mirror of
https://github.com/garrytan/gstack.git
synced 2026-10-31 22:14:25 +01:00
+8








28f1385eac
* fix: sourced helpers locate themselves under zsh (#3024) Skill blocks source bin/gstack-codex-probe and bin/gstack-egress-lib.sh into the host's shell. Under zsh (macOS default) BASH_SOURCE is empty, so Codex model selection resolved /../scripts/... and every outside pass reported CODEX_MODEL: invalid; the egress lib resolved to the caller's cwd. Resolve the file once at source time (BASH_SOURCE under bash, %x under zsh), and replace the bash-only ${!name} indirection in the auth probe. A free test derives every sourced helper from the generated skills and sources each under bash and zsh. Co-authored-by: marcosdetry <marcosdetry@users.noreply.github.com> Co-authored-by: akshaykalia <akshaykalia@users.noreply.github.com> Co-authored-by: scottgigante-hubflow <scottgigante-hubflow@users.noreply.github.com> Co-authored-by: woodsonl <woodsonl@users.noreply.github.com> Co-authored-by: bezdekt <bezdekt@users.noreply.github.com> * fix: reply and issue templates never pass untrusted text in double quotes (#3046) greptile-triage's reply APIs posted -f body="<reply text>" while the templates quote SHAs and diff code in backticks; in double quotes bash ran each span (replies posted as 'Fixed in .', and diff or reviewer text ran as a command). The pre-existing-failure issue template and /spec's issue title had the same shape. All three now go through quoted heredocs. The generated-bash lint gains untrusted-in-quotes and now also covers runtime reference docs in skill directories (review/greptile-triage.md). Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: a Codex usage limit is its own state, not 'inconclusive; proceeding' (#3051) The model probe matched only model and broken-install errors, so a usage or rate limit fell through to MODEL_PROBE_INCONCLUSIVE: Codex's own message (with the reset time) was dropped, the hint pointed at model config, and every later skill re-probed for 30 s. The probe now returns MODEL_QUOTA_EXHAUSTED (exit 4) on a failed, non-timeout call that names a usage limit, rate limit, insufficient_quota or 429, relays Codex's line verbatim and caches it for 15 minutes. Preflights map it to CODEX_MODE: quota_exhausted (missing coverage, no further Codex calls), and classifyOutsideReview reports a mid-run limit as unavailable (quota) with its own troubleshooting anchor. Co-authored-by: venkatareddyg <venkatareddyg@users.noreply.github.com> * test: record ship/review union sizes after the quota branch and issue template * fix: Codex watchdog freezes the command before reaping its children The bash-native watchdog (stock macOS, no timeout binary) ran pkill -P on the command's children and then killed the command. Between the two, the command continued past its killed child and printed output after its deadline. Stop it first. The test's minimal PATH also lacked pkill, which the watchdog uses to reap children (stock macOS ships it), so the orphaned child held stdout open and the capture waited out its sleep. * fix(autoplan): an oversize session journal is a hard too_large denial, not a retryable identity failure (#3050) The guard capped journal reads at 32 MiB and reported a longer journal as 'identity', which the hook treats as deny-and-retry. A journal only grows, so /autoplan could never enter Phase 1 in a long session, and the troubleshooting row sent people chasing symlinks. Oversize journals now report too_large as a hard cause that names the size and the supported fallback (run the reviews by hand or start a new session). Co-authored-by: venkatareddyg <venkatareddyg@users.noreply.github.com> * fix(deps): clear the weekly OSV findings (#3040) Raise overrides to fixed releases: ip-address 10.7.3 (our own pin was the vulnerable 10.3.1; reachable through socks in browse's proxy path), fast-uri 3.1.8, hono 4.13.13, proxy-addr 2.0.8, @modelcontextprotocol/sdk 1.31.0, sharp 0.35.5, and in lib/diagram-render dompurify 3.4.16 (bundled by mermaid) and source-map-js 1.2.2; the render page is rebuilt. Three advisories without a usable fix get justified, expiring ignores: braces (build-time only, excalidraw -> sass -> chokidar; a test checks the page carries no chokidar code), sprintf-js (roarr's constant-format logger) and katex (needs pre-existing prototype pollution; fix outside mermaid's range). osv-scanner with the repo config: no issues found. * fix(autoplan): name a shell-built snapshot init as the cause of the missing-evidence denial (#3045) The guard binds a phase entry only to a literal snapshot-init Bash call. When the model resolved the tool path into a variable, chained or piped the call, init ran and succeeded but the denial said 'Complete the existing snapshot init step', which reads as already done, so the run gave up. When no init binds but a Bash call looks like one, the denial now says the shape cannot be bound and prints the literal re-run (answers reused:true). The binding rule is unchanged, and the skill says to run init on its own. Co-authored-by: CharlesBerg85 <CharlesBerg85@users.noreply.github.com> * fix(browse): chain reads piped stdin synchronously, so Windows runs the flow (#3039) On Windows, an awaited Bun.stdin.text() inside the CLI's un-awaited main() let the process exit 0 before reading stdin: a replayed flow reported success with nothing sent to the daemon. Read fd 0 synchronously (skipped for a terminal); empty input now gets the server's usage error and a non-zero exit. Co-authored-by: VaibDeveloper <VaibDeveloper@users.noreply.github.com> * fix(browse): clearer config-mismatch refusal and a WebGL note for the macOS GPU flags (#3030) A plain call refused because another session started the daemon with --headed/--proxy was told to 'apply --proxy/--headed', flags it never passed. It now says who started the daemon and offers both remedies. The BROWSER.md env table notes that the default macOS GPU flags disable WebGL and that GSTACK_DISABLE_GPU=off restores it for WebGL QA. Co-authored-by: Hagetino <Hagetino@users.noreply.github.com> * fix(ship): skip Greptile triage when no PR exists yet (#3020) /ship's first run on a branch dispatched a triage subagent only to learn there was no PR. Step 10 now checks for a PR inline (gh's 'no pull requests found') and records that Greptile runs on the PR once it exists; other gh errors still dispatch so the child can report unavailable. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix(router): route only to loaded skills, and say when not to invoke (#3018) The root router listed every skill and ended 'When in doubt, invoke the skill'. Current Claude models follow that literally: they tried to invoke skills a user had turned off in Claude Code's skillOverrides (which gstack's own disabled_skills render cannot see), and borderline requests tipped into full workflows. The router now routes only to skills in the session's available-skills list and names when to answer directly; the CLAUDE.md routing block skill-start offers says the same. Co-authored-by: Rekin226 <Rekin226@users.noreply.github.com> * fix: spawned sessions never auto-grant consent or publish off the machine (#3016) Spawned sessions auto-choose the recommended option except for destructive or irreversible choices. Main already withholds the artifacts-sync consent prompt from spawned sessions; the remaining gap was any other prompt whose recommended option grants a consent or syncs data off the machine. Every spawned directive (skill-start's block, the escape sentence, the Conductor deny reason and the AUQ-error fallback) now names that as irreversible. A consent the user already gave interactively stays in force. Co-authored-by: ekson73 <ekson73@users.noreply.github.com> * test: fix the spawned-block slice and type the #3045 re-init events * fix(redact): TypeScript types and JSX expressions are code, not env.kv secrets (#3048) A credential-named parameter or prop with an unquoted code value (`session: SessionState,`, `key={turn.requestId + turn.role}`, `apiKey={settings.apiKey}`) cleared the entropy gate and blocked pushes as a MEDIUM env.kv finding. One rule replaces the PR's two keyword heuristics: a value that reads as a code name (no digits, word-like camelCase or dotted parts) with a code terminator, or a brace expression of such names with no quoted literal, is source syntax. Paired controls keep firing on quoted, brace-wrapped, concatenated, YAML-bare and random-letter literals. Co-authored-by: tianxchen <tianxchen@users.noreply.github.com> * fix(setup): refuse to register a hook that does not parse (#3032) Claude Code runs gstack's hook shims through /bin/sh; a shim that does not parse exits 2, which for a PreToolUse hook blocks the tool call in every session until someone finds the file. Before any hook registration is healed or written, setup now runs bash -n on each registered shim and a Bun parse of its TypeScript entry, and stops with the file, line and fix. A lean rewrite of the PR's idea (one gate block plus a test that runs it). Co-authored-by: BenjaminDSmithy <BenjaminDSmithy@users.noreply.github.com> * fix(browse): Windows runtime roots start the server from the source checkout's bundle (#3026) On Windows a host runtime root (~/.codex/skills/gstack and the other env-var hosts) holds file copies of browse/dist and no node_modules, so its server-node.mjs could not import playwright and every browser skill died (the module-scope server.ts half was fixed in v1.91.19.0). setup now records the source checkout in every runtime root's .source-path, and the CLI runs that checkout's bundle when the adjacent one cannot reach playwright. Verified by unit layout tests and the Windows CI lane, not live on Windows. Co-authored-by: ubay887 <ubay887@users.noreply.github.com> * fix(copilot): on Windows, run bash blocks in Git for Windows Bash (#3047) Copilot CLI on Windows ran gstack's extensionless helpers straight from PowerShell, which opened an app picker, and bare bash can resolve to WSL. The Copilot tool glossary (rendered into every Copilot skill) now says to run each bash block in Git for Windows Bash by its full path and never to translate a block into PowerShell. The PR's PowerShell repair skill and README section are not taken. Co-authored-by: nareshh74 <nareshh74@users.noreply.github.com> * test: record ship's union size after the Greptile no-PR check * fix: keep issue numbers out of model-read Copilot text; seed the new routing sentence in the autoplan fixture * docs: Oct 6 fix-wave plan, reviewed through /autoplan * fix(autoplan): every oversize-journal path reports too_large with a recovery that keeps the work Review follow-up to the #3050 fix. The summed batch read and the narrowed parent loop now report too_large with the real size instead of a plain error naming a hard-coded 32 MiB. The denial names the size and the cap, and gives the recovery that keeps work (/context-save, new session, /context-restore, /autoplan) with a docs anchor. A test seam that can only lower the cap replaces the 33 MiB fixture. * test(autoplan): negative controls and more shapes for the shell-built init denial * fix(redact): decide TS/JSX source syntax by file context, not by value shape Review follow-up to the #3048 fix. The word-shape heuristic also exempted real secrets made of words (API_KEY=VelvetRiverOrbitSunset; in a .env). The scanner now takes the file path (ScanOptions.sourcePath): the pre-push hook scans the pushed diff per file with its +++ b/ path, and gstack-redact --from-file passes its path. Only in a TypeScript/JSX file is an unquoted value after ':' or a JSX brace of names treated as source; the same line in .env/YAML, or with no path, is scanned as a literal, and the jwt pattern still fires in .ts files. Co-authored-by: tianxchen <tianxchen@users.noreply.github.com> * docs: gstack's own CLAUDE.md routing block matches the new router wording Live check (Claude Code 2.1.284): a skill set to "off" in skillOverrides is absent from the session's skill list while its sibling is listed, and both are listed without the override, so 'route only to skills in your available-skills list' holds. * fix(sync-gbrain): don't offer a call-graph build for a source with no code pages (#3029) Step 3.5 offered --dream for any source with pages, including one that holds only markdown (synced with gbrain's default strategy), where a build cannot help, and the injected CLAUDE.md guidance told future sessions to run --dream for any count: 0. On gbrain >= 0.60, code-def for an impossible symbol answers .status (probed live against gbrain 0.60.96: ready for a code source, out_of_scope for a markdown-only one). Step 3.5 now probes the pinned source first, reports the cause instead of asking on out_of_scope, and keeps today's offer when the probe is unknown (older gbrain, an error). gstack's own sync uses --strategy code, so the PR's soft-delete claim does not apply to gstack's path. Co-authored-by: syong <syong@users.noreply.github.com> * fix(setup): register the hooks that parse, refuse the rest with file:line (#3032) The landed gate stopped setup before any registration, so one unparseable hook left every other hook unregistered. On the Claude install the canonical root is the checkout Claude Code runs hooks from, so there is no staged swap to abort: setup now registers the hooks that parse, refuses the ones that do not with file and line, finishes the install and exits non-zero, saying it is a gstack bug and that a skipped hook keeps running from the checkout. bin/gstack-hook-check derives its list from setup's _hook_command_path calls plus the autoplan phase-publication hook (a test keeps them equal), parses each shim with its shebang's interpreter, and bundles each TypeScript entry with its local imports (packages external) into a temp dir, so a broken or missing import, or a missing entry, fails. Co-authored-by: BenjaminDSmithy <BenjaminDSmithy@users.noreply.github.com> * fix(auto-update): hold an update whose hooks do not parse, before the checkout moves bin/gstack-session-update fast-forwards the live checkout before setup runs, and Claude Code runs hooks straight from it, so setup's gate cannot protect an auto-update. After the Bun floor check the updater extracts the incoming revision with git archive and runs bin/gstack-hook-check on it; a failure holds the update like bun-too-old, records hook-does-not-parse with file:line in the update status, and the next session start says the current hooks keep running. Fixture-remote tests cover a broken shim, a broken imported module and a parseable control. Co-authored-by: BenjaminDSmithy <BenjaminDSmithy@users.noreply.github.com> * fix(browse): chain with no flow is a usage error before any daemon boots (#3039) Empty piped input, a terminal with no arguments, and a readFileSync(0) error (EAGAIN, EOF) now print the chain usage with the cause and its troubleshooting link and exit 1 before ensureServer(), instead of sending an empty argument to a daemon the CLI booted for it. Tests run the real CLI with no daemon and no browser, plus the stub-daemon flow. Co-authored-by: VaibDeveloper <VaibDeveloper@users.noreply.github.com> * fix(browse): refuse a runtime-root CLI and checkout server from different builds (#3026) A Windows runtime root runs the source checkout's Node server bundle through .source-path. When the root's CLI and the checkout's bundle carry different browse/dist/.version builds (the checkout was rebuilt without refreshing the root), the CLI now refuses to start the server with both builds, the ./setup fix and its troubleshooting anchor. Unversioned dev builds are not called skew. Co-authored-by: ubay887 <ubay887@users.noreply.github.com> * ci(windows): real isolated ./setup install, then a piped chain flow through the installed browse windows-setup-e2e built the checkout but never ran ./setup. It now installs the codex host into an isolated HOME (a runtime root copy with no node_modules), then from an unrelated directory pipes a two-command chain flow at a loopback page into the installed browse and asserts the page text, and that empty piped input exits 1 with the usage error. Covers the checkout-bundle path and the piped-stdin path end to end. * test(lint): free text never enters a generated command (structural rule) The generated-bash lint's untrusted-in-quotes rule (gh/glab/git/curl text flags in double quotes only) becomes free-text-placeholder: any <placeholder> anywhere in any command, in double, single or no quotes, and in heredoc bodies (quoted or not), unless it is on the reviewed identifier allowlist (test/helpers/placeholder-allowlist.ts) with the grammar a value must match. Quoted-only identifiers (URLs, selectors, paths printed by gstack) must sit inside quotes. Backslash-newline continuations now read as one command, so a flag and its value on separate lines are still seen together. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: reply, issue and spec text travel as agent-written files, never through a heredoc Replaces the quoted-heredoc bodies with CEO-12's file mechanism: a bash block creates a private mktemp file in the project's .gstack/tmp/ (kept out of git through .git/info/exclude), the agent writes the text with its file-write tool, and the next block posts with -F body=@file, --body-file or "$(cat file)". A heredoc ends early at any body line equal to its delimiter; a file has no such line. When the file was never written the block refuses with the path and the command to send it by hand, and it deletes the file after a successful post. - review/greptile-triage.md: replies; the fetched JSON moves from fixed /tmp/greptile_*.json paths to a mktemp directory; REPO and PR are re-derived per block instead of read from a previous shell. - Test-failure issues: title and body files, platform case on github/gitlab, assignee only when it is a valid login. - /spec: one draft file feeds the audit hash, every redaction scan, the issue and the archive; the archive no longer expands <title>/<body> inside an unquoted heredoc; the decision-log JSON is built by jq from files. Location verified against Claude Code 2.1.284 (Dockerfile.ci pin): project-local .gstack/tmp is writable under acceptEdits with no prompt and keeps mode 0600; $TMPDIR and the gstack state root prompt even under acceptEdits. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix(ship): Greptile triage runs only when a PR exists; every other case says why PR: unknown (gh missing, not logged in, a non-GitHub remote, or a failed lookup) used to dispatch the triage child anyway. The Step 10 check now tests gh, its login, the remote and the PR in that order, and every case but 'PR: exists' skips the dispatch and records 'Greptile: not run (<reason>); runs on the PR once it exists'. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: question-tuning commands carry no free text The question log took the summary in single quotes and the preference write took the user's own words; an apostrophe ended the quote and the rest ran as shell. The summary is now a kebab-case slug (an identifier with a grammar), and the optional free_text field is no longer passed from skills (the binary still accepts it). * fix(cso): Windows finds docker.exe only under known-folder roots and says transport is unsupported (#3028) On Windows /cso looked for docker only in Git's directory and System32, so Docker Desktop was never found and the gap read as 'not installed'. It now resolves Program Files, Program Files (x86) and the Windows directory through PowerShell's [Environment]::GetFolderPath (not the process environment), checks each candidate's real path against those roots and rejects reparse points. lib/cso/docker.ts admits only unix:/// endpoints, so a trusted docker.exe reports 'native Windows Docker transport is not supported yet; static assessment only'; one outside the roots is refused as outside the trusted install locations, not overridable. Both carry an anchor. Native named-pipe transport is not implemented. * fix: sourced helpers share one self-locate block and fail loudly when they cannot locate themselves (#3024 rework) The codex probe and the egress lib now carry one byte-identical block: BASH_SOURCE under bash or %x under zsh, made absolute with $PWD, else a set GSTACK_ROOT that holds the helper. Anything else (dash, sh, a shell that cannot name the file) prints "gstack: cannot locate <helper> (shell: ...)" with the docs/troubleshooting.md#sourced-helper-location URL and returns before any bash- or zsh-only syntax. The Claude-only default path and the $(pwd) fallback are gone. The generated Codex preflight dot-sources the probe and, when it cannot load, reports CODEX_MODE: helper_unavailable with the helper's own line instead of falling through to not_authed. The class guard enumerates nested sources, makes real resolver, auth, state-root and egress-home calls under bash and zsh (absolute, relative then cd, bare name via PATH), covers dash and an unknown shell, runs the full preflight, and lints BASH_SOURCE to the shared block plus other bash-only constructs. Co-authored-by: marcosdetry <marcosdetry@users.noreply.github.com> Co-authored-by: akshaykalia <akshaykalia@users.noreply.github.com> Co-authored-by: scottgigante-hubflow <scottgigante-hubflow@users.noreply.github.com> Co-authored-by: woodsonl <woodsonl@users.noreply.github.com> Co-authored-by: bezdekt <bezdekt@users.noreply.github.com> * fix(cso): a Maven or Gradle repository is Java, not Python (#3028) With no supported stack metadata, preparation defaulted to Python and asked a Java repository for requirements.txt. A root pom.xml, build.gradle or build.gradle.kts now names the stack as Java (unsupported for runtime preparation, so the prerequisite says so), and Python helper scripts or a root requirements.txt beside it no longer make it Python. * fix(browse): BROWSER.md says plainly how to enable WebGL; the mismatch refusal links Daemon discipline (#3030) The GSTACK_DISABLE_GPU row now states the effect and the switch: set GSTACK_DISABLE_GPU=off to enable WebGL (off is the only recognized value). The reworded config-mismatch refusal links BROWSER.md's Daemon discipline section. Free tests pin both. Co-authored-by: Hagetino <Hagetino@users.noreply.github.com> * test(windows): name the B2, B4, B5 and B12 tests in the curated Windows lane The runtime-root bundle tests lived in browse/test/config.test.ts and the Copilot glossary test in test/host-config.test.ts, both excluded from the Windows lane, so the two Windows-only fixes were never checked on Windows. Move each into its own Windows-safe file (unchanged assertions) and list it, the piped-stdin chain test and the /cso docker.exe discovery test as named Windows coverage in scripts/lib/windows-curation.ts. * fix: the free-text block exits loudly when mktemp fails; spec and redact tests follow the file mechanism The redact resolver test, the #2679 mktemp guard test, the /spec invariant and the spec secret-sink harness pinned the heredoc transport. They now pin the agent-written file: the harness runs the draft-file block, writes the draft the way the file-write tool would, then runs the real scan fence, with every HIGH/MEDIUM/scanner-failure assertion unchanged. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * test: hostile payloads through the generated posting blocks, identifier grammars, every quoting shape - The lint reads a whole command, so an unquoted multi-word placeholder is caught; negative controls cover double, single and no quotes, quoted and unquoted heredocs, continuation lines, and quoted-only identifiers used unquoted. - Every allowlisted grammar rejects quote-breaking characters (and shell operators and whitespace when used unquoted); git accepts 'fix;id' and 'fix`id`' as branch names, the ref grammar does not. - Greptile replies, test-failure issues (GitHub and GitLab) and /spec filing and archive run from the real generated text with stub gh/glab: a body with backticks, $(...), an apostrophe and lines equal to EOF and the old heredoc delimiters arrives byte for byte, nothing in it runs, and the temp file is gone. An unwritten file is refused with its path and the manual command, and nothing is sent. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * test(ship): the PR check precedes the Greptile dispatch, and each skip names its reason Runs Step 10's generated check against stub gh for gh missing, not logged in, a non-GitHub remote, no PR, a failed lookup and an existing PR. Fails on the previous text (which printed PR: unknown and dispatched). Co-authored-by: dgrant <dgrant@users.noreply.github.com> * feat(ship): ship-measure runner for the measure-then-fix loop (A6) scripts/ship-measure.ts owns /ship's diagnostic measurement of a red eval case: per-kind aggregation (rule trials; behavior panels of 3, each at its own 2 of 3 plus 90% of trials; judge outputs scored by 3-sample panels, standalone judge ids accepted), one artifact directory per baseline, repair round and trial, a diagnostic label with verdict null that refuses the eval history dir, an estimated admission budget that reserves every concurrent trial and reconciles actual cost, the ask-once rule, the repair-round limit, the unmeasured skip and the per-case PR table. Free-suite flakes (CEO-1): 'free' reruns one shard's exact file list N times, rerun 1 alone as the baseline and the rest in parallel at max(1, floor(cores / shard concurrency)), each with its own HOME, state root and flake ledger, GSTACK_FREE_RETRY_FLAKY=0, under a wall cap that reports a partial count. test:ubicloud --diagnostic runs the same on one VM with the retry off. bin/gstack-config gains the DX-7 keys with their CEO-9 defaults and rejecting validators. Free tests drive every kind, the budget stop, the unmeasured skip and the free reruns with fake runners. * fix: a Codex usage limit is quota_exhausted, a plain 429 is rate_limited, and only Codex's own error lines count (#3051 rework) The probe's quota signature used to treat any rate limit, "too many requests" or 429 as quota, and read the whole output, which echoes the prompt. Now only Codex's trailing ERROR: / stream error lines are read. A usage-limit or insufficient_quota line is MODEL_QUOTA_EXHAUSTED (exit 4, cached 15 minutes; quota wins over a 429 that names insufficient_quota); a plain 429 is MODEL_PROBE_RATE_LIMITED (exit 0, state rate_limited, never cached), reported as CODEX_MODE: unverified (rate_limited) so the review still runs and decides. The quota HINT says how many minutes gstack skips Codex and how to retry now (GSTACK_CODEX_PROBE_RETRY=1 skips the negative cache for that probe, or delete <state root>/.codex-model-probe), with the full troubleshooting URL. classifyOutsideReview reads the same trailing lines: a mid-review quota is unavailable (quota_exhausted), a mid-review 429 unavailable (rate_limited). New gate row rate_limited and docs anchor codex-rate-limited; the quota row documents the retry. Tests cover each line shape, the echoed prompt, exit 0, timeout, retry recovery, probe/classifier parity, and two fresh-shell preflight blocks where the second makes no paid Codex call. Co-authored-by: venkatareddyg <venkatareddyg@users.noreply.github.com> * ci: the eval image installs zsh, and the macOS lane sources the Codex probe from zsh Dockerfile.ci adds zsh so the sourced-helper guard runs (and fails loudly when zsh is missing) inside the eval container too. The macOS named regressions job sources bin/gstack-codex-probe from zsh in an unrelated directory and runs _gstack_codex_select_model exec (no paid call), then runs the sourced-helper guard on the macOS runner. A free test pins both lanes and runs the macOS step's script locally. * test: deterministic regression for the Codex watchdog freezing the command before reaping its children The existing stubborn-child test caught the reap-then-kill race only some of the time (2 of 3 runs on the pre-fix watchdog). A pkill that sleeps after reaping widens the window between reaping the children and killing the command, so a watchdog that does not SIGSTOP the command first always lets it print "late" after the deadline. Fails 3 of 3 on the pre-fix watchdog, passes 3 of 3 on the current one. * test: live codex smokes run codex in a private HOME and CODEX_HOME codex-resume-flag-semantics and codex-e2e-sol-scope probe the codex CLI (the latter at module load, in every free run), and every codex run writes $CODEX_HOME/tmp/arg0, so with a real codex on PATH they wrote into the developer's ~/.codex. Both now give codex a private HOME and CODEX_HOME and remove it afterwards. A free test runs each file against a stub codex that records HOME and CODEX_HOME and asserts neither is the caller's. * feat(ship): a red eval runs classify, measure, fix, re-measure, gate once (A6) Step 6 no longer stops at the first red paid eval: it reads the new lazily loaded measure section and runs the loop for each red case, and Step 5 measures a failed free-suite shard (diagnostic reruns, retry off) before the triage can call it pre-existing. The section states the diagnostic- not-verdict rule, the per-kind table, the ask-once and named-red stops, the unmeasured label and the honest-fix wording (observed k/n after fix at <cause>). A free test pins the steps and their order in the generated skill and checks every command and flag it names exists in the runner. Ship's parity cap rises to 1.455 (measured 272,944 / 187,706 = 1.4541). * docs(testing): diagnostic measurement is not a verdict; per-kind measure table (A6) * test(ship): seeded-flake paid case for the measure loop (A6, periodic, behavior) A fixture repo with no remote documents a single-case eval command and a gate command, both free local stubs. queue-priorities claims trial slots with atomic mkdir (reset per ship-measure round) and fails slots 2, 5 and 9 while src/priorities.js leaves its seed-shuffled list unsorted, printing failure_cause and failure_detail that name the line. The paid case gives /ship's generated Step 6 and measure section to a live session and asserts it measured the case alone, fixed the named line without touching the eval, re-measured at least 10 passing trials on the sorted file, and only then ran the gate once; any question other than the spend question is a contract violation. A free test measures the fixture through the real ship-measure CLI: 7/10 unsorted, 7/10 after an unrelated edit with the gate still red (ENG-13's negative control), 10/10 after the sort with the gate green. * test(cso): the missing-scanner case points Docker at a dead socket 'an optional scanner prerequisite does not turn completed manual coverage partial' assumed no usable Docker. On a machine with Docker and a pulled gitleaks image the scan completed and the test failed (also on main). The scan step now gets a DOCKER_HOST that cannot connect, which is the prerequisite-missing condition the test describes. * fix: Aside prompts travel as agent-written files, never inside the command The open-ended reading recipe (browse, devex-review), scrape's fuzzy-intent read and the shared web-research block put the model's question, query and reply format inside a double-quoted _aside_exec argument, where backticks and $(...) from page or repo text would run. The agent now writes the prompt into a private .gstack/tmp file; the send block rebuilds the path from the printed basename, refuses when the file is empty, and passes "$(cat "$PROMPT_FILE")". The fixed read-only sentence stays in the shell source. aside-driver pins move from the inline prompt to the file-based send block. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: design briefs and approval feedback travel as agent-written files Design briefs went into $D variants/generate/check as a double-quoted argument, and approval feedback was spliced into a single-quoted echo of JSON, so a quote, backtick or $(...) in either broke the command or ran. The agent now writes each brief into a private .gstack/tmp file and the block passes --brief "$(cat "$BRIEF_FILE")" (the CLI's --brief-file expects a JSON brief, so plain text uses --brief). variants and the later check reuse the same brief file. approved.json is built with jq -n from --arg values and --rawfile feedback, so the feedback is data. The office-hours mockup flow creates its brief and feedback files in one block. design-approval-flow now writes the feedback file (with quotes, backticks and $(...) in it) instead of substituting <FEEDBACK>, and checks the file is removed after the save. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: carve office-hours on every host to stay under the 160,000-byte ceiling With its brief, feedback and research prompts moved into files, the inlined office-hours render is 161,329 bytes on Codex and 160,302 on Factory, over SKILL_BYTE_CEILING, so gen-skill-docs --host all fails. Per the C4 policy (carve every skill whose external render exceeds the ceiling), office-hours joins ship and plan-ceo-review in CARVED_ON_EVERY_HOST. The Claude render is unchanged; external hosts read its sections relative to the installed skill. external-host-carve and qa-lazy-sections now include office-hours in their carved-skill lists. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * chore: regenerate after the free-text mktemp guard * fix(codex): prompts and review focus travel as agent-written files Co-authored-by: dgrant <dgrant@users.noreply.github.com> Challenge and consult prompts, and the /codex review focus text, are written by the agent into private .gstack/tmp files and fed to Codex on stdin or read back with cat; no heredoc carries them. Consult's new and resumed runs share one block (and one parser) so no placeholder stands in for the parser; a resumed run reads its session id from .context/codex-session-id and checks it. Adds the A3 group 2 placeholder allowlist block. * fix(benchmark-models): an inline prompt goes through a prompt file Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix(land-and-deploy): PR head repository and branch travel as data Co-authored-by: dgrant <dgrant@users.noreply.github.com> The branch check reads headRepository/headRefName from gh's JSON into quoted shell variables; ls-remote, the confirm-first delete and the local delete use those variables instead of model-substituted names. The postfail test now pins the variable form it replaces. * fix(office-hours): the reviewer receipt must match the receipt grammar before the check runs Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix(setup-gbrain): cleanup-orphans collects the PAT with read_secret_to_env Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix(ship): PR body drafts and rejected test paths are agent-written files Co-authored-by: dgrant <dgrant@users.noreply.github.com> The PR body is written into two private drafts (through ## Documentation, and from ## Test plan on); the scan block splices Step 14.5's section between them into the scanned temp file and deletes the drafts. Rejected test paths are read from a written list, and paths outside the repository are skipped. Tests that ran the old heredoc blocks now write the drafts and list first. * fix(sync-gbrain): orchestrator flags must match an allowlisted grammar Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: the gbrain page save and external-host placeholders pass the free-text rule The gbrain save block passed the whole page through a quoted heredoc inside --content; the agent now writes it to a mktemp file and gbrain put reads it on stdin. The Claude Code outside-voice runtime root and session mode, and the feature slug, join the allowlist with grammars; unused entries are dropped, and a test keeps every allowlisted placeholder tied to a real use. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: skill-specific Aside research queries travel as prompt files investigate, office-hours, plan-ceo-review, review, plan-eng-review, design-consultation Phase 2, the shared test bootstrap (B2) and QA's test bootstrap each carried their own _aside_exec call with the model-filled query inside a double-quoted argument. Skills that already carry the Web research block now give only the prompt-file text and use that block to create, write and send it; design-consultation, the shared bootstrap and QA create the file and send it with the new {{ASIDE_RESEARCH_SEND}} resolver, which is the same send block the research section uses (read-only sentence stays in the shell). aside-driver's B2 pin and qa-browser-preservation's bootstrap contract move from the inline call to the prompt file and the shared send resolver. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * fix: prose design commands pass briefs and feedback as data The DESIGN_SETUP command list taught `iterate --feedback "..."`, and design-consultation's late-change and plan-design-review's slop-regenerate lines showed `--feedback "<feedback>"` / `"..."`; design-shotgun's retry showed `generate --brief "<brief>"` and `evolve --brief "<brief>"`. Feedback now reads "$(cat "$FEEDBACK_FILE")" from an agent-written mktemp file (the setup note says how), and the shotgun retry reads the variant's own brief out of briefs.json with jq instead of retyping it. Co-authored-by: dgrant <dgrant@users.noreply.github.com> * test: record parity sizes after the free-text file mechanism; refresh ship goldens Measured union and skeleton sizes for ship, plan-eng-review, plan-design-review, office-hours, design-consultation, review, qa, browse and design-shotgun after every posting, prompt and brief site moved to agent-written mktemp files; the ship goldens pick up the question-tuning slug and the dropped free_text field. * test: parity caps for the combined wave head (rounded up past the measured ratio) * fix(plan-ceo-review): the mode handoff helper reminds the model to lead its chat with the line (#3038) Periodic run 37272185151's auto-decide-preserved red replays as the Step 0E handoff-line class: the model ran gstack-ceo-mode-handoff --auto, got the exact line, then opened its chat with a heading and a paraphrase, so the user never saw the auto-decide line (the terminal collapses tool output) and the detector saw no outcome. The helper now prints a one-line reminder on stderr beside the line; the detector accepts the line followed by that reminder and nothing else. The ship Greptile check test reads its template (test-of-test ratchet). * test: every troubleshooting link in shipped code resolves, relative or full URL * docs(todos): Oct 6 fix-wave follow-ups * v1.91.32.0 release notes: Codex works from zsh, free text never enters shell commands, honest failures * fix(plan-ceo-review): show the mode handoff line through a PostToolUse hook, not model compliance (#3038) The stderr reminder beside the handoff line measured 1/10 on auto-decide-preserved: the model still opened its chat with a paraphrase ('I've set review mode to HOLD SCOPE...'), so the user never saw the line and the detector saw no outcome. /plan-ceo-review now declares a Claude-only PostToolUse hook that shows the helper's exact line as a system message. Diagnostic measurement on this head: 10/10, the line visible in every trial while no trial's chat led with it. The reminder is reverted. * fix: type the handoff hook's line filter * fix: repairs from the integrated free suite - The self-locate markers looked like heredocs to the pipe-deadlock guard; they are plain comment fences now. - The spawned block keeps the canonical 'conservative non-destructive' phrase the cross-surface drift guard pins. - The free-text block's mktemp refusal says 'mktemp failed', as every /codex temp file must. - The batch transcript read reports sizes in MiB. - docs/TESTING_INTERNALS.md gets a real 'Ship measure' heading for its anchor. - The early-question harness admits /plan-ceo-review's handoff hook by its exact command (it only adds a system message); any other PostToolUse Bash hook is still refused. - Paid file count and the unconsumed-file example follow the new seeded case; ship goldens refreshed. - The CSO scanner test points Docker at a dead socket, so a machine with the pulled image does not scan. * test(cso): a host with no Docker binary is also a no-scan host The missing-image scan test failed on a Docker-less VM (docker is not installed in a trusted directory). Every accepted message is a reason the scan cannot run here; the not_assessed, helper-owned artifact assertions are unchanged. * ci: split the USERPROFILE assignment from export (SC2155) * fix(browse): Windows runtime roots read .source-path as a native path (#3026) The new windows-setup-e2e journey showed the installed browse could not start its server. setup runs under Git Bash and wrote .source-path as an MSYS path (/d/a/gstack); Node read it as \d\a\gstack on the current drive, found nothing, and fell back to the root's own bundle, which cannot import playwright. setup now writes cygpath -m, and the CLI maps an MSYS drive path for installs written by an older setup. The journey prints the daemon and startup logs when the chain fails. * fix(ship-measure): defects from the parallel #3034 plan review - A trial passes only with a JUnit testcase that executed and passed; a --case that selected nothing also exits 0. - measure runs a no-cost --list preflight that must plan the case's own file before any paid trial. - Diagnostic trials drop EVALS_CACHE_DIR/EVALS_CACHE_RUNTIME_ID, so they never reuse a stored judge pass or publish one the gate could reuse. - Gate once: when CI runs the gate on push, the push is the gate run; the skill no longer also starts eval:bg:pr (a second paid gate). - ubi-runner: down fails unless the VM is confirmed gone, and nothing sweeps other runners' VMs by default (UBI_GC_HOURS opt-in). * test(windows): the seeded-flake fixture drives a POSIX shell stub, so it stays off the Windows lane CreateProcess cannot exec the fixture's ./evals.sh shebang script (status null on windows-free-shard 6). The measure runner is covered on Windows by ship-measure.test.ts, which uses in-process fake runners. * fix: the free-text block refuses an unset temp dir instead of writing at / The rendered block ran mktemp "$_GT/..." with _GT unset in the portability test; on Linux that fails at /, but Git Bash on Windows can write to its install root, so the block went on (windows-free-shard 3). It now uses ${_GT:?}, so an unset directory always fails closed with the mktemp message. * test: pin the guarded ${_GT:?} temp-file form * test: the Greptile check's gh stub uses a portable shebang On Windows the absolute bash path has a space, so the stub never ran and every case read 'gh not logged in' (windows-free-shard 1). * test: the Greptile check runs with the native PATH and the gh stub first The PATH-only stub bin could not run its stub on Windows (every case read 'gh not logged in'). Stub cases now keep the real PATH with the stub first; only the no-gh case drops directories holding a real gh. --------- Co-authored-by: marcosdetry <marcosdetry@users.noreply.github.com> Co-authored-by: akshaykalia <akshaykalia@users.noreply.github.com> Co-authored-by: scottgigante-hubflow <scottgigante-hubflow@users.noreply.github.com> Co-authored-by: woodsonl <woodsonl@users.noreply.github.com> Co-authored-by: bezdekt <bezdekt@users.noreply.github.com> Co-authored-by: dgrant <dgrant@users.noreply.github.com> Co-authored-by: venkatareddyg <venkatareddyg@users.noreply.github.com> Co-authored-by: CharlesBerg85 <CharlesBerg85@users.noreply.github.com> Co-authored-by: VaibDeveloper <VaibDeveloper@users.noreply.github.com> Co-authored-by: Hagetino <Hagetino@users.noreply.github.com> Co-authored-by: Rekin226 <Rekin226@users.noreply.github.com> Co-authored-by: ekson73 <ekson73@users.noreply.github.com> Co-authored-by: tianxchen <tianxchen@users.noreply.github.com> Co-authored-by: BenjaminDSmithy <BenjaminDSmithy@users.noreply.github.com> Co-authored-by: ubay887 <ubay887@users.noreply.github.com> Co-authored-by: nareshh74 <nareshh74@users.noreply.github.com> Co-authored-by: syong <syong@users.noreply.github.com>
1171 lines
54 KiB
TypeScript
1171 lines
54 KiB
TypeScript
#!/usr/bin/env bun
|
|
/**
|
|
* Generate SKILL.md files from .tmpl templates.
|
|
*
|
|
* Pipeline:
|
|
* read .tmpl → find {{PLACEHOLDERS}} → resolve from source → format → write .md
|
|
*
|
|
* Supports --dry-run: generate to memory, exit 1 if different from committed file.
|
|
* Used by skill:check and CI freshness checks.
|
|
*/
|
|
|
|
import { discoverTemplates, discoverSectionTemplates, includesSkill } from './discover-skills';
|
|
import { externalSkillName, extractNameAndDescription } from './external-skill-names';
|
|
export { extractNameAndDescription } from './external-skill-names';
|
|
import { generateLlmsTxt } from './gen-llms-txt';
|
|
import { generateAgentsDigest, DIGEST_RELPATH, DIGEST_BYTE_BUDGET } from './gen-agents-digest';
|
|
import { generateDesignChecklistMd } from './resolvers/design-checklist';
|
|
import { DOM_DUMP_SCRIPT, DOM_DUMP_FILE } from '../lib/dom-dump-script';
|
|
import * as fs from 'fs';
|
|
import * as path from 'path';
|
|
import type { Host, TemplateContext } from './resolvers/types';
|
|
import { HOST_PATHS } from './resolvers/types';
|
|
import { RESOLVERS } from './resolvers/index';
|
|
import { rewriteCarvedSectionRefs, SKILL_BYTE_CEILING, usesLazySections } from './resolvers/sections';
|
|
import { insertRuntimePreludes } from './resolvers/runtime-root';
|
|
import { ALL_HOST_NAMES, resolveHostArg, getHostConfig } from '../hosts/index';
|
|
import type { HostConfig } from './host-config';
|
|
|
|
const ROOT = path.resolve(import.meta.dir, '..');
|
|
import { ALL_MODEL_NAMES, resolveModel, type Model } from './models';
|
|
import { resolveStateRoot } from '../lib/state-root';
|
|
import { mkdirpSync } from '../lib/fs-utils';
|
|
|
|
type HostArg = Host | 'all';
|
|
|
|
/** Internal render settings. Inputs always come from ROOT; output routing and
|
|
* content links are separate so checks can render canonical bytes into scratch. */
|
|
export interface GenerationOptions {
|
|
host?: HostArg;
|
|
dryRun?: boolean;
|
|
outputRoot?: string;
|
|
contentLinkRoot?: string | null;
|
|
/** Install-context render contract: absolute install root this render serves. */
|
|
installRoot?: string | null;
|
|
/** Skills this install leaves unregistered (gstack-config disabled_skills); the router omits them. */
|
|
disabledSkills?: string[];
|
|
model?: Model | null;
|
|
catalogMode?: 'trim' | 'full';
|
|
explainLevel?: 'default' | 'terse';
|
|
respectDetection?: boolean;
|
|
log?: (message: string) => void;
|
|
}
|
|
|
|
interface RenderOptions {
|
|
outputRoot: string;
|
|
contentLinkRoot: string | null;
|
|
installRoot: string | null;
|
|
disabledSkills: string[];
|
|
model: Model | null;
|
|
catalogMode: 'trim' | 'full';
|
|
explainLevel: 'default' | 'terse';
|
|
gbrainDetected: boolean;
|
|
}
|
|
|
|
export interface GeneratedArtifact {
|
|
relativePath: string;
|
|
kind: 'skill' | 'section' | 'metadata' | 'openclaw' | 'index' | 'digest' | 'asset';
|
|
host?: Host;
|
|
}
|
|
|
|
export interface GenerationDiagnostic {
|
|
kind: 'stale' | 'error' | 'warning' | 'skipped';
|
|
message: string;
|
|
host?: Host;
|
|
relativePath?: string;
|
|
}
|
|
|
|
export interface GenerationResult {
|
|
exitCode: 0 | 1;
|
|
artifacts: GeneratedArtifact[];
|
|
diagnostics: GenerationDiagnostic[];
|
|
}
|
|
|
|
/** Canonical generation never reads local detection state unless opted in. */
|
|
function loadGbrainOverride(respectDetection: boolean): boolean {
|
|
if (!respectDetection) return false;
|
|
const stateDir = resolveStateRoot();
|
|
try {
|
|
const json = JSON.parse(fs.readFileSync(path.join(stateDir, 'gbrain-detection.json'), 'utf-8'));
|
|
// Slow, remote, locked and briefly unreachable engines are still usable (#1964/#2051/#2456, A2).
|
|
return ['ok', 'timeout', 'db-unreachable', 'thin-client', 'engine-locked'].includes(json.gbrain_local_status ?? '');
|
|
} catch {
|
|
return false;
|
|
}
|
|
}
|
|
|
|
function effectiveSuppressedResolvers(hostConfig: HostConfig, options: RenderOptions): Set<string> {
|
|
let list = hostConfig.suppressedResolvers || [];
|
|
if (options.gbrainDetected) {
|
|
list = list.filter(r => r !== 'GBRAIN_CONTEXT_LOAD' && r !== 'GBRAIN_SAVE_RESULTS');
|
|
}
|
|
return new Set(list);
|
|
}
|
|
|
|
/** Parse CLI settings only when executing, never when imported by tests/checks. */
|
|
function parseGenerationArgs(args: string[]): GenerationOptions {
|
|
const value = (flag: string): string | undefined => {
|
|
const index = args.findIndex(arg => arg === flag || arg.startsWith(`${flag}=`));
|
|
if (index < 0) return undefined;
|
|
const arg = args[index];
|
|
const result = arg.startsWith(`${flag}=`) ? arg.slice(flag.length + 1) : args[index + 1];
|
|
if (!result || result.startsWith('--')) throw new Error(`${flag} requires a value`);
|
|
return result;
|
|
};
|
|
const hostValue = value('--host') ?? 'claude';
|
|
const host = hostValue === 'all' ? 'all' : resolveHostArg(hostValue) as Host;
|
|
const modelValue = value('--model');
|
|
const model = modelValue === undefined ? null : resolveModel(modelValue);
|
|
if (modelValue !== undefined && !model) {
|
|
throw new Error(`Unknown model: ${modelValue}. Use ${ALL_MODEL_NAMES.join(', ')}, or a family variant (e.g., claude-opus-4-7, gpt-5.4-mini, o3).`);
|
|
}
|
|
const catalogMode = value('--catalog-mode') ?? 'trim';
|
|
if (catalogMode !== 'trim' && catalogMode !== 'full') {
|
|
throw new Error(`Unknown catalog mode: ${catalogMode}. Use 'trim' (default) or 'full'.`);
|
|
}
|
|
const explainLevel = value('--explain-level') ?? 'default';
|
|
if (explainLevel !== 'default' && explainLevel !== 'terse') {
|
|
throw new Error(`Unknown explain level: ${explainLevel}. Use 'default' or 'terse'.`);
|
|
}
|
|
const outDir = value('--out-dir');
|
|
const linkRoot = value('--link-root');
|
|
const disabledValue = value('--disabled-skills');
|
|
const disabledSkills = (disabledValue ?? '').split(/[\s,]+/).filter(Boolean);
|
|
const badDisabled = disabledSkills.find(n => !/^\/?[a-z0-9-]+$/.test(n));
|
|
if (badDisabled) throw new Error(`--disabled-skills takes skill names (got ${JSON.stringify(badDisabled)})`);
|
|
const installRoot = value('--install-root');
|
|
if (installRoot !== undefined && !INSTALL_ROOT_PATTERN.test(installRoot)) {
|
|
throw new Error(`--install-root must be an absolute path of letters, digits and . _ - + @ / (got ${JSON.stringify(installRoot)})`);
|
|
}
|
|
// Swap-in callers use --link-root for the FINAL serving path (#2692).
|
|
// Direct --out-dir callers retain their existing links into the render.
|
|
return {
|
|
host, model, catalogMode, explainLevel,
|
|
dryRun: args.includes('--dry-run'),
|
|
respectDetection: args.includes('--respect-detection'),
|
|
outputRoot: outDir === undefined ? ROOT : path.resolve(outDir),
|
|
contentLinkRoot: linkRoot !== undefined ? path.resolve(linkRoot)
|
|
: outDir !== undefined ? path.resolve(outDir) : null,
|
|
installRoot: installRoot ?? null,
|
|
disabledSkills,
|
|
};
|
|
}
|
|
|
|
/** Install roots are spliced into shell lines unquoted, so only plain absolute paths. */
|
|
const INSTALL_ROOT_PATTERN = /^\/[A-Za-z0-9_.@+\/-]*$/;
|
|
|
|
/**
|
|
* Install-context render contract (docs/ADDING_A_HOST.md): a per-install render
|
|
* names its own install root instead of the host's default global root. Null
|
|
* keeps the committed bytes.
|
|
*/
|
|
function rewriteInstallRoot(content: string, hostConfig: HostConfig, installRoot: string | null): string {
|
|
if (!installRoot) return content;
|
|
const root = installRoot.replace(/\/+$/, '');
|
|
const defaults = hostConfig.usesEnvVars
|
|
? [`$HOME/${hostConfig.globalRoot}`, `~/${hostConfig.globalRoot}`]
|
|
: ['$HOME/.claude/skills/gstack', '~/.claude/skills/gstack'];
|
|
return defaults.reduce((text, from) => text.split(from).join(root), content);
|
|
}
|
|
|
|
/**
|
|
* C8 (#3018): the router names every skill, so it routed to skills the user
|
|
* disabled (it also said "When in doubt, invoke the skill" until v1.91.31). A per-install render that
|
|
* knows its disabled skills drops their routing rules (keeping the other side
|
|
* of an "A or B" rule) and says which skills are off. The router itself and
|
|
* the way back (gstack-upgrade) can never be disabled.
|
|
*/
|
|
function omitDisabledSkills(content: string, disabled: string[]): string {
|
|
const off = new Set(disabled.map(n => n.replace(/^\//, '')).filter(n => n !== 'gstack' && n !== 'gstack-upgrade').map(n => n.replace(/^gstack-/, '')));
|
|
if (off.size === 0) return content;
|
|
const isOff = (ref: string) => off.has(ref.replace(/^`\/(?:gstack-)?|`$/g, ''));
|
|
const routed = content.split('\n').flatMap(line => {
|
|
const m = line.match(/^(- .*→ invoke )(`\/[a-z0-9-]+`(?: or `\/[a-z0-9-]+`)*)(.*)$/);
|
|
if (!m) return [line];
|
|
const kept = m[2].split(' or ').filter(ref => !isOff(ref));
|
|
return kept.length ? [`${m[1]}${kept.join(' or ')}${m[3]}`] : [];
|
|
}).join('\n');
|
|
const names = [...off].map(n => `\`/${n}\``).join(', ');
|
|
return routed.replace('\n## Route first\n', `\n## Route first\n\nDisabled on this install (never invoke or suggest them): ${names}.\n`);
|
|
}
|
|
|
|
/** Repoint only Claude section links, retaining global bin/browse/doc paths. */
|
|
function rewriteSectionBase(content: string, linkRoot: string | null): string {
|
|
if (!linkRoot) return content;
|
|
// Callback replacement preserves literal $ sequences in configured paths.
|
|
return content.replace(
|
|
/~\/\.claude\/skills\/gstack\/([^\s)`"'*]+\/sections\/)/g,
|
|
(_m, p1: string) => `${linkRoot}/${p1}`,
|
|
);
|
|
}
|
|
|
|
// HostPaths, HOST_PATHS, and TemplateContext imported from ./resolvers/types (line 7-8)
|
|
// Design constants (AI_SLOP_BLACKLIST, OPENAI_HARD_REJECTIONS, OPENAI_LITMUS_CHECKS)
|
|
// live in ./resolvers/constants and are consumed by resolvers directly.
|
|
|
|
// ─── External Host Helpers ───────────────────────────────────
|
|
|
|
// ─── Voice Trigger Processing ────────────────────────────────
|
|
|
|
/**
|
|
* Extract voice-triggers YAML list from frontmatter.
|
|
* Returns an array of trigger strings, or [] if no voice-triggers field.
|
|
*/
|
|
function extractVoiceTriggers(content: string): string[] {
|
|
const fmStart = content.indexOf('---\n');
|
|
if (fmStart !== 0) return [];
|
|
const fmEnd = content.indexOf('\n---', fmStart + 4);
|
|
if (fmEnd === -1) return [];
|
|
const frontmatter = content.slice(fmStart + 4, fmEnd);
|
|
|
|
const triggers: string[] = [];
|
|
let inVoice = false;
|
|
for (const line of frontmatter.split('\n')) {
|
|
if (/^voice-triggers:/.test(line)) { inVoice = true; continue; }
|
|
if (inVoice) {
|
|
const m = line.match(/^\s+-\s+"(.+)"$/);
|
|
if (m) triggers.push(m[1]);
|
|
else if (!/^\s/.test(line)) break;
|
|
}
|
|
}
|
|
return triggers;
|
|
}
|
|
|
|
/**
|
|
* Preprocess voice triggers: fold voice-triggers YAML field into description,
|
|
* then strip the field from frontmatter. Must run BEFORE transformFrontmatter
|
|
* and extractNameAndDescription so all hosts see the updated description.
|
|
*/
|
|
function processVoiceTriggers(content: string): string {
|
|
const triggers = extractVoiceTriggers(content);
|
|
if (triggers.length === 0) return content;
|
|
|
|
// Strip voice-triggers block from frontmatter
|
|
content = content.replace(/^voice-triggers:\n(?:\s+-\s+"[^"]*"\n?)*/m, '');
|
|
|
|
// Get current description (after stripping voice-triggers, so it's clean)
|
|
const { description } = extractNameAndDescription(content);
|
|
if (!description) return content;
|
|
|
|
// Build new description with voice triggers appended
|
|
const voiceLine = `Voice triggers (speech-to-text aliases): ${triggers.map(t => `"${t}"`).join(', ')}.`;
|
|
const newDescription = description + '\n' + voiceLine;
|
|
|
|
// Replace old indented description with new in frontmatter
|
|
const oldIndented = description.split('\n').map(l => ` ${l}`).join('\n');
|
|
const newIndented = newDescription.split('\n').map(l => ` ${l}`).join('\n');
|
|
content = content.replace(oldIndented, newIndented);
|
|
|
|
return content;
|
|
}
|
|
|
|
// Export for testing
|
|
export { extractVoiceTriggers, processVoiceTriggers };
|
|
|
|
// ─── Catalog Trim (v1.45.0.0 T4) ─────────────────────────────
|
|
//
|
|
// Frontmatter `description:` blocks today pack: a one-line outcome, "Use when
|
|
// asked to..." voice triggers, "Proactively..." routing guidance, and a
|
|
// "(gstack)" tag. This pile is the always-loaded catalog surface — every
|
|
// session pays for the full text. The catalog trim splits the description
|
|
// into a one-line catalog entry (lead sentence + "(gstack)") that stays in
|
|
// the frontmatter, and a "## When to invoke" body section that holds the
|
|
// routing/voice triggers prose for in-skill discovery.
|
|
//
|
|
// Opt-out: `--catalog-mode=full` keeps v1.44 behavior (no trim, full
|
|
// description in frontmatter). Use when debugging routing regressions or
|
|
// when shipping skills to hosts that depend on the legacy fat catalog.
|
|
|
|
export interface CatalogParts {
|
|
lead: string; // First sentence — kept in catalog
|
|
routingProse: string; // "Use when asked to...", "Proactively..." paragraphs
|
|
voiceLine: string | null; // "Voice triggers (speech-to-text aliases): ..." line if present
|
|
hasGstackTag: boolean;
|
|
}
|
|
|
|
export function splitCatalogDescription(description: string): CatalogParts {
|
|
// Voice triggers line (folded in by processVoiceTriggers earlier)
|
|
const voiceMatch = description.match(/Voice triggers \(speech-to-text aliases\):[^\n]+/);
|
|
const voiceLine = voiceMatch ? voiceMatch[0] : null;
|
|
let working = voiceLine ? description.replace(voiceLine, '').trim() : description.trim();
|
|
|
|
const hasGstackTag = /\(gstack\)/.test(working);
|
|
if (hasGstackTag) working = working.replace(/\(gstack\)/, '').trim();
|
|
|
|
// Lead = first sentence, ending at the first `.`/`!`/`?` that is followed by
|
|
// whitespace or end-of-text. Terminator chars NOT followed by whitespace/end
|
|
// (embedded periods in "TODOS.md", URLs, "v1.45.0.0") are consumed by the
|
|
// second alternative `[.!?](?!\s|$)` and do NOT end the sentence. The two
|
|
// alternatives are disjoint character classes, so there is no ambiguity and
|
|
// no catastrophic-backtracking risk. If no terminator-followed-by-boundary
|
|
// exists at all, we fall back to a 20-word cut below.
|
|
// First normalize to single-line for sentence detection, then back out.
|
|
const collapsed = working.replace(/\s+/g, ' ').trim();
|
|
const sentenceMatch = collapsed.match(/^((?:[^.!?]|[.!?](?!\s|$))*[.!?])(?:\s|$)/);
|
|
// sentenceLead is the FULL first sentence (no truncation). We compute routing
|
|
// from this position, then optionally truncate the displayed lead afterwards.
|
|
// Truncating first then computing routing was the v1.45.0.0 bug — when the
|
|
// first sentence exceeded 200 chars, the routing extraction would lose the
|
|
// entire tail of the description (design-consultation's "Use when..."
|
|
// routing prose silently dropped).
|
|
const sentenceLead = sentenceMatch ? sentenceMatch[1].trim() : collapsed.split(/\s/).slice(0, 20).join(' ');
|
|
|
|
// Routing prose: everything AFTER the first sentence boundary in the collapsed view.
|
|
const leadInCollapsed = collapsed.indexOf(sentenceLead);
|
|
const routingCollapsed = leadInCollapsed >= 0
|
|
? collapsed.slice(leadInCollapsed + sentenceLead.length).trim()
|
|
: '';
|
|
|
|
// Now produce the displayed lead — truncated if too long. The original
|
|
// sentenceLead is preserved for routing extraction below.
|
|
let lead = sentenceLead;
|
|
if (lead.length > 200) {
|
|
const trunc = lead.slice(0, 197);
|
|
const lastSpace = trunc.lastIndexOf(' ');
|
|
lead = (lastSpace > 60 ? trunc.slice(0, lastSpace) : trunc) + '...';
|
|
}
|
|
// Restore line breaks for routing prose by mapping back to original layout.
|
|
// Use original whitespace structure where possible; fall back to collapsed.
|
|
// Anchor recovery on sentenceLead (the untruncated first sentence) — not
|
|
// `lead` (which may have a "..." suffix and won't substring-match `working`).
|
|
let routingProse = routingCollapsed;
|
|
const collapsedLeadIdx = working.replace(/\s+/g, ' ').indexOf(sentenceLead);
|
|
if (collapsedLeadIdx >= 0) {
|
|
let consumed = 0;
|
|
let cut = 0;
|
|
for (let i = 0; i < working.length && consumed < collapsedLeadIdx + sentenceLead.length; i++) {
|
|
if (/\s/.test(working[i])) {
|
|
if (i === 0 || /\s/.test(working[i - 1])) continue;
|
|
consumed += 1;
|
|
} else {
|
|
consumed += 1;
|
|
}
|
|
cut = i + 1;
|
|
}
|
|
const tail = working.slice(cut).trim();
|
|
if (tail.length > 0) routingProse = tail;
|
|
}
|
|
|
|
return { lead, routingProse, voiceLine, hasGstackTag };
|
|
}
|
|
|
|
/** Build the catalog-trimmed `description:` block. */
|
|
export function buildTrimmedDescription(parts: CatalogParts): string {
|
|
const lead = parts.lead.trim();
|
|
const suffix = parts.hasGstackTag ? ' (gstack)' : '';
|
|
return `${lead}${suffix}`;
|
|
}
|
|
|
|
/** Build the body section that holds the routing/voice prose. */
|
|
export function buildWhenToInvokeSection(parts: CatalogParts): string {
|
|
const lines: string[] = ['## When to invoke this skill', ''];
|
|
if (parts.routingProse) {
|
|
lines.push(parts.routingProse);
|
|
lines.push('');
|
|
}
|
|
if (parts.voiceLine) {
|
|
lines.push(parts.voiceLine);
|
|
lines.push('');
|
|
}
|
|
return lines.join('\n');
|
|
}
|
|
|
|
/**
|
|
* Render a string as a YAML inline scalar value (the text after `key: `),
|
|
* quoting only when a plain scalar would be invalid or ambiguous.
|
|
*
|
|
* The bug this guards (#1778): a description like "Ship workflow: detect..."
|
|
* emitted as a plain scalar has an interior ": " that a strict YAML parser
|
|
* (Codex/OpenAI skill loading) reads as a nested mapping and rejects with
|
|
* "mapping values are not allowed in this context". When quoting is needed we
|
|
* fall back to JSON.stringify, which produces a double-quoted scalar that YAML
|
|
* accepts verbatim (YAML is a superset of JSON for flow scalars). Strings that
|
|
* are already valid plain scalars pass through unchanged to keep regen diffs small.
|
|
*/
|
|
export function toYamlInlineScalar(s: string): string {
|
|
const needsQuote =
|
|
s.length === 0 ||
|
|
s !== s.trim() || // leading/trailing whitespace
|
|
/:(\s|$)/.test(s) || // "foo: bar" / trailing colon → mapping ambiguity
|
|
/\s#/.test(s) || // " #" → inline comment
|
|
/\.\.\./.test(s) || // "..." → document-end marker; strict parsers reject mid-scalar (catalog-trim truncation appends it)
|
|
/^[\s>|&*!%@`"'#,\[\]{}?-]/.test(s); // leading YAML indicator char
|
|
return needsQuote ? JSON.stringify(s) : s;
|
|
}
|
|
|
|
/**
|
|
* Apply catalog trim to a SKILL.md body:
|
|
* - shorten frontmatter `description:` to lead + (gstack)
|
|
* - insert "## When to invoke" body section AFTER the generated header
|
|
* (so it lands near the top of body content, where routing guidance
|
|
* belongs)
|
|
*
|
|
* Returns the rewritten content plus the extracted parts.
|
|
*/
|
|
export function applyCatalogTrim(content: string, skillName: string): { content: string; parts: CatalogParts } | null {
|
|
// Locate description block in frontmatter
|
|
if (!content.startsWith('---\n')) return null;
|
|
const fmEnd = content.indexOf('\n---', 4);
|
|
if (fmEnd === -1) return null;
|
|
const frontmatter = content.slice(4, fmEnd);
|
|
|
|
// Match `description: |` block + indented body lines
|
|
const descMatch = frontmatter.match(/^description:\s*\|?\s*\n((?:\s{2,}.*(?:\n|$))+)/m)
|
|
|| frontmatter.match(/^description:\s+(.+)$/m);
|
|
if (!descMatch) return null;
|
|
|
|
// Extract full description text
|
|
let descText: string;
|
|
if (descMatch[0].startsWith('description: |') || /^description:\s*\|/.test(descMatch[0])) {
|
|
descText = descMatch[1].split('\n').map(l => l.replace(/^\s{2}/, '')).join('\n').trim();
|
|
} else {
|
|
descText = descMatch[1].trim();
|
|
}
|
|
|
|
// Skip skills with very short descriptions (already trimmed or no routing prose).
|
|
// Below ~120 chars, splitting adds no value.
|
|
if (descText.length < 120) return null;
|
|
|
|
const parts = splitCatalogDescription(descText);
|
|
// If lead + (gstack) is already most of the text, no trim needed.
|
|
const trimmedLen = buildTrimmedDescription(parts).length;
|
|
if (trimmedLen >= descText.length - 20) return null;
|
|
|
|
// Replace description in frontmatter — keep trailing newline so the next
|
|
// YAML field doesn't collide on the same line as the description value.
|
|
// Quote the value when it would be an invalid YAML plain scalar (the common
|
|
// case: an interior ": " like "Ship workflow: detect..." which a strict YAML
|
|
// parser reads as a nested mapping and rejects — #1778). toYamlInlineScalar
|
|
// only quotes when needed, so descriptions without special chars stay plain.
|
|
const newDesc = buildTrimmedDescription(parts);
|
|
// Function replacer (not a string) so a `$` in the description — e.g. a future
|
|
// skill referencing `$B`/`$D` — can't be interpreted as a `$&`/`$1` replacement
|
|
// pattern and silently corrupt the frontmatter.
|
|
const newDescLine = `description: ${toYamlInlineScalar(newDesc)}\n`;
|
|
const newFrontmatter = frontmatter.replace(descMatch[0], () => newDescLine);
|
|
let newContent = '---\n' + newFrontmatter + content.slice(fmEnd);
|
|
|
|
// Insert body section after frontmatter (after the closing ---\n and any
|
|
// existing GENERATED header). We insert before the first non-comment line.
|
|
const bodyStart = newContent.indexOf('\n---\n') + 5;
|
|
const whenToInvoke = '\n' + buildWhenToInvokeSection(parts).trim() + '\n';
|
|
// Skip past the generated header if present (it lives after frontmatter close)
|
|
const headerMatch = newContent.slice(bodyStart).match(/^(<!--[^>]*-->\s*\n)+/);
|
|
const insertAt = bodyStart + (headerMatch ? headerMatch[0].length : 0);
|
|
newContent = newContent.slice(0, insertAt) + whenToInvoke + '\n' + newContent.slice(insertAt);
|
|
|
|
return { content: newContent, parts };
|
|
}
|
|
|
|
const OPENAI_SHORT_DESCRIPTION_LIMIT = 120;
|
|
|
|
function condenseOpenAIShortDescription(description: string): string {
|
|
const firstParagraph = description.split(/\n\s*\n/)[0] || description;
|
|
const collapsed = firstParagraph.replace(/\s+/g, ' ').trim();
|
|
if (collapsed.length <= OPENAI_SHORT_DESCRIPTION_LIMIT) return collapsed;
|
|
|
|
const truncated = collapsed.slice(0, OPENAI_SHORT_DESCRIPTION_LIMIT - 3);
|
|
const lastSpace = truncated.lastIndexOf(' ');
|
|
const safe = lastSpace > 40 ? truncated.slice(0, lastSpace) : truncated;
|
|
return `${safe}...`;
|
|
}
|
|
|
|
function generateOpenAIYaml(displayName: string, shortDescription: string): string {
|
|
return `interface:
|
|
display_name: ${JSON.stringify(displayName)}
|
|
short_description: ${JSON.stringify(shortDescription)}
|
|
default_prompt: ${JSON.stringify(`Use ${displayName} for this task.`)}
|
|
policy:
|
|
allow_implicit_invocation: true
|
|
`;
|
|
}
|
|
|
|
/**
|
|
* Transform frontmatter for external hosts.
|
|
* Claude: strips `sensitive:` field (only Factory uses it).
|
|
* Codex: keeps name + description only, enforces 1024-char limit.
|
|
* Factory: keeps name + description + user-invocable, conditionally adds disable-model-invocation.
|
|
*/
|
|
function transformFrontmatter(content: string, host: Host): string {
|
|
const hostConfig = getHostConfig(host);
|
|
const fm = hostConfig.frontmatter;
|
|
|
|
if (fm.mode === 'denylist') {
|
|
// Denylist mode: strip listed fields, keep everything else
|
|
for (const field of fm.stripFields || []) {
|
|
if (field === 'voice-triggers') {
|
|
content = content.replace(/^voice-triggers:\n(?:\s+-\s+"[^"]*"\n?)*/m, '');
|
|
} else {
|
|
content = content.replace(new RegExp(`^${field}:\\s*.*\\n`, 'm'), '');
|
|
}
|
|
}
|
|
return content;
|
|
}
|
|
|
|
// Allowlist mode: reconstruct frontmatter with only allowed fields
|
|
const fmStart = content.indexOf('---\n');
|
|
if (fmStart !== 0) return content;
|
|
const fmEnd = content.indexOf('\n---', fmStart + 4);
|
|
if (fmEnd === -1) return content;
|
|
const frontmatter = content.slice(fmStart + 4, fmEnd);
|
|
const body = content.slice(fmEnd + 4);
|
|
const { name, description } = extractNameAndDescription(content);
|
|
|
|
// Description limit enforcement
|
|
if (fm.descriptionLimit) {
|
|
const behavior = fm.descriptionLimitBehavior || 'error';
|
|
if (description.length > fm.descriptionLimit) {
|
|
if (behavior === 'error') {
|
|
throw new Error(
|
|
`${hostConfig.displayName} description for "${name}" is ${description.length} chars (max ${fm.descriptionLimit}). ` +
|
|
`Compress the description in the .tmpl file.`
|
|
);
|
|
} else if (behavior === 'warn') {
|
|
console.warn(`WARNING: ${hostConfig.displayName} description for "${name}" exceeds ${fm.descriptionLimit} chars`);
|
|
}
|
|
// 'truncate' — silently proceed
|
|
}
|
|
}
|
|
|
|
// Build frontmatter with allowed fields
|
|
const indentedDesc = description.split('\n').map(l => ` ${l}`).join('\n');
|
|
const fmName = fm.nameMatchesDirectory && name !== 'gstack' && !name.startsWith('gstack-') ? `gstack-${name}` : name;
|
|
let newFm = `---\nname: ${fmName}\ndescription: |\n${indentedDesc}\n`;
|
|
|
|
// Add extra fields (host-wide)
|
|
if (fm.extraFields) {
|
|
for (const [key, value] of Object.entries(fm.extraFields)) {
|
|
if (key !== 'name' && key !== 'description') {
|
|
newFm += `${key}: ${value}\n`;
|
|
}
|
|
}
|
|
}
|
|
|
|
// Add conditional fields
|
|
if (fm.conditionalFields) {
|
|
for (const rule of fm.conditionalFields) {
|
|
const match = Object.entries(rule.if).every(([k, v]) =>
|
|
new RegExp(`^${k}:\\s*${v}`, 'm').test(frontmatter)
|
|
);
|
|
if (match) {
|
|
for (const [key, value] of Object.entries(rule.add)) {
|
|
newFm += `${key}: ${value}\n`;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// Preserve additional keepFields beyond name and description
|
|
if (fm.keepFields) {
|
|
for (const field of fm.keepFields) {
|
|
if (field === 'name' || field === 'description') continue;
|
|
// Match YAML field with possible multi-line/array value (indented lines after colon)
|
|
const fieldMatch = frontmatter.match(new RegExp(`^${field}:(.*(?:\\n(?:[ \\t]+.+))*)`, 'm'));
|
|
if (fieldMatch) {
|
|
newFm += `${field}:${fieldMatch[1]}\n`;
|
|
}
|
|
}
|
|
}
|
|
|
|
// Rename fields (copy values from template frontmatter with new keys)
|
|
if (fm.renameFields) {
|
|
for (const [oldName, newName] of Object.entries(fm.renameFields)) {
|
|
const fieldMatch = frontmatter.match(new RegExp(`^${oldName}:(.+(?:\\n(?:\\s+.+)*)?)`, 'm'));
|
|
if (fieldMatch) {
|
|
newFm += `${newName}:${fieldMatch[1]}\n`;
|
|
}
|
|
}
|
|
}
|
|
|
|
newFm += '---';
|
|
return newFm + body;
|
|
}
|
|
|
|
/**
|
|
* Extract hook descriptions from frontmatter for inline safety prose.
|
|
* Returns a description of what the hooks do, or null if no hooks.
|
|
*/
|
|
function extractHookSafetyProse(tmplContent: string, hostConfig: HostConfig): string | null {
|
|
if (!tmplContent.match(/^hooks:/m)) return null;
|
|
if (hostConfig.capabilities.safetyHooks === 'enforced') return null;
|
|
|
|
// Parse the hook matchers to build a human-readable safety description
|
|
const matchers: string[] = [];
|
|
const matcherRegex = /matcher:\s*"(\w+)"/g;
|
|
let m;
|
|
while ((m = matcherRegex.exec(tmplContent)) !== null) {
|
|
if (!matchers.includes(m[1])) matchers.push(m[1]);
|
|
}
|
|
|
|
if (matchers.length === 0) return null;
|
|
|
|
// Build safety prose based on what tools are hooked
|
|
const toolDescriptions: Record<string, string> = {
|
|
Bash: 'check bash commands for destructive operations (rm -rf, DROP TABLE, force-push, git reset --hard, etc.) before execution',
|
|
Edit: 'verify file edits are within the allowed scope boundary before applying',
|
|
Write: 'verify file writes are within the allowed scope boundary before applying',
|
|
};
|
|
|
|
const safetyChecks = matchers
|
|
.map(t => toolDescriptions[t] || `check ${t} operations for safety`)
|
|
.join(', and ');
|
|
|
|
return `> **Safety Advisory — not enforced on ${hostConfig.displayName}:** advisory, not blocked. ${hostConfig.displayName} runs no gstack safety hooks, so nothing stops a command automatically. On Claude Code this skill's hooks ${safetyChecks}; here, do those checks yourself: always pause and verify before executing potentially destructive operations. If uncertain about a command's safety, ask the user for confirmation before proceeding.`;
|
|
}
|
|
|
|
// ─── External Host Config (now derived from hosts/*.ts) ──────
|
|
// EXTERNAL_HOST_CONFIG replaced by getHostConfig() from hosts/index.ts
|
|
|
|
// ─── Template Processing ────────────────────────────────────
|
|
|
|
const GENERATED_HEADER = `<!-- AUTO-GENERATED from {{SOURCE}} — do not edit directly -->\n<!-- Regenerate: bun run gen:skill-docs -->\n`;
|
|
|
|
/**
|
|
* Apply a host's configured path + tool rewrites. Extracted so both SKILL.md
|
|
* (via processExternalHost) and section files (via processSectionTemplate) get
|
|
* identical per-host treatment — a section's cross-references must rewrite the
|
|
* same way the parent skill's do, or external hosts get wrong paths.
|
|
*/
|
|
function applyHostRewrites(content: string, hostConfig: HostConfig): string {
|
|
let result = content;
|
|
for (const rewrite of hostConfig.pathRewrites) {
|
|
result = result.replaceAll(rewrite.from, rewrite.to);
|
|
}
|
|
if (hostConfig.toolRewrites) {
|
|
for (const [from, to] of Object.entries(hostConfig.toolRewrites)) {
|
|
result = result.replaceAll(from, to);
|
|
}
|
|
}
|
|
return result;
|
|
}
|
|
|
|
/**
|
|
* Resolve {{PLACEHOLDER}} / {{NAME:arg}} tokens against the RESOLVERS registry,
|
|
* honoring host suppression and appliesTo gating, then assert nothing is left
|
|
* unresolved. Extracted so SKILL.md and section templates resolve through the
|
|
* exact same path — a security/sanitization fix to one can't miss the other.
|
|
*/
|
|
/**
|
|
* A second {{PREAMBLE}} in one template re-expands the entire ~12K-token
|
|
* preamble mid-document (#2508/#2362 — a PROSE mention of the macro in
|
|
* spec/SKILL.md.tmpl expanded it a second time, +43KB per /spec load).
|
|
* Resolution is context-blind, so any second occurrence — code fence, prose,
|
|
* anywhere — is a generation error, never intentional. Throw at render time
|
|
* so the mistake cannot reach a generated SKILL.md again.
|
|
*/
|
|
export function assertSinglePreamble(tmplContent: string, relTmplPath: string): void {
|
|
const count = (tmplContent.match(/\{\{PREAMBLE\}\}/g) || []).length;
|
|
if (count > 1) {
|
|
throw new Error(
|
|
`${relTmplPath} contains {{PREAMBLE}} ${count} times — a template may reference it `
|
|
+ `at most once (each occurrence expands the full preamble; see #2508/#2362). `
|
|
+ `Refer to "the preamble" in prose instead of the macro.`,
|
|
);
|
|
}
|
|
}
|
|
|
|
function resolvePlaceholders(
|
|
tmplContent: string,
|
|
ctx: TemplateContext,
|
|
hostConfig: HostConfig,
|
|
relTmplPath: string,
|
|
options: RenderOptions,
|
|
): string {
|
|
assertSinglePreamble(tmplContent, relTmplPath);
|
|
// effectiveSuppressedResolvers() honors --respect-detection: when gbrain is
|
|
// detected locally, GBRAIN_* resolvers un-suppress. Shared by SKILL.md and
|
|
// section generation so both paths get the same gbrain-aware behavior.
|
|
const suppressed = effectiveSuppressedResolvers(hostConfig, options);
|
|
const onePass = (input: string): string =>
|
|
input.replace(/\{\{(\w+(?::[^}]+)?)\}\}/g, (_match, fullKey) => {
|
|
const parts = fullKey.split(':');
|
|
const resolverName = parts[0];
|
|
const args = parts.slice(1);
|
|
if (suppressed.has(resolverName)) return '';
|
|
const resolve = RESOLVERS[resolverName];
|
|
if (!resolve) throw new Error(`Unknown placeholder {{${resolverName}}} in ${relTmplPath}`);
|
|
return args.length > 0 ? resolve(ctx, args) : resolve(ctx);
|
|
});
|
|
|
|
// Multi-pass: a resolver may emit content that itself contains {{TOKENS}} — the
|
|
// {{SECTION:id}} resolver inlines a section template (with its own resolvers)
|
|
// for non-Claude hosts. .replace() doesn't re-scan inserted text, so loop until
|
|
// the output stabilizes. Bounded to avoid an infinite loop if a resolver ever
|
|
// emits its own placeholder; 6 passes is far more nesting than any skill needs.
|
|
let content = tmplContent;
|
|
for (let pass = 0; pass < 6; pass++) {
|
|
const next = onePass(content);
|
|
if (next === content) break;
|
|
content = next;
|
|
}
|
|
|
|
const remaining = content.match(/\{\{(\w+(?::[^}]+)?)\}\}/g);
|
|
if (remaining) {
|
|
throw new Error(`Unresolved placeholders in ${relTmplPath}: ${remaining.join(', ')}`);
|
|
}
|
|
return content;
|
|
}
|
|
|
|
/**
|
|
* Build the TemplateContext from a template's frontmatter. Shared by SKILL.md
|
|
* and section generation so sections inherit the SAME context the parent skill
|
|
* resolves with (skillName, tier, benefitsFrom, interactive) — enforced by
|
|
* test/template-context-parity.test.ts. skillNameOverride lets section
|
|
* generation pin the parent skill's name instead of deriving "sections".
|
|
*/
|
|
function buildContext(
|
|
tmplContent: string,
|
|
tmplPath: string,
|
|
host: Host,
|
|
options: RenderOptions,
|
|
skillNameOverride?: string,
|
|
): TemplateContext {
|
|
const { name: extractedName } = extractNameAndDescription(tmplContent);
|
|
const skillName = skillNameOverride || extractedName || path.basename(path.dirname(tmplPath));
|
|
const benefitsMatch = tmplContent.match(/^benefits-from:\s*\[([^\]]*)\]/m);
|
|
const benefitsFrom = benefitsMatch
|
|
? benefitsMatch[1].split(',').map(s => s.trim()).filter(Boolean)
|
|
: undefined;
|
|
const tierMatch = tmplContent.match(/^preamble-tier:\s*(\d+)$/m);
|
|
const preambleTier = tierMatch ? parseInt(tierMatch[1], 10) : undefined;
|
|
const interactiveMatch = tmplContent.match(/^interactive:\s*(true|false)\s*$/m);
|
|
const interactive = interactiveMatch ? interactiveMatch[1] === 'true' : undefined;
|
|
return {
|
|
skillName, tmplPath, benefitsFrom, host, paths: HOST_PATHS[host],
|
|
preambleTier, model: options.model ?? getHostConfig(host).defaultModel, interactive, explainLevel: options.explainLevel, installRoot: options.installRoot,
|
|
};
|
|
}
|
|
|
|
/**
|
|
* Process external host output: routing, frontmatter, path rewrites, metadata.
|
|
* Shared between Codex and Factory (and future external hosts).
|
|
*/
|
|
function processExternalHost(
|
|
content: string,
|
|
tmplContent: string,
|
|
host: Host,
|
|
skillDir: string,
|
|
extractedDescription: string,
|
|
ctx: TemplateContext,
|
|
options: RenderOptions,
|
|
frontmatterName?: string,
|
|
): { content: string; outputPath: string; symlinkLoop: boolean; metadata?: { outputPath: string; content: string } } {
|
|
const hostConfig = getHostConfig(host);
|
|
|
|
const name = externalSkillName(skillDir === '.' ? '' : skillDir, frontmatterName);
|
|
// --out-dir mirrors the host tree (outputs only; inputs read from ROOT).
|
|
const outputDir = path.join(options.outputRoot, hostConfig.hostSubdir, 'skills', name);
|
|
const outputPath = path.join(outputDir, 'SKILL.md');
|
|
|
|
// Guard against symlink loops
|
|
let symlinkLoop = false;
|
|
const claudePath = ctx.tmplPath.replace(/\.tmpl$/, '');
|
|
try {
|
|
const resolvedClaude = fs.realpathSync(claudePath);
|
|
const resolvedExternal = path.join(fs.realpathSync(path.dirname(outputPath)), path.basename(outputPath));
|
|
if (resolvedClaude === resolvedExternal) {
|
|
symlinkLoop = true;
|
|
}
|
|
} catch {
|
|
// realpathSync fails if file doesn't exist yet — no symlink loop
|
|
}
|
|
|
|
// Extract hook safety prose BEFORE transforming frontmatter (which strips hooks)
|
|
const safetyProse = extractHookSafetyProse(tmplContent, hostConfig);
|
|
|
|
// Transform frontmatter (host-aware)
|
|
let result = transformFrontmatter(content, host);
|
|
|
|
// Insert safety advisory at the top of the body (after frontmatter)
|
|
if (safetyProse) {
|
|
const bodyStart = result.indexOf('\n---') + 4;
|
|
result = result.slice(0, bodyStart) + '\n' + safetyProse + '\n' + result.slice(bodyStart);
|
|
}
|
|
|
|
// Config-driven path + tool rewrites (shared with processSectionTemplate so
|
|
// section cross-references get the same per-host treatment as SKILL.md).
|
|
result = applyHostRewrites(result, hostConfig);
|
|
|
|
// Config-driven: generate metadata (e.g., openai.yaml for Codex)
|
|
const metadata = hostConfig.generation.generateMetadata && !symlinkLoop ? {
|
|
outputPath: path.join(outputDir, 'agents', 'openai.yaml'),
|
|
content: generateOpenAIYaml(name, condenseOpenAIShortDescription(extractedDescription)),
|
|
} : undefined;
|
|
|
|
return { content: result, outputPath, symlinkLoop, metadata };
|
|
}
|
|
|
|
function processTemplate(tmplPath: string, host: Host, options: RenderOptions): { outputPath: string; content: string; symlinkLoop?: boolean; metadata?: { outputPath: string; content: string } } {
|
|
// Normalize to LF at the entry point. Templates may have CRLF on disk when
|
|
// checked out on Windows with core.autocrlf=true. Downstream regexes
|
|
// (processVoiceTriggers, transformFrontmatter) hardcode \n, so without
|
|
// normalization they silently no-op on CRLF — producing different output
|
|
// than CI (Linux, LF) and breaking the Skill Docs Freshness check.
|
|
// (catalogParts left the return type with the proactive-suggestions
|
|
// retirement — merge of the two v1.64 waves.)
|
|
const tmplContent = fs.readFileSync(tmplPath, 'utf-8').replace(/\r\n/g, '\n');
|
|
const relTmplPath = path.relative(ROOT, tmplPath);
|
|
let outputPath = tmplPath.replace(/\.tmpl$/, '');
|
|
|
|
// Determine skill directory relative to ROOT
|
|
const skillDir = path.relative(ROOT, path.dirname(tmplPath));
|
|
|
|
// --out-dir: mirror the skill tree into the out-dir instead of writing in
|
|
// place (external hosts compute their own output paths below).
|
|
if (host === 'claude') {
|
|
outputPath = path.join(options.outputRoot, skillDir, path.basename(tmplPath).replace(/\.tmpl$/, ''));
|
|
}
|
|
|
|
// Extract name/description: name drives external skill naming + setup symlinks
|
|
// (and TemplateContext.skillName via buildContext); description feeds external
|
|
// host metadata. When frontmatter name: differs from directory name (e.g.
|
|
// run-tests/ with name: test), the frontmatter name wins.
|
|
const { name: extractedName, description: extractedDescription } = extractNameAndDescription(tmplContent);
|
|
|
|
const currentHostConfig = getHostConfig(host);
|
|
const ctx = buildContext(tmplContent, tmplPath, host, options);
|
|
const skillName = ctx.skillName;
|
|
|
|
// Replace placeholders + assert none remain (shared path with section generation).
|
|
let content = resolvePlaceholders(tmplContent, ctx, currentHostConfig, relTmplPath, options);
|
|
|
|
// Preprocess voice triggers: fold into description, strip field from frontmatter.
|
|
// Must run BEFORE transformFrontmatter so all hosts see the updated description,
|
|
// and BEFORE extractedDescription is used by external host metadata.
|
|
content = processVoiceTriggers(content);
|
|
|
|
// Re-extract description AFTER voice trigger preprocessing so Codex openai.yaml
|
|
// metadata gets the updated description with voice triggers included.
|
|
const postProcessDescription = extractNameAndDescription(content).description;
|
|
|
|
// For Claude: strip sensitive: field (only Factory uses it)
|
|
// For external hosts: route output, transform frontmatter, rewrite paths
|
|
let symlinkLoop = false;
|
|
let metadata: { outputPath: string; content: string } | undefined;
|
|
if (host === 'claude') {
|
|
content = transformFrontmatter(content, host);
|
|
} else {
|
|
const result = processExternalHost(content, tmplContent, host, skillDir, postProcessDescription, ctx, options, extractedName || undefined);
|
|
content = result.content;
|
|
outputPath = result.outputPath;
|
|
symlinkLoop = result.symlinkLoop;
|
|
metadata = result.metadata;
|
|
}
|
|
|
|
// Prepend generated header (after frontmatter)
|
|
const header = GENERATED_HEADER.replace('{{SOURCE}}', path.basename(tmplPath));
|
|
const fmEnd = content.indexOf('---', content.indexOf('---') + 3);
|
|
if (fmEnd !== -1) {
|
|
const insertAt = content.indexOf('\n', fmEnd) + 1;
|
|
content = content.slice(0, insertAt) + header + content.slice(insertAt);
|
|
} else {
|
|
content = header + content;
|
|
}
|
|
|
|
// Catalog trim (Claude only — external hosts have their own frontmatter shapes)
|
|
if (host === 'claude' && options.catalogMode === 'trim') {
|
|
const trimmed = applyCatalogTrim(content, skillName);
|
|
if (trimmed) content = trimmed.content;
|
|
}
|
|
|
|
// --out-dir: repoint section-base paths to the out-dir (no-op otherwise).
|
|
if (host === 'claude') content = rewriteSectionBase(content, options.contentLinkRoot);
|
|
if (skillDir === '' || skillDir === '.') content = omitDisabledSkills(content, options.disabledSkills);
|
|
content = rewriteInstallRoot(insertRuntimePreludes(rewriteCarvedSectionRefs(content, ctx), ctx), currentHostConfig, options.installRoot);
|
|
|
|
return { outputPath, content, symlinkLoop, metadata };
|
|
}
|
|
|
|
/**
|
|
* Generate one on-demand section file (`<skill>/sections/<name>.md.tmpl` →
|
|
* `<name>.md`). Sections are BODY FRAGMENTS — no frontmatter, no catalog trim,
|
|
* no voice triggers. They resolve placeholders through the SAME path as
|
|
* SKILL.md (resolvePlaceholders) using the PARENT skill's TemplateContext
|
|
* (so appliesTo gating + tier behave identically — a section's {{PREAMBLE}}-
|
|
* style resolver renders the same content it would in the parent, not empty).
|
|
*
|
|
* Output routing mirrors SKILL.md: Claude writes in-tree at
|
|
* `<skill>/sections/<name>.md`; external hosts write to
|
|
* `<hostSubdir>/skills/<externalName>/sections/<name>.md`. External hosts get
|
|
* applyHostRewrites so cross-references resolve per host.
|
|
*/
|
|
function processSectionTemplate(
|
|
sectionTmplPath: string,
|
|
skillDir: string,
|
|
host: Host,
|
|
options: RenderOptions,
|
|
): { outputPath: string; content: string } {
|
|
const tmplContent = fs.readFileSync(sectionTmplPath, 'utf-8');
|
|
const relTmplPath = path.relative(ROOT, sectionTmplPath);
|
|
const hostConfig = getHostConfig(host);
|
|
|
|
// Read the owning SKILL.md.tmpl so the section inherits the parent's name +
|
|
// tier + benefits-from (TemplateContext parity). Fall back to the dir name.
|
|
const parentTmplPath = path.join(ROOT, skillDir, 'SKILL.md.tmpl');
|
|
const parentContent = fs.existsSync(parentTmplPath) ? fs.readFileSync(parentTmplPath, 'utf-8') : '';
|
|
const parentName = (parentContent && extractNameAndDescription(parentContent).name) || skillDir;
|
|
const ctx = buildContext(parentContent || tmplContent, parentTmplPath, host, options, parentName);
|
|
|
|
// Resolve placeholders against the section body (shared guard catches stragglers).
|
|
let content = resolvePlaceholders(tmplContent, ctx, hostConfig, relTmplPath, options);
|
|
|
|
// External hosts: rewrite cross-reference paths/tools (no frontmatter to transform).
|
|
if (host !== 'claude') {
|
|
content = applyHostRewrites(content, hostConfig);
|
|
} else {
|
|
// --out-dir: a section may cross-reference another section by absolute path;
|
|
// repoint those to the out-dir too (no-op when --out-dir is unset).
|
|
content = rewriteSectionBase(content, options.contentLinkRoot);
|
|
}
|
|
content = rewriteInstallRoot(insertRuntimePreludes(rewriteCarvedSectionRefs(content, ctx), ctx), hostConfig, options.installRoot);
|
|
|
|
// Plain generated header (no frontmatter to insert after).
|
|
content = GENERATED_HEADER.replace('{{SOURCE}}', path.basename(sectionTmplPath)) + content;
|
|
|
|
const fileName = path.basename(sectionTmplPath).replace(/\.tmpl$/, '');
|
|
let outputPath: string;
|
|
if (host === 'claude') {
|
|
outputPath = path.join(options.outputRoot, skillDir, 'sections', fileName);
|
|
} else {
|
|
const externalName = externalSkillName(skillDir, parentName);
|
|
outputPath = path.join(options.outputRoot, hostConfig.hostSubdir, 'skills', externalName, 'sections', fileName);
|
|
}
|
|
return { outputPath, content };
|
|
}
|
|
|
|
// ─── Main ───────────────────────────────────────────────────
|
|
|
|
/** Render each artifact once. Artifact writes go through emit(); stale external
|
|
* caches are pruned only after a successful host render. Dry runs use the same
|
|
* inventory as normal generation, including metadata and
|
|
* shared outputs. Options are per invocation so imports/concurrent runs cannot
|
|
* inherit another caller's model, detection, or output paths.
|
|
*
|
|
* templates + host settings -> render -> emit -> dry-run: compare only
|
|
* | -> normal: mkdir + write
|
|
* shared index/digest ---------------+ -> artifact inventory + diagnostics
|
|
* successful external host -----------------> normal only: prune retired caches
|
|
*/
|
|
export async function runGeneration(settings: GenerationOptions = {}): Promise<GenerationResult> {
|
|
const options: RenderOptions = {
|
|
outputRoot: path.resolve(settings.outputRoot ?? ROOT),
|
|
contentLinkRoot: settings.contentLinkRoot ?? null,
|
|
installRoot: settings.installRoot ?? null,
|
|
disabledSkills: settings.disabledSkills ?? [],
|
|
model: settings.model ?? null,
|
|
catalogMode: settings.catalogMode ?? 'trim',
|
|
explainLevel: settings.explainLevel ?? 'default',
|
|
gbrainDetected: loadGbrainOverride(settings.respectDetection ?? false),
|
|
};
|
|
const hosts = settings.host === 'all' ? ALL_HOST_NAMES as Host[] : [settings.host ?? 'claude'];
|
|
const log = settings.log ?? (() => {});
|
|
const artifacts: GeneratedArtifact[] = [];
|
|
const diagnostics: GenerationDiagnostic[] = [];
|
|
const templates = discoverTemplates(ROOT);
|
|
const sections = discoverSectionTemplates(ROOT);
|
|
const rel = (outputPath: string) => path.relative(options.outputRoot, outputPath).split(path.sep).join('/');
|
|
|
|
function emit(outputPath: string, content: string, kind: GeneratedArtifact['kind'], host?: Host): void {
|
|
const relativePath = rel(outputPath);
|
|
artifacts.push({ relativePath, kind, ...(host ? { host } : {}) });
|
|
try {
|
|
if (settings.dryRun) {
|
|
let existing: string | undefined;
|
|
try {
|
|
existing = fs.readFileSync(outputPath, 'utf-8');
|
|
} catch (error) {
|
|
if ((error as NodeJS.ErrnoException).code !== 'ENOENT') throw error;
|
|
// Windows reports ENOENT for a child of a regular file. Distinguish
|
|
// that filesystem error from a missing artifact without writing.
|
|
let parent = path.dirname(outputPath);
|
|
while (true) {
|
|
try {
|
|
if (!fs.statSync(parent).isDirectory()) {
|
|
throw Object.assign(new Error(`ENOTDIR: output ancestor is not a directory: ${parent}`, { cause: error }), { code: 'ENOTDIR' });
|
|
}
|
|
break;
|
|
} catch (ancestorError) {
|
|
if ((ancestorError as NodeJS.ErrnoException).code !== 'ENOENT') throw ancestorError;
|
|
const next = path.dirname(parent);
|
|
if (next === parent) throw error;
|
|
parent = next;
|
|
}
|
|
}
|
|
}
|
|
if (existing !== content) {
|
|
diagnostics.push({ kind: 'stale', relativePath, host, message: `STALE: ${relativePath}` });
|
|
log(`STALE: ${relativePath}`);
|
|
} else {
|
|
log(`FRESH: ${relativePath}`);
|
|
}
|
|
} else {
|
|
mkdirpSync(path.dirname(outputPath));
|
|
fs.writeFileSync(outputPath, content);
|
|
log(`GENERATED: ${relativePath}`);
|
|
}
|
|
} catch (error) {
|
|
throw Object.assign(new Error(`${relativePath}: ${(error as Error).message}`), { relativePath });
|
|
}
|
|
}
|
|
|
|
function failed(error: unknown, host?: Host): void {
|
|
const message = error instanceof Error ? error.message : String(error);
|
|
diagnostics.push({ kind: 'error', host, message, relativePath: (error as { relativePath?: string })?.relativePath });
|
|
}
|
|
|
|
for (const host of hosts) {
|
|
try {
|
|
const hostConfig = getHostConfig(host);
|
|
const tokenBudget: Array<{ skill: string; lines: number; tokens: number }> = [];
|
|
const renderedNames = new Set<string>();
|
|
for (const template of templates) {
|
|
const skillDir = path.dirname(template.tmpl);
|
|
if (!includesSkill(hostConfig, skillDir)) continue;
|
|
const result = processTemplate(path.join(ROOT, template.tmpl), host, options);
|
|
const relativePath = rel(result.outputPath);
|
|
if (host !== 'claude') renderedNames.add(path.basename(path.dirname(result.outputPath)));
|
|
if (result.symlinkLoop) {
|
|
diagnostics.push({ kind: 'skipped', relativePath, host, message: `SKIPPED (symlink loop): ${relativePath}` });
|
|
log(`SKIPPED (symlink loop): ${relativePath}`);
|
|
continue;
|
|
}
|
|
emit(result.outputPath, result.content, 'skill', host);
|
|
if (result.metadata) emit(result.metadata.outputPath, result.metadata.content, 'metadata', host);
|
|
if (skillDir === 'qa') {
|
|
const report = fs.readFileSync(path.join(ROOT, 'qa', 'templates', 'functional-report-template.md'), 'utf-8');
|
|
emit(path.join(path.dirname(result.outputPath), 'templates', 'functional-report-template.md'),
|
|
(host === 'claude' ? '' : GENERATED_HEADER.replace('{{SOURCE}}', 'qa/templates/functional-report-template.md')) + report, 'asset', host);
|
|
}
|
|
tokenBudget.push({ skill: relativePath, lines: result.content.split('\n').length, tokens: Math.round(result.content.length / 4) });
|
|
const bytes = Buffer.byteLength(result.content, 'utf8');
|
|
if (bytes > SKILL_BYTE_CEILING) {
|
|
// Setup renders per install (--install-root): an oversized skill must never leave a user with none.
|
|
const message = `${host}/${path.basename(path.dirname(result.outputPath))}/SKILL.md is ${bytes} bytes, over the ${SKILL_BYTE_CEILING.toLocaleString('en-US')}-byte limit. Fix: carve sections with usesLazySections() for this skill.`;
|
|
diagnostics.push({ kind: options.installRoot ? 'warning' : 'error', host, relativePath, message });
|
|
}
|
|
}
|
|
|
|
for (const section of sections) {
|
|
if (!includesSkill(hostConfig, section.skillDir) || !usesLazySections(host, section.skillDir)) continue;
|
|
const result = processSectionTemplate(path.join(ROOT, section.tmpl), section.skillDir, host, options);
|
|
emit(result.outputPath, result.content, 'section', host);
|
|
tokenBudget.push({ skill: rel(result.outputPath), lines: result.content.split('\n').length, tokens: Math.round(result.content.length / 4) });
|
|
}
|
|
|
|
// Claude owns these catalog-derived runtime assets. Use the same host
|
|
// inclusion rule and compare-or-write path as every other artifact.
|
|
if (host === 'claude' && includesSkill(hostConfig, 'review')) {
|
|
emit(path.join(options.outputRoot, 'review', 'design-checklist.md'),
|
|
generateDesignChecklistMd(), 'asset', host);
|
|
emit(path.join(options.outputRoot, DOM_DUMP_FILE), DOM_DUMP_SCRIPT + '\n', 'asset', host);
|
|
}
|
|
|
|
if (host === 'openclaw') {
|
|
for (const variant of ['lite', 'full', 'plan'] as const) {
|
|
const fileName = `gstack-${variant}-CLAUDE.md`;
|
|
emit(path.join(options.outputRoot, 'openclaw', fileName),
|
|
fs.readFileSync(path.join(ROOT, 'openclaw', 'templates', fileName), 'utf-8'), 'openclaw', host);
|
|
}
|
|
}
|
|
|
|
// A failed render exits this try before pruning: its inventory is partial.
|
|
// Only remove generated directories owned by this host; sidecars and user
|
|
// skills survive. Dry runs never create, rewrite, or remove any directory.
|
|
if (!settings.dryRun && host !== 'claude') {
|
|
const skillsRoot = path.join(options.outputRoot, hostConfig.hostSubdir, 'skills');
|
|
let entries: fs.Dirent[] = [];
|
|
try {
|
|
entries = fs.readdirSync(skillsRoot, { withFileTypes: true });
|
|
} catch (error) {
|
|
if ((error as NodeJS.ErrnoException).code !== 'ENOENT') throw error;
|
|
}
|
|
for (const entry of entries) {
|
|
if (entry.isSymbolicLink() || !entry.isDirectory() || !entry.name.startsWith('gstack-') || renderedNames.has(entry.name)) continue;
|
|
// Keep the old render usable until setup has migrated installed copies/links.
|
|
if (entry.name === 'gstack-claude' && process.env.GSTACK_DEFER_CLAUDE_RENAME_PRUNE === '1') continue;
|
|
let generated = false;
|
|
try {
|
|
generated = fs.readFileSync(path.join(skillsRoot, entry.name, 'SKILL.md'), 'utf-8').includes('<!-- AUTO-GENERATED from');
|
|
} catch (error) {
|
|
if ((error as NodeJS.ErrnoException).code !== 'ENOENT') throw error;
|
|
}
|
|
if (!generated) {
|
|
log(` kept ${host} skills/${entry.name}: not a gstack render (no generated banner)`);
|
|
continue;
|
|
}
|
|
fs.rmSync(path.join(skillsRoot, entry.name), { recursive: true, force: true });
|
|
log(` pruned stale ${host} render: ${entry.name}`);
|
|
if (entry.name === 'gstack-claude') {
|
|
log(' /claude is now /claude-code. Run ./setup to migrate installed skill links; generation only updates render files.');
|
|
}
|
|
}
|
|
}
|
|
|
|
if (!settings.dryRun && tokenBudget.length > 0) {
|
|
tokenBudget.sort((a, b) => b.lines - a.lines);
|
|
log(`\nToken Budget (${host} host)`);
|
|
log('═'.repeat(60));
|
|
for (const item of tokenBudget) {
|
|
const name = item.skill.replace(/\/SKILL\.md$/, '').replace(`${hostConfig.hostSubdir}/skills/`, '');
|
|
log(` ${name.padEnd(30)} ${String(item.lines).padStart(5)} lines ~${String(item.tokens).padStart(6)} tokens`);
|
|
}
|
|
log('─'.repeat(60));
|
|
log(` ${'TOTAL'.padEnd(30)} ${String(tokenBudget.reduce((sum, t) => sum + t.lines, 0)).padStart(5)} lines ~${String(tokenBudget.reduce((sum, t) => sum + t.tokens, 0)).padStart(6)} tokens\n`);
|
|
}
|
|
} catch (error) {
|
|
failed(error, host);
|
|
}
|
|
}
|
|
|
|
// Shared artifacts are awaited in both modes. A failure must reach the CLI
|
|
// exit code, never disappear in a fire-and-forget auxiliary writer.
|
|
try {
|
|
const index = await generateLlmsTxt();
|
|
emit(path.join(options.outputRoot, 'gstack', 'llms.txt'), index.content, 'index');
|
|
for (const warning of index.warnings) diagnostics.push({ kind: 'warning', message: `[gen-llms-txt] ${warning}` });
|
|
} catch (error) {
|
|
failed(error);
|
|
}
|
|
try {
|
|
const digest = generateAgentsDigest();
|
|
emit(path.join(options.outputRoot, DIGEST_RELPATH), digest.content, 'digest');
|
|
if (!settings.dryRun) log(`[gen-agents-digest] ${DIGEST_RELPATH}: ${digest.bytes} bytes (budget ${DIGEST_BYTE_BUDGET})`);
|
|
} catch (error) {
|
|
failed(error);
|
|
}
|
|
|
|
return { exitCode: diagnostics.some(d => d.kind === 'error' || d.kind === 'stale') ? 1 : 0, artifacts, diagnostics };
|
|
}
|
|
|
|
/** Importing this module never executes generation or reads CLI/user settings.
|
|
* Async main is require()-compatible: only top-level await would break callers. */
|
|
export async function main(args = process.argv.slice(2)): Promise<number> {
|
|
try {
|
|
const settings = parseGenerationArgs(args);
|
|
const result = await runGeneration({ ...settings, log: console.log });
|
|
for (const diagnostic of result.diagnostics) {
|
|
if (diagnostic.kind === 'error') console.error(`ERROR${diagnostic.host ? ` (${diagnostic.host})` : ''}: ${diagnostic.message}`);
|
|
if (diagnostic.kind === 'warning') console.error(diagnostic.message);
|
|
}
|
|
if (result.diagnostics.some(d => d.kind === 'stale')) {
|
|
console.error(`\nGenerated files are stale. Run: bun run gen:skill-docs --host ${settings.host ?? 'claude'}`);
|
|
}
|
|
if (!settings.dryRun) {
|
|
try {
|
|
const config = fs.readFileSync(path.join(resolveStateRoot(), 'config.yaml'), 'utf-8');
|
|
if (/^skill_prefix:\s*true/m.test(config)) {
|
|
console.log('\nNote: skill_prefix is true. Run gstack-relink to re-apply name: patches (it patches both the install and any active gbrain render).');
|
|
}
|
|
} catch { /* optional local install note */ }
|
|
}
|
|
return result.exitCode;
|
|
} catch (error) {
|
|
console.error(`ERROR: ${(error as Error).message}`);
|
|
return 1;
|
|
}
|
|
}
|
|
|
|
if (import.meta.main) {
|
|
void main().then(code => { process.exitCode = code; });
|
|
}
|