map sections.
Add a comprehensive mixed-shape test covering multiple GSD entries (both
legacy map and new array-of-tables, double- and single-quoted names)
interleaved with multiple user-authored agents in both shapes — verifies
all GSD entries are stripped and every user entry is preserved.
installCodexConfig() applied a narrow path-only regex pass before
generateCodexAgentToml(), skipping the convertClaudeToCodexMarkdown() +
neutralizeAgentReferences(..., 'AGENTS.md') pipeline used on the .md emit
path. Result: emitted Codex agent TOMLs carried stale Claude-specific
references (CLAUDE.md, .claude/skills/, .claude/commands/, .claude/agents/,
.claudeignore, bare "Claude" agent-name mentions).
Route the TOML path through convertClaudeToCodexMarkdown and extend that
pipeline to cover bare .claude/<subdir>/ references and .claudeignore
(both previously unhandled on the .md path too). The $HOME/.claude/
get-shit-done prefix substitution still runs first so the absolute Codex
install path is preserved before the generic .claude → .codex rewrite.
Regression test: tests/issue-2639-codex-toml-neutralization.test.cjs —
drives installCodexConfig against a fixture containing every flagged
marker and asserts the emitted TOML contains zero CLAUDE.md / .claude/
/ .claudeignore occurrences and that Claude Code / Claude Opus product
names survive.
Fixes#2639
Flat-skills installs write SKILL.md files under gsd-<cmd>/ dirs, but
Claude Code resolves skills by their frontmatter `name:`, not directory
name. PR #2595 normalized every `/gsd-<cmd>` to `/gsd:<cmd>` across
workflows — including inside `Skill(skill="...")` args — but the
installer still emitted `name: gsd-<cmd>`, so every Skill() call on a
flat-skills install resolved to nothing.
Fix: emit `name: gsd:<cmd>` (colon form) in
`convertClaudeCommandToClaudeSkill`. Keep the hyphen-form directory
name for Windows path safety.
Codex stays on hyphen form: its adapter invokes skills as `$gsd-<cmd>`
(shell-var syntax) and a colon would terminate the variable name.
`convertClaudeCommandToCodexSkill` uses `yamlQuote(skillName)` directly
and is untouched.
- Extract `skillFrontmatterName(dirName)` helper (exported for tests).
- Update claude-skills-migration and qwen-skills-migration assertions
that encoded the old hyphen emission.
- Add `tests/bug-2643-skill-frontmatter-name.test.cjs` asserting every
`Skill(skill="gsd:<cmd>")` reference in workflows resolves to an
emitted frontmatter name.
Full suite: 5452/5452 passing.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Root cause shared with #2647: a broken 1.38.3 tarball shipped without
sdk/dist/. The pre-#2441-decouple installer reacted by running
spawnSync('npm.cmd', ['install'], { cwd: sdkDir }) inside the npx cache
on Windows, where the cache is read-only, producing the misleading
"Failed to npm install in sdk/" error.
Defensive changes here (user-facing behavior only; packaging fix lives
in the sibling PR for #2647):
- Classify the install context (classifySdkInstall): detect npx cache
paths, node_modules-based installs, and dev clones via path heuristics
plus a side-effect-free write probe. Exported for test.
- Rewrite the dist-missing error to branch on context:
tarball + npxCache -> "don't touch npx cache; npm i -g ...@latest"
tarball (other) -> upgrade path + clone-build escape hatch
dev-clone -> keep existing cd sdk && npm install && npm run build
- Preserve the invariant that the installer never shells out to
npm install itself — users always drive that.
- Add tests/bug-2649-sdk-fail-fast.test.cjs covering the classifier and
both failure messages, with spawnSync/execSync interceptors that
assert no nested npm install is attempted.
Cross-ref: #2647 (packaging).
Fixes#2649
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Codex and OpenCode install paths read `model_overrides` only from
`~/.gsd/defaults.json` (global). A per-project override set in
`.planning/config.json` — the reporter's exact setup for
`gsd-codebase-mapper` — was silently dropped, so the child agent inherited
the runtime's default model regardless of `model_overrides`.
Neither runtime has an inline `model` parameter on its spawn API
(Codex `spawn_agent(agent_type, message)`, OpenCode `task(description,
prompt, subagent_type, task_id, command)`), so the per-agent model must
reach the child via the static config GSD writes at install time. That
config was being populated from the wrong source.
Fix: add `readGsdEffectiveModelOverrides(targetDir)` which merges
`~/.gsd/defaults.json` with per-project `.planning/config.json`, with
per-project keys winning on conflict. Both install sites now call it and
walk up from the install root to locate `.planning/` — matching the
precedence `readGsdRuntimeProfileResolver` already uses for #2517.
Also update the Codex Task()->spawn_agent mapping block so it no longer
says "omit" without context: it now documents that per-agent overrides
are embedded in the agent TOML and notes the restriction that Codex
only permits `spawn_agent` when the user explicitly requested sub-agents
(do the work inline otherwise).
Regression tests (`tests/bug-2256-model-overrides-transport.test.cjs`)
cover: global-only, project-only, project-wins-on-conflict, walking up
from a nested `targetDir`, Codex TOML `model =` emission, and OpenCode
frontmatter `model:` emission.
Closes#2256
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(#2620): detect HOME-relative PATH entries before suggesting absolute export
When the installer reported `gsd-sdk` not on PATH and suggested
appending an absolute `export PATH="/home/user/.npm-global/bin:$PATH"`
line to the user's rc file, a user who had the equivalent
`export PATH="$HOME/.npm-global/bin:$PATH"` already in their shell
profile would get a duplicate entry — the installer only compared the
absolute form.
Add `homePathCoveredByRc(globalBin, homeDir, rcFileNames?)` to
`bin/install.js` and export it for test-mode callers. The helper scans
`~/.zshrc`, `~/.bashrc`, `~/.bash_profile`, `~/.profile`, grepping each
file for `export PATH=` / bare `PATH=` lines and substituting the
common HOME forms (\$HOME, \${HOME}, leading ~/) with the real home
directory before comparing each resolved PATH segment against
globalBin. Trailing slashes are normalised so `.npm-global/bin/`
matches `.npm-global/bin`. Missing / unreadable / malformed rc files
are swallowed — the caller falls back to the existing absolute
suggestion.
Tests cover $HOME, \${HOME}, and ~/ forms, absolute match,
trailing-slash match, commented-out lines, missing rc files, and
unreadable rc files (directory where a file is expected).
Closes#2620
* fix(#2620): skip relative PATH segments in homePathCoveredByRc
CodeRabbit flagged that the helper unconditionally resolved every
non-$-containing segment against homeAbs via path.resolve(homeAbs, …),
which silently turns a bare relative segment like `bin` or
`node_modules/.bin` into `$HOME/bin` / `$HOME/node_modules/.bin`. That
is wrong: bare PATH segments depend on the shell's cwd at lookup time,
not on $HOME — so the helper was returning true for rc files that do
not actually cover globalBin.
Guard the compare with path.isAbsolute(expanded) after HOME expansion.
Only segments that are absolute on their own (or that became absolute
via $HOME / \${HOME} / ~ substitution) are compared against targetAbs.
Relative segments are skipped.
Add two regression tests covering a bare `bin` segment and a nested
`node_modules/.bin` segment; both previously returned true when home
happened to contain a matching subdirectory and now correctly return
false.
Closes#2620 (CodeRabbit follow-up)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(#2620): wire homePathCoveredByRc into installer suggestion path
CodeRabbit flagged that homePathCoveredByRc was added in the previous
commit but never called from the installer, so the user-facing PATH
warning stayed unchanged — users with `export PATH="$HOME/.npm-global/bin:$PATH"`
in their rc would still get a duplicate absolute-path suggestion.
Add `maybeSuggestPathExport(globalBin, homeDir)` that:
- skips silently when globalBin is already on process.env.PATH;
- prints a "try reopening your shell" diagnostic when homePathCoveredByRc
returns true (the directory IS on PATH via an rc entry — just not in
the current shell);
- otherwise falls through to the absolute-path
`echo 'export PATH="…:$PATH"' >> ~/.zshrc` suggestion.
Call it from installSdkIfNeeded after the sdk/dist check succeeds,
resolving globalBin via `npm prefix -g` (plus `/bin` on POSIX). Swallow
any exec failure so the installer keeps working when npm is weird.
Export maybeSuggestPathExport for tests. Add three new regression tests
(installer-flow coverage per CodeRabbit nitpick):
- rc covers globalBin via $HOME form → no absolute suggestion emitted
- rc covers only an unrelated directory → absolute suggestion emitted
- globalBin already on process.env.PATH → no output at all
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(sdk): decouple SDK from build-from-source install path, close#2441 and #2453
Ship sdk/dist prebuilt in the tarball and replace the npm-install-g
sub-install with a parent-package bin shim (bin/gsd-sdk.js). npm chmods
bin entries from a packed tarball correctly, eliminating the mode-644
failure (#2453) and the full class of NPM_CONFIG_PREFIX/ignore-scripts/
corepack/air-gapped failure modes that caused #2439 and #2441.
Changes:
- sdk/package.json: prepublishOnly runs `rm -rf dist && tsc && chmod +x
dist/cli.js` (stale-build guard + execute-bit fix at publish time)
- package.json: add "gsd-sdk": "bin/gsd-sdk.js" bin entry; add sdk/dist
to files so the prebuilt CLI ships in the tarball
- bin/gsd-sdk.js: new back-compat shim — resolves sdk/dist/cli.js relative
to the package root and delegates via `node`, so all existing PATH call
sites (slash commands, agents, hooks) continue to work unchanged (S1 shim)
- bin/install.js: replace installSdkIfNeeded() build-from-source + global-
install dance with a dist-verify + chmod-in-place guard; delete
resolveGsdSdk(), detectShellRc(), emitSdkFatal() helpers now unused
- .github/workflows/install-smoke.yml: add smoke-unpacked job that strips
execute bit from sdk/dist/cli.js before install to reproduce the exact
#2453 failure mode
- tests/bug-2441-sdk-decouple.test.cjs: new regression tests asserting all
invariants (no npm install -g from sdk/, shim exists, sdk/dist in files,
prepublishOnly has rm -rf + chmod)
- tests/bugs-1656-1657.test.cjs: update stale assertions that required
build-from-source behavior (now asserts new prebuilt-dist invariants)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* chore(release): bump to 1.38.2, wire release.yml to build SDK dist
- Bump version 1.38.1 -> 1.38.2 for the #2441/#2453 fix shipped in 0f6903d.
- Add `build:sdk` script (`cd sdk && npm ci && npm run build`).
- `prepublishOnly` now runs hooks + SDK builds as a safety net.
- release.yml (rc + finalize): build SDK dist before `npm publish` so the
published tarball always ships fresh `sdk/dist/` (kept gitignored).
- CHANGELOG: document 1.38.2 entry and `--sdk` flag semantics change.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* ci: build SDK dist before tests and smoke jobs
sdk/dist/ is gitignored (built fresh at publish time via release.yml),
but both the test suite and install-smoke jobs run `bin/install.js`
or `npm pack` against the checked-out tree where dist doesn't exist yet.
- test.yml: `npm run build:sdk` before `npm run test:coverage`, so tests
that spawn `bin/install.js` don't hit `installSdkIfNeeded()`'s fatal
missing-dist check.
- install-smoke.yml (both smoke and smoke-unpacked): build SDK before
pack/chmod so the published tarball contains dist and the unpacked
install has a file to strip exec-bit from.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(sdk): lift SDK runtime deps to parent so tarball install can resolve them
The SDK's runtime deps (ws, @anthropic-ai/claude-agent-sdk) live in
sdk/package.json, but sdk/node_modules is NOT shipped in the parent
tarball — only sdk/dist, sdk/src, sdk/prompts, and sdk/package.json are.
When a user runs `npm install -g get-shit-done-cc`, npm installs the
parent's node_modules but never runs `npm install` inside the nested
sdk/ directory.
Result: `node sdk/dist/cli.js` fails with ERR_MODULE_NOT_FOUND for 'ws'.
The smoke tarball job caught this; the unpacked variant masked it
because `npm install -g <dir>` copies the entire workspace including
sdk/node_modules (left over from `npm run build:sdk`).
Fix: declare the same deps in the parent package.json so they land in
<pkg>/node_modules, which Node's resolution walks up to from
<pkg>/sdk/dist/cli.js. Keep them declared in sdk/package.json too so
the SDK remains a self-contained package for standalone dev.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(lockfile): regenerate package-lock.json cleanly
The previous `npm install` run left the lockfile internally inconsistent
(resolved esbuild@0.27.7 referenced but not fully written), causing
`npm ci` to fail in CI with "Missing from lock file" errors.
Clean regen via rm + npm install fixes all three failed jobs
(test, smoke, smoke-unpacked), which were all hitting the same
`npm ci` sync check.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(deps): remove unused esbuild + vitest from root devDependencies
Both were declared but never imported anywhere in the root package
(confirmed via grep of bin/, scripts/, tests/). They lived in sdk/
already, which is the only place they're actually used.
The transitive tree they pulled in (vitest → vite → esbuild 0.28 →
@esbuild/openharmony-arm64) was the root of the CI npm ci failures:
the openharmony platform package's `optional: true` flag was not being
applied correctly by npm 10 on Linux runners, causing EBADPLATFORM.
After removal: 800+ transitive packages → 155. Lockfile regenerated
cleanly. All 4170 tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(sdk): pretest:coverage builds sdk; tighten shim test assertions
Add "pretest:coverage": "npm run build:sdk" so npm run test:coverage
works in clean checkouts where sdk/dist/ hasn't been built yet.
Tighten the two loose shim assertions in bug-2441-sdk-decouple.test.cjs:
- forwards-to test now asserts path.resolve() is called with the
'sdk','dist','cli.js' path segments, not just substring presence
- node-invocation test now asserts spawnSync(process.execPath, [...])
pattern, ruling out matches in comments or the shebang line
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: address PR review — pretest:coverage + tighten shim tests
Review feedback from trek-e on PR 2457:
1. pretest:coverage + pretest hooks now run `npm run build:sdk` so
`npm run test[:coverage]` in a clean checkout produces the required
sdk/dist/ artifacts before running the installer-dependent tests.
CI already does this explicitly; local contributors benefit.
2. Shim tests in bug-2441-sdk-decouple.test.cjs tightened from loose
substring matches (which would pass on comments/shebangs alone) to
regex assertions on the actual path.resolve call, spawnSync with
process.execPath, process.argv.slice(2), and process.exit pattern.
These now provide real regression protection for #2453-class bugs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix: correct CHANGELOG entry and add [1.38.2] reference link
Two issues in the 1.38.2 CHANGELOG entry:
- installSdkIfNeeded() was described as deleted but it still exists in
bin/install.js (repurposed to verify sdk/dist/cli.js and fix execute bit).
Corrected the description to say 'repurposes' rather than 'deletes'.
- The reference-link block at the bottom of the file was missing a [1.38.2]
compare URL and [Unreleased] still pointed to v1.37.1...HEAD. Added the
[1.38.2] link and updated [Unreleased] to compare/v1.38.2...HEAD.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(sdk): double-cast WorkflowConfig to Record for strict tsc build
TypeScript error on main (introduced in #2611) blocks `npm run build`
in sdk/, which now runs as part of this PR's tarball build path. Apply
the double-cast via `unknown` as the compiler suggests.
Same fix as #2622; can be dropped if that lands first.
* test: remove bug-2598 test obsoleted by SDK decoupling
The bug-2598 test guards the Windows CVE-2024-27980 fix in the old
build-from-source path (npm spawnSync with shell:true + formatSpawnFailure
diagnostics). This PR removes that entire code path — installSdkIfNeeded
no longer spawns npm, it just verifies the prebuilt sdk/dist/cli.js
shipped in the tarball.
The test asserts `installSdkIfNeeded.toString()` contains a
formatSpawnFailure helper. After decoupling, no such helper exists
(nothing to format — there's no spawn). Keeping the test would assert
invariants of the rejected architecture.
The original #2598 defect (silent failure of npm spawn on Windows) is
structurally impossible in the shim path: bin/gsd-sdk.js invokes
`node sdk/dist/cli.js` directly via child_process.spawn with an
explicit argv array. No .cmd wrapper, no shell delegation.
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Tom Boucher <trekkie@nomorestars.com>
* feat: make model profiles runtime-aware for Codex/non-Claude runtimes (closes#2517)
Adds an optional top-level `runtime` config key plus a
`model_profile_overrides[runtime][tier]` map. When `runtime` is set,
profile tiers (opus/sonnet/haiku) resolve to runtime-native model IDs
(and reasoning_effort where supported) instead of bare Claude aliases.
Codex defaults from the spec:
opus -> gpt-5.4 reasoning_effort: xhigh
sonnet -> gpt-5.3-codex reasoning_effort: medium
haiku -> gpt-5.4-mini reasoning_effort: medium
Claude defaults mirror MODEL_ALIAS_MAP. Unknown runtimes fall back to
the Claude-alias safe default rather than emit IDs the runtime cannot
accept. reasoning_effort is only emitted into Codex install paths;
never returned from resolveModelInternal and never written to Claude
agent frontmatter.
Backwards compatible: any user without `runtime` set sees identical
behavior — the new branch is gated on `config.runtime != null`.
Precedence (highest to lowest):
1. per-agent model_overrides
2. runtime-aware tier resolution (when `runtime` is set)
3. resolve_model_ids: "omit"
4. Claude-native default
5. inherit (literal passthrough)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(#2517): address adversarial review of #2609 (findings 1-16)
Addresses all 16 findings from the adversarial review of PR #2609.
Each finding is enumerated below with its resolution.
CRITICAL
- F1: readGsdRuntimeProfileResolver(targetDir) now probes per-project
.planning/config.json AND ~/.gsd/defaults.json with per-project winning,
so the PR's headline claim ("set runtime in project config and Codex
TOML emit picks it up") actually holds end-to-end.
- F2: resolveTierEntry field-merges user overrides with built-in defaults.
The CONFIGURATION.md string-shorthand example
`{ codex: { opus: "gpt-5-pro" } }`
now keeps reasoning_effort from the built-in entry. Partial-object
overrides like `{ opus: { reasoning_effort: 'low' } }` keep the
built-in model. Both paths regression-tested.
MAJOR
- F3: resolveReasoningEffortInternal gates strictly on the
RUNTIMES_WITH_REASONING_EFFORT allowlist regardless of override
presence. Override + unknown-runtime no longer leaks reasoning_effort.
- F4: runtime:"claude" is now a no-op for resolution (it is the implicit
default). It no longer hijacks resolve_model_ids:"omit". Existing
tests for `runtime:"claude"` returning Claude IDs were rewritten to
reflect the no-op semantics; new test asserts the omit case returns "".
- F5: _readGsdConfigFile in install.js writes a stderr warning on JSON
parse failure instead of silently returning null. Read failure and
parse failure are warned separately. Library require is hoisted to top
of install.js so it is not co-mingled with config-read failure modes.
- F6: install.js requires for core.cjs / model-profiles.cjs are hoisted
to the top of the file with __dirname-based absolute paths so global
npm install works regardless of cwd. Test asserts both lib paths exist
relative to install.js __dirname.
- F7: docs/CONFIGURATION.md `runtime` row no longer lists `opencode` as
a valid runtime — install-path emission for non-Codex runtimes is
explicitly out of scope per #2517 / #2612, and the doc now points at
#2612 for the follow-on work. resolveModelInternal still accepts any
runtime string (back-compat) and falls back safely for unknown values.
- F8: Tests now isolate HOME (and GSD_HOME) to a per-test tmpdir so the
developer's real ~/.gsd/defaults.json cannot bleed into assertions.
Same pattern CodeRabbit caught on PRs #2603 / #2604.
- F9: `runtime` and `model_profile_overrides` documented as flat-only
in core.cjs comments — not routed through `get()` because they are
top-level keys per docs/CONFIGURATION.md and introducing nested
resolution for two new keys was not worth the edge-case surface.
- F10/F13: loadConfig now invokes _warnUnknownProfileOverrides on the
raw parsed config so direct .planning/config.json edits surface
unknown runtime values (e.g. typo `runtime: "codx"`) and unknown
tier values (e.g. `model_profile_overrides.codex.banana`) at read
time. Warnings only — preserves back-compat for runtimes added
later. Per-process warning cache prevents log spam across repeated
loadConfig calls.
MINOR / NIT
- F11: Removed dead `tier || 'sonnet'` defensive shortcut. The local
is now `const alias = tier;` with a comment explaining why `tier`
is guaranteed truthy at that point (every MODEL_PROFILES entry
defines `balanced`, the fallback profile).
- F12: Extracted resolveTierEntry() in core.cjs as the single source
of truth for runtime-aware tier resolution. core.cjs and bin/install.js
both consume it — no duplicated lookup logic between the two files.
- F14: Added regression tests for findings #1, #2, #3, #4, #6, #10, #13
in tests/issue-2517-runtime-aware-profiles.test.cjs. Each must-fix
path has a corresponding test that fails against the pre-fix code
and passes against the post-fix code.
- F15: docs/CONFIGURATION.md `model_profile` row cross-references
#1713 / #1806 next to the `adaptive` enum value.
- F16: RUNTIME_PROFILE_MAP remains in core.cjs as the single source of
truth; install.js imports it through the exported resolveTierEntry
helper rather than carrying its own copy. Doc files (CONFIGURATION.md,
USER-GUIDE.md, settings.md) intentionally still embed the IDs as text
— code comment in core.cjs flags that those doc files must be updated
whenever the constant changes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(#2598): pass shell: true to npm spawnSync on Windows
Since Node's CVE-2024-27980 fix (>= 18.20.2 / >= 20.12.2 / >= 21.7.3),
spawnSync refuses to launch .cmd/.bat files on Windows without
`shell: true`. installSdkIfNeeded picks npmCmd='npm.cmd' on win32 and
then calls spawnSync five times — every one returns
{ status: null, error: EINVAL } before npm ever runs. The installer
checks `status !== 0`, trips the failure path, and emits a bare
"Failed to `npm install` in sdk/." with zero diagnostic output because
`stdio: 'inherit'` never had a child to stream.
Every fresh install on Windows has failed at the SDK build step on any
supported Node version for the life of the post-CVE bin/install.js.
Introduce a local `spawnNpm(args, opts)` helper inside
installSdkIfNeeded that injects `shell: process.platform === 'win32'`
when the caller doesn't override it. Route all five npm invocations
through it: `npm install`, `npm run build`, `npm install -g .`, and
both `npm config get prefix` calls.
Adds a static regression test that parses installSdkIfNeeded and
asserts no bare `spawnSync(npmCmd, ...)` remains, a shell-aware
wrapper exists, and at least five invocations go through it.
Closes#2598
* fix(#2598): surface spawnSync diagnostics in SDK install fatal paths
Thread result.error / result.signal / result.status into emitSdkFatal for
the three npm failure branches (install, run build, install -g .) via a
formatSpawnFailure helper. The root cause of #2598 went silent precisely
because `{ status: null, error: EINVAL }` was reduced to a generic
"Failed to `npm install` in sdk/." with no diagnostic — stdio: 'inherit'
had no child process to stream and result.error was swallowed. Any future
regression in the same area (EINVAL, ENOENT, signal termination) now
prints its real cause in the red fatal banner.
Also strengthen the regression test so it cannot pass with only four
real npm call sites: the previous `spawnSync(npmCmd, ..., shell)` regex
double-counted the spawnNpm helper's own body when a helper existed.
Separate arrow-form vs function-form helper detection and exclude the
wrapper body from explicitShellNpm so the `>= 5` assertion reflects real
invocations only. Add a new test that asserts all three fatal branches
now reference formatSpawnFailure / result.error / signal / status.
Addresses CodeRabbit review comments on PR #2600:
- r3126987409 (bin/install.js): surface underlying spawnSync failure
- r3126987419 (test): explicitShellNpm overcounts by one via helper def
The Copilot content converter only replaced `~/.claude/` and
`$HOME/.claude/` when followed by a literal `/`. Bare references
(e.g. `configDir = ~/.claude` at end of line) slipped through and
triggered the post-install "Found N unreplaced .claude path reference(s)"
warning, since the leak scanner uses `(?:~|$HOME)/\.claude\b`.
Switched both replacements to a `(\/|\b)` capture group so trailing-slash
and bare forms are handled in a single pass — matching the pattern
already used by Antigravity, OpenCode, Kilo, and Codex converters.
Closes#2545
\$CLAUDE_PROJECT_DIR is Claude Code-specific. Gemini CLI doesn't set it, and on
Windows its path-join logic doubled the value producing unresolvable paths like
D:\Projects\GSD\'D:\Projects\GSD'. Gemini runs project hooks with project root
as cwd, so bare relative paths (e.g. node .gemini/hooks/gsd-check-update.js)
are cross-platform and correct. Claude Code and others still use the env var.
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Use process.platform !== 'win32' guard in catch instead of a comment, and add
regression test for bug #2525 (gsd-sdk bin symlink points at non-executable file).
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(tests): update 5 source-text tests to read config-schema.cjs
VALID_CONFIG_KEYS moved from config.cjs to config-schema.cjs in the
drift-prevention companion PR. Tests that read config.cjs source text
and checked for key literal includes() now point to the correct file.
Closes#2480
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(cli): add /gsd-sync-skills for cross-runtime managed skill sync (#2380)
Adds /gsd-sync-skills command so multi-runtime users can keep gsd-* skill
directories aligned across runtime roots after updating one runtime with gsd-update.
Changes:
- bin/install.js: add --skills-root <runtime> flag that prints the skills root
path for any supported runtime, reusing the existing getGlobalDir() table.
Banner is suppressed when --skills-root is used (machine-readable output).
- commands/gsd/sync-skills.md: slash command definition
- get-shit-done/workflows/sync-skills.md: full workflow spec covering argument
parsing, path resolution via --skills-root, diff computation (CREATE/UPDATE/
REMOVE/SKIP), dry-run report (default), apply execution, idempotency guarantee,
and safety rules (only gsd-* touched, dry-run performs no writes).
Safety rules: only gsd-* directories are ever created/updated/removed; non-GSD
skills in destination roots are never touched; --dry-run is the default.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
#2388 (plan-phase silently renames feature branch): add explicit Git
Branch Invariant section to plan-phase.md prohibiting branch
creation/rename/switch during planning; phase slug changes are
plan-level only and must not affect the git branch.
#2431 (worktree teardown silently swallows errors): replace
`git worktree remove --force 2>/dev/null || true` with a lock-aware
block in quick.md and execute-phase.md that detects locked worktrees,
attempts unlock+retry, and surfaces a user-visible recovery message
when removal still fails.
#2396 (hardcoded test commands bypass Makefile): add a three-tier
test command resolver (project config → Makefile/Justfile → language
sniff) in execute-phase.md, verify-phase.md, and audit-fix.md.
Makefile with a `test:` target now takes priority over npm/cargo/go.
#2376 (OpenCode @$HOME not mapped on Windows): add platform guard in
bin/install.js so OpenCode on win32 uses the absolute path instead of
`$HOME/...`, which OpenCode does not expand in @file references on
Windows.
Tests: 29 new assertions across 4 regression test files (all passing).
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
- #2418: convertClaudeToAntigravityContent now replaces bare ~/.claude and
$HOME/.claude (no trailing slash) for both global and local installs,
eliminating the "unreplaced .claude path reference" warnings in
gsd-debugger.md and update.md during Antigravity installs.
- #2399: plan-phase workflow gains step 13c that commits PLAN.md files
and STATE.md via gsd-sdk query commit when commit_docs is true.
Previously commit_docs:true was read but never acted on in plan-phase.
- #2419: new-project.md and new-milestone.md now parse agents_installed
and missing_agents from the init JSON and warn users clearly when GSD
agents are not installed, rather than silently failing with "agent type
not found" when trying to spawn gsd-project-researcher subagents.
- #2421: gsd-planner.md gains a "Grep gate hygiene" rule immediately after
the Nyquist Rule explaining the self-invalidating grep gate anti-pattern
and providing comment-stripping alternatives (grep -v, ast-grep).
Tests: 4 new test files (30 tests) all passing.
Closes#2418Closes#2399Closes#2419Closes#2421
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Bug #2453: After tsc builds sdk/dist/cli.js, npm install -g from a local
directory does not chmod the bin-script target (unlike tarball extraction).
The file lands at mode 644, the gsd-sdk symlink points at a non-executable
file, and command -v gsd-sdk fails on every first install. Fix: explicitly
chmodSync(cliPath, 0o755) immediately after npm install -g completes,
mirroring the pattern used for hook files throughout the installer.
Bug #2451: gsd-context-monitor warning messages over-reported usage by ~13
percentage points vs CC native /context. Root cause: gsd-statusline.js
wrote a buffer-normalized used_pct (accounting for the 16.5% autocompact
reserve) to the bridge file, inflating values. The bridge used_pct is now
raw (Math.round(100 - remaining_percentage)), consistent with what CC's
native /context command reports. The statusline progress bar continues to
display the normalized value; only the bridge value changes. Updated the
existing #2219 tests to check the normalized display via hook stdout rather
than bridge.used_pct, and added a new assertion that bridge.used_pct is raw.
Closes#2453Closes#2451
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
## Why
#2386 added `installSdkIfNeeded()` to build @gsd-build/sdk from bundled
source and `npm install -g .`, because the npm-published @gsd-build/sdk
is intentionally frozen and version-mismatched with get-shit-done-cc.
But every failure path in that function was warning-only — including
the final `which gsd-sdk` verification. When npm's global bin is off a
user's PATH (common on macOS), the installer printed a yellow warning
then exited 0. Users saw "install complete" and then every `/gsd-*`
command crashed with `command not found: gsd-sdk` (the #2439 symptom).
No CI job executed the install path, so this class of regression could
ship undetected — existing "install" tests only read bin/install.js as
a string.
## What changed
**bin/install.js — installSdkIfNeeded() is now transactional**
- All build/install failures exit non-zero (not just warn).
- Post-install `which gsd-sdk` check is fatal: if the binary landed
globally but is off PATH, we exit 1 with a red banner showing the
resolved npm bin dir, the user's shell, the target rc file, and the
exact `export PATH=…` line to add.
- Escape hatch: `GSD_ALLOW_OFF_PATH=1` downgrades off-PATH to exit 2
for users with intentionally restricted PATH who will wire up the
binary manually.
- Resolver uses POSIX `command -v` via `sh -c` (replaces `which`) so
behavior is consistent across sh/bash/zsh/fish.
- Factored `resolveGsdSdk()`, `detectShellRc()`, `emitSdkFatal()`.
**.github/workflows/install-smoke.yml (new)**
- Executes the real install path: `npm pack` → `npm install -g <tgz>`
→ run installer non-interactively → `command -v gsd-sdk` → run
`gsd-sdk --version`.
- PRs: path-filtered to installer-adjacent files, ubuntu + Node 22 only.
- main/release branches: full matrix (ubuntu+macos × Node 22+24).
- Reusable via workflow_call with `ref` input for release gating.
**.github/workflows/release.yml — pre-publish gate**
- New `install-smoke-rc` and `install-smoke-finalize` jobs invoke the
reusable workflow against the release branch. `rc` and `finalize`
now `needs: [validate-version, install-smoke-*]`, so a broken SDK
install blocks `npm publish`.
## Test plan
- Local full suite: 4154/4154 pass
- install-smoke.yml will self-validate on this PR (ubuntu+Node22 only)
Addresses root cause of #2439 (the per-command pre-flight in #2440 is
the complementary defensive layer).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
PR #2386 v1 installed the published @gsd-build/sdk from npm, which ships an
older version that lacks query handlers needed by current workflows. Every
GSD release would drift further from what the installer put on PATH.
This commit rewires installSdkIfNeeded() to build from the in-repo sdk/
source tree instead:
1. cd sdk && npm install (build-time deps incl. tsc)
2. npm run build (tsc → sdk/dist/)
3. npm install -g . (global install; gsd-sdk on PATH)
Each step is a hard gate — failures warn loudly and point users at the
manual equivalent command. No more silent drift between installed SDK and
the rest of the GSD system.
Root package.json `files` now ships sdk/src, sdk/prompts, sdk/package.json,
sdk/package-lock.json, and sdk/tsconfig.json so npm-registry installs also
carry the source tree needed to build gsd-sdk locally.
Also fixes a blocking tsc error in sdk/src/event-stream.ts:313 — the cast
to `Array<{ type: string; [key: string]: unknown }>` needed a double-cast
via `unknown` because BetaContentBlock's variants don't carry an index
signature. Runtime-neutral type-widening; sdk vitest suite unchanged
(1256 passing; the lone failure is a pre-existing integration test that
requires external API access).
Updates the #1657/#2385 regression test to assert the new build-from-source
path (path.resolve(__dirname, '..', 'sdk') + `npm run build` + `npm install
-g .`) plus a new assertion that root package.json files array ships sdk
source.
Refs #2385
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
`npm install -g` can succeed while the binary lands in a prefix that
isn't on the current shell's PATH (common with Homebrew, nvm, or an
unconfigured npm prefix). Re-probe via `which gsd-sdk` (or `where` on
Windows) after install; if it doesn't resolve, downgrade the success
message to a warning with a shell-restart hint.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Previously passing both silently had --no-sdk win. Exit non-zero with
a clear error to match how other exclusive flag pairs (--global/--local,
--config-dir/--local) are handled.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Every /gsd-* command shells out to `gsd-sdk query …`, but the SDK was
never installed by bin/install.js — the `--sdk` flag documented in
README was never implemented. Users upgrading to 1.36+ hit
"command not found: gsd-sdk" on every command.
- Implement SDK install in finishInstall's finalize path
- Default on; --no-sdk to skip; --sdk to force when already present
- Idempotent probe via `which gsd-sdk` before reinstalling
- Failures are warnings, not fatal — install hint printed
Closes#2385
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* feat: add /gsd-spec-phase — Socratic spec refinement with ambiguity scoring (#2213)
Introduces `/gsd-spec-phase <phase>` as an optional pre-step before discuss-phase.
Clarifies WHAT a phase delivers (requirements, boundaries, acceptance criteria) with
quantitative ambiguity scoring before discuss-phase handles HOW to implement.
- `commands/gsd/spec-phase.md` — slash command routing to workflow
- `get-shit-done/workflows/spec-phase.md` — full Socratic interview loop (up to 6
rounds, 5 rotating perspectives: Researcher, Simplifier, Boundary Keeper, Failure
Analyst, Seed Closer) with weighted 4-dimension ambiguity gate (≤ 0.20 to write SPEC.md)
- `get-shit-done/templates/spec.md` — SPEC.md template with falsifiable requirements
(Current/Target/Acceptance per requirement), Boundaries, Acceptance Criteria,
Ambiguity Report, and Interview Log; includes two full worked examples
- `get-shit-done/workflows/discuss-phase.md` — new `check_spec` step detects
`{padded_phase}-SPEC.md` at startup; displays "Found SPEC.md — N requirements
locked. Focusing on implementation decisions."; `analyze_phase` respects `spec_loaded`
flag to skip "what/why" gray areas; `write_context` emits `<spec_lock>` section
with boundary summary and canonical ref to SPEC.md
- `docs/ARCHITECTURE.md` — update command/workflow counts (74→75, 71→72)
Closes#2213
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(hooks): add gsd-read-injection-scanner PostToolUse hook (#2201)
Adds a new PostToolUse hook that scans content returned by the Read tool
for prompt injection patterns, including four summarisation-specific patterns
(retention-directive, permanence-claim, etc.) that survive context compression.
Defense-in-depth for long GSD sessions where the context summariser cannot
distinguish user instructions from content read from external files.
- Advisory-only (warns without blocking), consistent with gsd-prompt-guard.js
- LOW severity for 1-2 patterns, HIGH for 3+
- Inlined pattern library (hook independence)
- Exclusion list: .planning/, REVIEW.md, CHECKPOINT, security docs, hook sources
- Wired in install.js as PostToolUse matcher: Read, timeout: 5s
- Added to MANAGED_HOOKS for staleness detection
- 19 tests covering all 13 acceptance criteria (SCAN-01–07, EXCL-01–06, EDGE-01–06)
Closes#2201
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(ci): add read-injection-scanner files to prompt-injection-scan allowlist
Test payloads in tests/read-injection-scanner.test.cjs and inlined patterns
in hooks/gsd-read-injection-scanner.js legitimately contain injection strings.
Add both to the CI script allowlist to prevent false-positive failures.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(test): assert exitCode, stdout, and signal explicitly in EDGE-05
Addresses CodeRabbit feedback: the success path discarded the return
value so a malformed-JSON input that produced stdout would still pass.
Now captures and asserts all three observable properties.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(install): replace all ~/.claude/ paths in generated Codex .toml files (#2320)
installCodexConfig() only rewrote get-shit-done/-scoped paths; all other
~/.claude/ references (hooks, skills, configDir) leaked into generated .toml
files unchanged. Add three additional regex replacements to catch $HOME/.claude/,
~/.claude/, and ./.claude/ patterns and rewrite them to .codex equivalents.
Adds regression test PATHS-01.
Closes#2320
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(install): handle bare .claude end-of-string and scan all .toml files (CR feedback)
- Use capture group (\/|$) so replacements handle both ~/.claude/ and bare
~/.claude at end of string, not just the trailing-slash form
- Expand PATHS-01 test to scan agents/*.toml + top-level config.toml
- Broaden leak pattern to match ./.claude, ~, and $HOME variants with or
without trailing slash
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* docs: sync ARCHITECTURE.md command count to 74
commands/gsd/ has 74 .md files; the two count references in
ARCHITECTURE.md still said 73. Fixes the command-count-sync
regression test.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: embed model_overrides in Codex TOML and OpenCode agent files (#2256)
Codex and OpenCode use static agent files (TOML / markdown frontmatter)
rather than inline Task(model=...) parameters, so model_overrides set in
~/.gsd/defaults.json was silently ignored — all subagents fell through to
the runtime's default model.
Fix: at install time, read model_overrides from ~/.gsd/defaults.json and
embed the matching model ID into each agent file:
- Codex: model = "..." field in the agent TOML (generateCodexAgentToml)
- OpenCode: model: ... field in agent frontmatter (convertClaudeToOpencodeFrontmatter)
Also adds readGsdGlobalModelOverrides() helper and passes the result
through installCodexConfig() and the OpenCode agent install loop.
Closes#2256
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(commands): add gsd:inbox command for GitHub issue/PR triage
inbox.md was created but not committed, causing the command count
to read 73 in git while ARCHITECTURE.md correctly stated 74.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Absolute hook paths in settings.json break when ~/.claude is bind-mounted
into a container at a different path, or when running under WSL with a
Windows Node.js that resolves a different home directory.
Add `--portable-hooks` CLI flag and `GSD_PORTABLE_HOOKS=1` env var opt-in.
When set, buildHookCommand() emits `$HOME`-relative paths instead of resolved
absolute paths, making the generated hook commands portable across bind mounts.
Fixes#2190
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Local installs write to .claude/settings.json inside the project, which
takes precedence over the user's global ~/.claude/settings.json. Writing
statusLine here silently clobbers any profile-level statusLine the user
configured. Guard the write with !isGlobal && !forceStatusline; pass
--force-statusline to override.
Closes#2248
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(hooks): stamp gsd-hook-version in .sh hooks and fix stale detection regex (#2136, #2206)
Three-part fix for the persistent "⚠ stale hooks — run /gsd-update" false
positive that appeared on every session after a fresh install.
Root cause: the stale-hook detector (gsd-check-update.js) could only match
the JS comment syntax // in its version regex — never the bash # syntax used
in .sh hooks. And the bash hooks had no version header at all, so they always
landed in the "unknown / stale" branch regardless.
Neither partial fix (PR #2207 regex only, PR #2215 install stamping only) was
sufficient alone:
- Regex fix without install stamping: hooks install with literal
"{{GSD_VERSION}}", the {{-guard silently skips them, bash hook staleness
permanently undetectable after future updates.
- Install stamping without regex fix: hooks are stamped correctly with
"# gsd-hook-version: 1.36.0" but the detector's // regex can't read it;
still falls to the unknown/stale branch on every session.
Fix:
1. Add "# gsd-hook-version: {{GSD_VERSION}}" header to
gsd-phase-boundary.sh, gsd-session-state.sh, gsd-validate-commit.sh
2. Extend install.js (both bundled and Codex paths) to substitute
{{GSD_VERSION}} in .sh files at install time (same as .js hooks)
3. Extend gsd-check-update.js versionMatch regex to handle bash "#"
comment syntax: /(?:\/\/|#) gsd-hook-version:\s*(.+)/
Tests: 11 new assertions across 5 describe blocks covering all three fix
parts independently plus an E2E install+detect round-trip. 3885/3885 pass.
Approach credit: PR #2207 (j2h4u / Maxim Brashenko) for the regex fix;
PR #2215 (nitsan2dots) for the install.js substitution approach.
Closes#2136, #2206, #2209, #2210, #2212
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(hooks): extract check-update worker to dedicated file, eliminating template-literal regex escaping
Move stale-hook detection logic from inline `node -e '<template literal>'` subprocess
to a standalone gsd-check-update-worker.js. Benefits:
- Regex is plain JS with no double-escaping (root cause of the (?:\\/\\/|#) confusion)
- Worker is independently testable and can be read directly by tests
- Uses execFileSync (array args) to satisfy security hook that blocks execSync
- MANAGED_HOOKS now includes gsd-check-update-worker.js itself
Update tests to read worker file instead of main hook for regex/configDir assertions.
All 3886 tests pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Codex install registered gsd-check-update.js in config.toml but never
copied the hook file to ~/.codex/hooks/. The hook-copy block in install()
was gated by !isCodex, leaving a broken reference on every fresh Codex
global install.
Adds a dedicated hook-copy step inside the isCodex branch that mirrors
the existing copy logic (template substitution, chmod). Adds a regression
test that verifies the hook file physically exists after install.
Closes#2153
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Three install code paths were leaking Claude-specific references into
Qwen installs: copyCommandsAsClaudeSkills lacked runtime-aware content
replacement, the agents copy loop had no isQwen branch, and the hooks
template loop only replaced the quoted '.claude' form. Added CLAUDE.md,
Claude Code, and .claude/ replacements across all three paths plus
copyWithPathReplacement's Qwen .md branch. Includes regression test
that walks the full .qwen/ tree after install and asserts zero Claude
references outside CHANGELOG.md.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add isQwen branch in copyWithPathReplacement for .md files converting
CLAUDE.md to QWEN.md and 'Claude Code' to 'Qwen Code'
- Add isQwen branch in copyWithPathReplacement for .js/.cjs files
converting .claude paths to .qwen equivalents
- Add Qwen Code program and command labels in finishInstall() so the
post-install message shows 'Qwen Code' instead of 'Claude Code'
Closes#2081
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Adds Qwen Code as a supported installation target. Users can now run
`npx get-shit-done-cc --qwen` to install all 68+ GSD commands as skills
to `~/.qwen/skills/gsd-*/SKILL.md`, following the same open standard as
Claude Code 2.1.88+.
Changes:
- `bin/install.js`: --qwen flag, getDirName/getGlobalDir/getConfigDirFromHome
support, QWEN_CONFIG_DIR env var, install/uninstall pipelines, interactive
picker option 12 (Trae→13, Windsurf→14, All→15), .qwen path replacements in
copyCommandsAsClaudeSkills and copyWithPathReplacement, legacy commands/gsd
cleanup, fix processAttribution hardcoded 'claude' → runtime-aware
- `README.md`: Qwen Code in tagline, runtime list, verification commands,
skills format NOTE, install/uninstall examples, flag reference, env vars
- `tests/qwen-install.test.cjs`: 13 tests covering directory mapping, env var
precedence, install/uninstall lifecycle, artifact preservation
- `tests/qwen-skills-migration.test.cjs`: 11 tests covering frontmatter
conversion, path replacement, stale skill cleanup, SKILL.md format validation
- `tests/multi-runtime-select.test.cjs`: Updated for new option numbering
Closes#2019
Co-authored-by: Muhammad <basirovmb1988@gmail.com>
Co-authored-by: Jonathan Lima <eezyjb@gmail.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(install): guard writeSettings against null settingsPath for cline runtime
Cline returns settingsPath: null from install() because it uses .clinerules
instead of settings.json. The finishInstall() guard was missing !isCline,
causing a crash with ERR_INVALID_ARG_TYPE when installing with the cline runtime.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
* test(cline): add regression tests for ERR_INVALID_ARG_TYPE null settingsPath guard
Adds two regression tests to tests/cline-install.test.cjs for gsd-build/get-shit-done#2044:
- Assert install(false, 'cline') does not throw ERR_INVALID_ARG_TYPE
- Assert settings.json is not written for cline runtime
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
* test(cline): fix regression tests to directly call finishInstall with null settingsPath
The previous regression tests called install() which returns early for cline
before reaching finishInstall(), so the crash was never exercised. Fix by:
- Exporting finishInstall from bin/install.js
- Calling finishInstall(null, null, ..., 'cline') directly so the null
settingsPath guard is actually tested
Tests now fail (ERR_INVALID_ARG_TYPE) without the fix and pass with it.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
* fix(autonomous): add Agent to allowed-tools in gsd-autonomous skill
Closes#2043
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(install): extend buildHookCommand to .sh hooks — absolute quoted paths
- Extend buildHookCommand() to branch on .sh suffix, using 'bash' runner
instead of 'node', so all hook paths go through the same quoted-path
construction: bash "/absolute/path/hooks/gsd-*.sh"
- Replace three manual 'bash ' + targetDir + '...' concatenations for
gsd-validate-commit.sh, gsd-session-state.sh, gsd-phase-boundary.sh
with buildHookCommand(targetDir, hookName) for the global-install branch
- Global .sh hook paths are now double-quoted, fixing invocation failure
when the config dir path contains spaces (Windows usernames, #2045)
- Adds regression tests in tests/sh-hook-paths.test.cjs
Closes#2045Closes#2046
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Cline was documented as a supported runtime but was absent from
bin/install.js. This adds full Cline support:
- Registers --cline CLI flag and adds 'cline' to --all list
- Adds getDirName/getConfigDirFromHome/getGlobalDir entries (CLINE_CONFIG_DIR env var respected)
- Adds convertClaudeToCliineMarkdown() and convertClaudeAgentToClineAgent()
- Wires Cline into copyWithPathReplacement(), install(), writeManifest(), finishInstall()
- Local install writes to project root (like Claude Code), not .cline/ subdirectory
- Generates .clinerules at install root with GSD integration rules
- Installs get-shit-done engine and agents with path/brand replacement
- Adds Cline as option 4 in interactive menu (13-runtime menu, All = 14)
- Updates banner description to include Cline
- Exports convertClaudeToCliineMarkdown and convertClaudeAgentToClineAgent for testing
- Adds tests/cline-install.test.cjs with 17 regression tests
- Updates multi-runtime-select, copilot-install, kilo-install tests for new option numbers
Fixes#1991
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The convertSlashCommandsToCodexSkillMentions function only converted
colon-style skill invocations (/gsd:command) but not hyphen-style
command references (/gsd-command) used in workflow output templates
(Next Up blocks, phase completion messages, etc.). This caused Codex
users to see /gsd- prefixed commands instead of $gsd- in chat output.
- Add regex to convert /gsd-command → $gsd-command with negative
lookbehind to exclude file paths (e.g. bin/gsd-tools.cjs)
- Strip /clear references in Codex output (no Codex equivalent)
- Add 5 regression tests covering command conversion, path
preservation, and /clear removal
Co-authored-by: Lakshman <lakshman@lakshman-GG9LQ90J61.local>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Running gsd-update (re-running the installer) silently deleted two
user-generated files:
- get-shit-done/USER-PROFILE.md (created by /gsd-profile-user)
- commands/gsd/dev-preferences.md (created by /gsd-profile-user)
Root causes:
1. copyWithPathReplacement() calls fs.rmSync(destDir, {recursive:true})
before copying, wiping USER-PROFILE.md with no preserve allowlist.
2. The legacy commands/gsd/ cleanup at ~line 5211 rmSync'd the entire
directory, wiping dev-preferences.md.
3. The backup path in profile-user.md pointed to the same directory
that gets wiped, so the backup was also lost.
Fix:
- Add preserveUserArtifacts(destDir, fileNames) and restoreUserArtifacts()
helpers that save/restore listed files around destructive wipes.
- Call them in install() before the get-shit-done/ copy (preserves
USER-PROFILE.md) and before the legacy commands/gsd/ cleanup
(preserves dev-preferences.md).
- Fix profile-user.md backup path from ~/.claude/get-shit-done/USER-PROFILE.backup.md
to ~/.claude/USER-PROFILE.backup.md (outside the wiped directory).
Closes#1924
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Local installs wrote bare relative paths (e.g. `node .claude/hooks/...`)
into settings.json. Claude Code persists the shell's cwd between tool
calls, so a single `cd subdir` broke every hook for the rest of the
session.
Prefix all 9 local hook commands with "$CLAUDE_PROJECT_DIR"/ so path
resolution is always anchored to the project root regardless of cwd.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The installer writes gsd-file-manifest.json to the runtime config root
at install time but uninstall() never removed it, leaving stale metadata
after every uninstall. Add fs.rmSync for MANIFEST_NAME at the end of the
uninstall cleanup sequence.
Regression test: tests/bug-1908-uninstall-manifest.test.cjs covers both
global and local uninstall paths.
Closes#1908
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Add fs.existsSync() guards to all .js hook registrations in install.js,
matching the pattern already used for .sh hooks (#1817). When hooks/dist/
is missing from the npm package, the copy step produces no files but the
registration step previously ran unconditionally for .js hooks, causing
"PreToolUse:Bash hook error" on every tool invocation.
Each .js hook (check-update, context-monitor, prompt-guard, read-guard,
workflow-guard) now verifies the target file exists before registering
in settings.json, and emits a skip warning when the file is absent.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(installer): deploy commands directory in local installs (#1736)
Local Claude installs now populate .claude/commands/gsd/ with command .md
files. Claude Code reads local project commands from .claude/commands/gsd/,
not .claude/skills/ — only the global ~/.claude/skills/ is used for the
skills format. The previous code deployed skills/ for both global and local
installs, causing all /gsd-* commands to return "Unknown skill" after a
local install.
Global installs continue to use skills/gsd-xxx/SKILL.md (Claude Code 2.1.88+
format). Local installs now use commands/gsd/xxx.md (the format Claude Code
reads for local project commands).
Also adds execute-phase.md to the prompt-injection scan allowlist (the
workflow grew past 50K chars, matching the existing discuss-phase.md exemption).
Closes#1736
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(installer): fix test cleanup pattern and uninstall local/global split (#1736)
Replace try/finally with t.after() in all 3 regression tests per CONTRIBUTING.md
conventions. Split the Claude Code uninstall branch on isGlobal: global removes
skills/gsd-*/ directories (with legacy commands/gsd/ cleanup), local removes
commands/gsd/ as the primary install location since #1736.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(install): preserve non-array hook entries during uninstall
Uninstall filtering returned null for hook entries without a hooks
array, silently deleting user-owned entries with unexpected shapes.
Return the entry unchanged instead so only GSD hooks are removed.
* test(install): add regression test for non-array hook entry preservation (#1825)
Fix mirrored filterGsdHooks helper to match production code and add
test proving non-array hook entries survive uninstall filtering.
Before registering each .sh hook (validate-commit, session-state, phase-boundary),
check that the target file was actually copied. If the .sh file is missing (e.g.
omitted from the npm package as in v1.32.0), skip registration and emit a warning
instead of writing a broken hook entry that errors on every tool invocation.
Closes#1817
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Community PRs repeatedly add marketing commentary in parentheses next to
product names (licensing model, parent company, architecture). Product
listings should contain only the product name.
Cleaned across 8 files in 5 languages (EN, KO, JA, ZH, PT) plus
install.js code comments and CHANGELOG. Added static analysis guard
test that prevents this pattern from recurring.
Fixes#1777
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
The hook was built, copied to hooks/dist/, and installed to disk — but
never registered as a PreToolUse entry in settings.json, making the
hooks.workflow_guard config flag permanently inert.
Adds the registration block following the same pattern as the other
community hooks (prompt-guard, read-guard, validate-commit, etc.).
Includes regression test that verifies every JS hook in gsdHooks has a
corresponding command construction and registration block.
Fixes#1767
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Addresses three findings from Codex adversarial review of #1768:
- Uninstall settings cleanup now filters at per-hook granularity instead of
per-entry. User hooks that share an entry with a GSD hook are preserved
instead of being removed as collateral damage.
- Add gsd-workflow-guard to PreToolUse/BeforeTool uninstall settings filter
so opt-in users don't get dangling references after uninstall.
- Codex install now strips legacy gsd-update-check.js hook entries before
appending the corrected gsd-check-update.js, preventing duplicate hooks
on upgrade from affected versions.
- 8 new regression tests covering per-hook filtering, legacy migration regex.
Fixes#1755
