stoll/gstack <- garrytan/gstack#3
Merged
Merged
Conversation
* fix: remove auth token from /health, secure extension bootstrap (CRITICAL-02 + HIGH-03) - Remove token from /health response (was leaked to any localhost process) - Write .auth.json to extension dir for Manifest V3 bootstrap - sidebar-agent reads token from state file via BROWSE_STATE_FILE env var - Remove getToken handler from extension (token via health broadcast) - Extension loads token before first health poll to prevent race condition * fix: require auth on cookie-picker data routes (CRITICAL-01) - Add Bearer token auth gate on all /cookie-picker/* data/action routes - GET /cookie-picker HTML page stays unauthenticated (UI shell) - Token embedded in served HTML for picker's fetch calls - CORS preflight now allows Authorization header * fix: add state file TTL and plaintext cookie warning (HIGH-02) - Add savedAt timestamp to state save output - Warn on load if state file older than 7 days - Auto-delete stale state files (>7 days) on server startup - Warning about plaintext cookie storage in save message * fix: innerHTML XSS in extension content script and sidepanel (MEDIUM-01) - content.js: replace innerHTML with createElement/textContent for ref panel - sidepanel.js: escape entry.command with escapeHtml() in activity feed - Both found by security audit + Codex adversarial red team * fix: symlink bypass in validateReadPath (MEDIUM-02) - Always resolve to absolute path first (fixes relative path bypass) - Use realpathSync to follow symlinks before boundary check - Throw on non-ENOENT realpathSync failures (explicit over silent) - Resolve SAFE_DIRECTORIES through realpathSync (macOS /tmp → /private/tmp) - Resolve directory part for non-existent files (ENOENT with symlinked parent) * fix: freeze hook symlink bypass and prefix collision (MEDIUM-03) - Add POSIX-portable path resolution (cd + pwd -P, works on macOS) - Fix prefix collision: /project-evil no longer matches /project freeze dir - Use trailing slash in boundary check to require directory boundary * fix: shell script injection in gstack-config and telemetry (MEDIUM-04) - gstack-config: validate keys (alphanumeric+underscore only) - gstack-config: use grep -F (fixed string) instead of -E (regex) - gstack-config: escape sed special chars in values, drop newlines - gstack-telemetry-log: sanitize REPO_SLUG and BRANCH via json_safe() * test: 20 security tests for audit remediation - server-auth: verify token removed from /health, auth on /refs, /activity/* - cookie-picker: auth required on data routes, HTML page unauthenticated - path-validation: symlink bypass blocked, realpathSync failure throws - gstack-config: regex key rejected, sed special chars preserved - state-ttl: savedAt timestamp, 7-day TTL warning - telemetry: branch/repo with quotes don't corrupt JSON - adversarial: sidepanel escapes entry.command, freeze prefix collision * chore: bump version and changelog (v0.13.1.0) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * docs: tone down changelog — defense in depth, not catastrophic bugs Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
#603) * feat: user sovereignty — AI models recommend, users decide When Claude and Codex agree on a scope change, they now present it to the user instead of auto-incorporating it. Adds User Sovereignty as the third core principle in ETHOS.md. Fixes the cross-model tension template in review.ts to present both perspectives neutrally instead of judging. Adds User Challenge category to autoplan with proper contract updates (intro, important rules, audit trail, gate handling). Adds Outside Voice Integration Rule to CEO and eng review templates. * chore: regenerate SKILL.md files from updated templates * chore: bump version and changelog (v0.13.2.0) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: proper gstack description in openai.yaml + block Codex from rewriting it Codex kept overwriting agents/openai.yaml with a browse-only description. Two fixes: (1) better description covering full PM/dev/eng/CEO/QA scope, (2) add agents/ to the filesystem boundary so Codex stops modifying it. * chore: regenerate SKILL.md files with updated filesystem boundary --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(security): commit bun.lock to pin dependency versions Remove bun.lock from .gitignore and commit the lockfile. Every bun install now uses exact pinned versions instead of resolving floating ^ ranges from npm fresh. Closes the supply-chain vector from #566. Co-Authored-By: boinger <boinger@users.noreply.github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: gstack-slug falls back to dirname/unknown when git context is absent Add || true to git commands and fallback defaults so gstack-slug works outside git repos. Prevents unbound variable crash that kills every review skill when no git context exists. Co-Authored-By: collinstraka-clov <collinstraka-clov@users.noreply.github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: setup auto-selects default after 10s timeout to prevent CI hangs Add -t 10 to the read command in the skill-prefix prompt. In CI, Docker, and Conductor workspaces where a TTY exists but nobody is watching, the prompt now auto-selects short names after 10 seconds instead of blocking forever. Co-Authored-By: stedfn <stedfn@users.noreply.github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: browse CLI Windows lockfile — use string flag instead of numeric constants Bun compiled binaries on Windows don't handle numeric fs.constants correctly. The string flag 'wx' is semantically identical to O_CREAT | O_EXCL | O_WRONLY per Node docs and works on all platforms. Fixes #599 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: add ~/.gstack/projects/ to plan file search path /office-hours writes design docs to ~/.gstack/projects/$SLUG/ but /ship and /review only searched ~/.claude/plans, ~/.codex/plans, and .gstack/plans. Add the project-scoped directory as the first search location so plan validation finds design docs created by the standard workflow. Fixes #591 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: autoplan dual-voice — sequential foreground execution instead of broken parallel Background subagents don't inherit tool permissions in Claude Code, so the Claude subagent in dual-voice mode was silently failing on every invocation. Every autoplan run was degrading to single-reviewer mode without warning. Change all three phases (CEO, Design, Eng) from "simultaneously" to sequential foreground execution: Claude subagent first (Agent tool, foreground), then Codex (Bash). Both complete before the consensus table. Fixes #497 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: regenerate SKILL.md files from updated templates Regenerated from autoplan/SKILL.md.tmpl (dual-voice fix) and scripts/resolvers/review.ts (plan search path fix). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add community PR guardrails — protect ETHOS.md and voice Add explicit CLAUDE.md rule requiring AskUserQuestion before accepting any community PR that touches ETHOS.md, removes promotional material, or changes Garry's voice. No exceptions, no auto-merging. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.13.2.0) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: gen-skill-docs detects symlink loop, skips codex write that overwrites Claude SKILL.md When .agents/skills/gstack is symlinked to the repo root (vendored dev mode), gen-skill-docs --host codex was writing the Codex-transformed SKILL.md through the symlink, overwriting the Claude version. This caused SKILL.md and agents/openai.yaml to silently revert to Codex paths after every build. Now detects when the codex output path resolves to the same real file as the Claude output and skips the write. Content is still generated for token budget tracking. The openai.yaml write is also skipped for the same symlink case. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: resolve all 7 test failures — version sync, zsh glob guard, symlink-aware codex tests 1. package.json version synced with VERSION file (0.13.3.0) 2. design-shotgun/SKILL.md.tmpl: added setopt +o nomatch guard to bash block with variant-*.png glob 3. Codex generation tests: skip skills where .agents/skills/{name} is a symlink back to repo root (vendored dev mode). These can't have proper codex content since gen-skill-docs skips the write to avoid overwriting the Claude SKILL.md. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: boinger <boinger@users.noreply.github.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: collinstraka-clov <collinstraka-clov@users.noreply.github.com> Co-authored-by: stedfn <stedfn@users.noreply.github.com>
* fix: sidebar prompt injection defense — XML framing, command allowlist, arg plumbing Three security fixes for the Chrome sidebar: 1. XML-framed prompts with trust boundaries and escape of < > & in user messages to prevent tag injection attacks. 2. Bash command allowlist in system prompt — only browse binary commands ($B goto, $B click, etc.) allowed. All other bash commands forbidden. 3. Fix sidebar-agent.ts ignoring queued args — server-side --model and --allowedTools changes were silently dropped because the agent rebuilt args from scratch instead of using the queue entry. Also defaults sidebar to Opus (harder to manipulate). 12 new tests covering XML escaping, command allowlist, Opus default, trust boundary instructions, and arg plumbing. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.13.4.0) ML prompt injection defense design doc + P0 TODO for follow-up PR. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: clear stale worktree and claude session on sidebar reconnect loadSession() was restoring worktreePath and claudeSessionId from prior crashes. The worktree directory no longer existed (deleted on cleanup) and --resume with a dead session ID caused claude to fail silently. Now validates worktree exists on load and clears stale claude session IDs on every server restart. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…and Factory (v0.13.5.0) (#621) * refactor: extract processExternalHost() shared helper for multi-host generation Refactor the Codex-specific output routing block in gen-skill-docs.ts into a shared processExternalHost() function. Both Codex and future external hosts (Factory Droid) will use this helper for output routing, symlink loop detection, frontmatter transformation, path rewrites, and metadata generation. - Rename codexSkillName() to externalSkillName() everywhere - Extract ExternalHostConfig interface with per-host settings - Codex output is byte-identical (verified via --dry-run) - Skip /codex skill for all non-Claude hosts (not just codex) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: add Factory Droid host type, preamble, and co-author trailer - Add 'factory' to Host union type with .factory/skills/gstack paths - Extend preamble runtime root detection for Factory ($HOME/.factory/) - Add GSTACK_DESIGN env var to preamble (was missing for Codex too) - Add Factory Droid co-author trailer for git commits Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: Factory Droid generation, --host all, and host-aware frontmatter - Add --host factory (alias: --host droid) to gen-skill-docs - Add --host all: generates for claude, codex, and factory in one invocation with fault-tolerant per-host error handling (only fails if claude fails) - Factory frontmatter: name + description + user-invocable: true - Factory sensitive skills: disable-model-invocation: true (from sensitive: field) - Claude: strips sensitive: field from output (only Factory uses it) - Factory tool name translation: Claude tool names → generic phrasing - Replace chained gen:skill-docs calls with --host all in package.json build Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: sensitive frontmatter for Factory Droid auto-invocation safety Add sensitive: true to 6 skill templates with side effects that Factory Droids shouldn't auto-invoke (ship, land-and-deploy, guard, careful, freeze, unfreeze). The field is: - Factory: emitted as disable-model-invocation: true - Claude/Codex: stripped from output by transformFrontmatter() Also fix Claude host path: call transformFrontmatter() for Claude to strip the sensitive: field from Claude output. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: gstack-platform-detect binary for multi-host debugging Bash script that prints a table of installed AI coding agents (Claude, Codex, Factory Droid, Kiro) with versions, skill paths, and gstack installation status. Useful for debugging multi-host setups. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: Factory Droid support in setup script - Add factory to --host values (auto-detected via command -v droid) - Add .factory/ skill doc generation step alongside .agents/ - Add create_factory_runtime_root() and link_factory_skill_dirs() helpers mirroring the Codex equivalents - Factory install section creates ~/.factory/skills/ with symlinks Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: Factory Droid awareness in skill-check and uninstall - skill-check.ts: add Factory skills validation and freshness check - gstack-uninstall: add Factory artifact cleanup (~/.factory/skills/gstack* and per-project .factory/ sidecar) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * test: Factory Droid generation + --host all test suites Add 13 new tests: - Factory output paths, frontmatter (user-invocable, disable-model-invocation) - Sensitive vs non-sensitive skill classification - Path rewrites (no .claude/skills/ in Factory output) - /codex skill exclusion, openai.yaml absence - Factory keeps Codex integration blocks (for second opinions) - --host droid alias, --dry-run freshness, preamble paths - --host all generates for all 3 hosts - Setup script host validation updated for factory Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: Factory Droid install instructions + CI freshness check - README: add Factory Droid section with install instructions and restart note (Factory requires restart to rescan skills) - CI: add Factory skill doc freshness verification to skill-docs.yml Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: generated Factory Droid skill output (.factory/skills/) 29 skills generated for Factory Droid with: - user-invocable: true on all skills - disable-model-invocation: true on 6 sensitive skills - .factory/skills/ paths (no .claude/skills/ references) - $GSTACK_ROOT env vars for runtime root detection - Tool name translation (Claude tool names → generic phrasing) Committed to git for CI freshness checks and direct consumption. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: add Factory Droid P1 TODO for browse MCP server Add 3 TODOs under new ## Factory Droid section: - P1: Browse MCP server (Option B, deeper Factory integration) - P3: .agent/skills/ dual output for cross-agent compatibility - P3: Custom Droid definitions alongside skills Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.13.5.0) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore: gitignore .factory and remove tracked files The .factory/ directory contains generated skill definitions for Factory Droid compatibility. These should not be tracked in git, same as .claude/skills/ and .agents/. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * chore: bump version and changelog (v0.13.5.1) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
….4.0) (#622) * feat: learnings + confidence resolvers — cross-skill memory infrastructure Three new resolvers for the self-learning system: - LEARNINGS_SEARCH: tells skills to load prior learnings before analysis - LEARNINGS_LOG: tells skills to capture discoveries after completing work - CONFIDENCE_CALIBRATION: adds 1-10 confidence scoring to all review findings Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: learnings bin scripts — append-only JSONL read/write gstack-learnings-log: validates JSON, auto-injects timestamp, appends to ~/.gstack/projects/$SLUG/learnings.jsonl. Append-only (no mutation). gstack-learnings-search: reads/filters/dedupes learnings with confidence decay (observed/inferred lose 1pt/30d), cross-project discovery, and "latest winner" resolution per key+type. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: learnings count in preamble output Every skill now prints "LEARNINGS: N entries loaded" during preamble, making the compounding loop visible to the user. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate learnings + confidence into 9 skill templates Add {{LEARNINGS_SEARCH}}, {{LEARNINGS_LOG}}, and {{CONFIDENCE_CALIBRATION}} placeholders to review, ship, plan-eng-review, plan-ceo-review, office-hours, investigate, retro, and cso templates. Regenerated all SKILL.md files. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: /learn skill — manage project learnings New skill for reviewing, searching, pruning, and exporting what gstack has learned across sessions. Commands: /learn, /learn search, /learn prune, /learn export, /learn stats, /learn add. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: self-learning roadmap — 5-release design doc Covers: R1 GStack Learns (v0.14), R2 Review Army (v0.15), R3 Smart Ceremony (v0.16), R4 /autoship (v0.17), R5 Studio (v0.18). Inspired by Compound Engineering, adapted to GStack's architecture. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * test: learnings bin script unit tests — 13 tests, free Tests gstack-learnings-log (valid/invalid JSON, timestamp injection, append-only) and gstack-learnings-search (dedup, type/query/limit filters, confidence decay, user-stated no-decay, malformed JSONL skip). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.13.4.0) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * test: learnings resolver + bin script edge case tests — 21 new tests, free Adds gen-skill-docs coverage for LEARNINGS_SEARCH, LEARNINGS_LOG, and CONFIDENCE_CALIBRATION resolvers. Adds bin script edge cases: timestamp preservation, special characters, files array, sort order, type grouping, combined filtering, missing fields, confidence floor at 0. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: sync package.json version with VERSION file (0.13.4.0) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: gitignore .factory/ — generated output, not source Same pattern as .claude/skills/ and .agents/. These SKILL.md files are generated from .tmpl templates by gen:skill-docs --host factory. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * test: /learn E2E — seed 3 learnings, verify agent surfaces them Seeds N+1 query pattern, stale cache pitfall, and rubocop preference into learnings.jsonl, then runs /learn and checks that at least 2/3 appear in the agent's output. Gate tier, ~$0.25/run. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…y (v0.13.5.0) (#641) * test: add 16 failing tests for 6 community fixes Tests-first for all fixes in this PR wave: - #594 discoverability: gstack tag in descriptions, 120-char first line - #573 feature signals: ship/SKILL.md Step 4 detection - #510 context warnings: no preemptive warnings in generated files - #474 Safety Net: no find -delete in generated files - #467 telemetry: JSONL writes gated by _TEL conditional - #584 sidebar: Write in allowedTools, stderr capture - #578 relink: prefixed/flat symlinks, cleanup, error, config hook Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: replace find -delete with find -exec rm for Safety Net (#474) -delete is a non-POSIX extension that fails on Safety Net environments. -exec rm {} + is POSIX-compliant and works everywhere. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: gate local JSONL writes by telemetry setting (#467) When telemetry is off, nothing is written anywhere — not just remote, but local JSONL too. Clean trust contract: off means off everywhere. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: remove preemptive context warnings from plan-eng-review (#510) The system handles context compaction automatically. Preemptive warnings waste tokens and create false urgency. Skills should not warn about context limits — just describe the compression priority order. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: add (gstack) tag to skill descriptions for discoverability (#594) Every SKILL.md.tmpl description now contains "gstack" on the last line, making skills findable in Claude Code's command palette. First-line hooks stay under 120 chars. Split ship description to fix wrapping. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: auto-relink skill symlinks on prefix config change (#578) New bin/gstack-relink creates prefixed (gstack-*) or flat symlinks based on skill_prefix config. gstack-config auto-triggers relink when skill_prefix changes. Setup guards against recursive calls with GSTACK_SETUP_RUNNING env var. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: add feature signal detection to version bump heuristic (#573) /ship Step 4 now checks for feature signals (new routes, migrations, test+source pairs, feat/ branches) when deciding version bumps. PATCH requires no feature signals. MINOR asks the user if any signal is detected or 500+ lines changed. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: sidebar Write tool, stderr capture, cross-platform URL opener (#584) Add Write to sidebar allowedTools (both sidebar-agent.ts and server.ts). Write doesn't expand attack surface beyond what Bash already provides. Replace empty stderr handler with buffer capture for better error diagnostics. New bin/gstack-open-url for cross-platform URL opening. Does NOT include Search Before Building intro flow (deferred). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: update sidebar-security test for Write tool addition The fallback allowedTools string now includes Write, matching the sidebar-agent.ts change from commit 68dc957. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.13.5.0) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: prevent gstack-relink from double-prefixing gstack-upgrade gstack-relink now checks if a skill directory is already named gstack-* before prepending the prefix. Previously, setting skill_prefix=true would create gstack-gstack-upgrade, breaking the /gstack-upgrade command. Matches setup script behavior (setup:260) which already has this guard. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: add double-prefix fix to changelog Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: remove .factory/ from git tracking and add to .gitignore Generated Factory Droid skills are build output, same as .agents/. They should not be committed to the repo. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: chrome-cdp localhost-only binding Restrict Chrome CDP to localhost by adding --remote-debugging-address=127.0.0.1 and --remote-allow-origins to prevent network-accessible debugging sessions. Clears 1 Socket anomaly (Chrome CDP session exposure). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: extension sender validation + message type allowlist Add sender.id check and ALLOWED_TYPES allowlist to the Chrome extension's message handler. Defense-in-depth against message spoofing from external extensions or future externally_connectable changes. Clears 2 Socket anomalies (extension permissions). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: checksum-verified bun install Replace unverified curl|bash bun installation with checksum-verified download-then-execute pattern. The install script is downloaded, sha256 verified against a known hash, then executed. Preserves the Bun-native install path without adding a Node/npm dependency. Clears Snyk W012 + 3 Socket anomalies. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: content trust boundary markers in browse output Wrap page-content commands (text, html, links, forms, accessibility, console, dialog, snapshot) with --- BEGIN/END UNTRUSTED EXTERNAL CONTENT --- markers. Covers direct commands (server.ts), chain sub-commands, and snapshot output (meta-commands.ts). Adds PAGE_CONTENT_COMMANDS set and wrapUntrustedContent() helper in commands.ts (single source of truth, DRY). Expands the SKILL.md trust warning with explicit processing rules for agents. Clears Snyk W011 (third-party content exposure). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: harden trust boundary markers against escape attacks - Sanitize URLs in markers (remove newlines, cap at 200 chars) to prevent marker injection via history.pushState - Escape marker strings in content (zero-width space) so malicious pages can't forge the END marker to break out of the untrusted block - Wrap resume command snapshot with trust boundary markers - Wrap diff command output with trust boundary markers - Wrap watch stop last snapshot with trust boundary markers Found by cross-model adversarial review (Claude + Codex). * chore: bump version and changelog (v0.13.4.0) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * chore: gitignore .factory/ and remove from tracking Factory Droid support was removed in this branch. The .factory/ directory was re-added by merging main (which had v0.13.5.0 Factory support). Gitignore it so it stays out. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…cture (v0.13.7.0) (#644) * feat: add parameterized resolver support to gen-skill-docs Extend the placeholder regex from {{WORD}} to {{WORD:arg1:arg2}}, enabling parameterized resolvers like {{INVOKE_SKILL:plan-ceo-review}}. - Widen ResolverFn type to accept optional args?: string[] - Update RESOLVERS record to use ResolverFn type - Both replacement and unresolved-check regexes updated - Fully backward compatible: existing {{WORD}} patterns unchanged Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: add INVOKE_SKILL resolver for composable skill loading New composition.ts resolver module that emits prose instructing Claude to read another skill's SKILL.md and follow it, skipping preamble sections. Supports optional skip= parameter for additional sections. Usage: {{INVOKE_SKILL:plan-ceo-review}} or {{INVOKE_SKILL:plan-ceo-review:skip=Outside Voice}} Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: use frontmatter name: for skill symlinks and Codex paths Patch all 3 name-derivation paths to read name: from SKILL.md frontmatter instead of relying solely on directory basenames. This enables directory names that differ from invocation names (e.g., run-tests/ directory with name: test). - setup: link_claude_skill_dirs reads name: via grep, falls back to basename - gen-skill-docs.ts: codexSkillName uses frontmatter name for Codex output paths - gen-skill-docs.ts: moved frontmatter extraction before Codex path logic Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: extract CHANGELOG_WORKFLOW resolver from /ship Move changelog generation logic into a reusable resolver. The resolver is changelog-only (no version bump per Codex review recommendation). Adds voice rules inline. /ship Step 5 now uses {{CHANGELOG_WORKFLOW}}. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * refactor: use INVOKE_SKILL resolver for plan-ceo-review office-hours fallback Replace inline skill loading prose (read file, skip sections) with {{INVOKE_SKILL:office-hours}} in the mid-session detection path. The BENEFITS_FROM prerequisite offer is unchanged (separate use case). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * refactor: BENEFITS_FROM resolver delegates to INVOKE_SKILL Eliminate duplicated skip-list logic by having generateBenefitsFrom call generateInvokeSkill internally. The wrapper (AskUserQuestion, design doc re-check) stays in BENEFITS_FROM. The loading instructions (read file, skip sections, error handling) come from INVOKE_SKILL. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * test: add resolver tests for INVOKE_SKILL, CHANGELOG_WORKFLOW, parameterized args 12 new tests covering: - INVOKE_SKILL: template placeholder, default skip list, error handling, BENEFITS_FROM delegation - CHANGELOG_WORKFLOW: content, cross-check, voice guidance, format - Parameterized resolver infra: colon-separated args processing, no unresolved placeholders across all generated SKILL.md files Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.13.7.0) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: journey routing tests — CLAUDE.md routing rules + stronger descriptions Three journey E2E tests (ideation, ship, debug) were failing because Claude answered directly instead of invoking the Skill tool. Root cause: skill descriptions in system-reminder are too weak to override Claude's default behavior for tasks it can handle natively. Fix has two parts: 1. CLAUDE.md routing rules in test workdir — Claude weighs project-level instructions higher than skill description metadata 2. "Proactively invoke" (not "suggest") in office-hours, investigate, ship descriptions — reinforces the routing signal 10/10 journey tests now pass (was 7/10). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: one-time CLAUDE.md routing injection prompt Add a preamble section that checks if the project's CLAUDE.md has skill routing rules. If not (and user hasn't declined), asks once via AskUserQuestion to inject a "## Skill routing" section. Root cause: skill descriptions in system-reminder metadata are too weak to reliably trigger proactive Skill tool invocation. CLAUDE.md project instructions carry higher weight in Claude's decision making. - Preamble bash checks for "## Skill routing" in CLAUDE.md - Stores decline in gstack-config (routing_declined=true) - Only asks once per project (HAS_ROUTING check + config check) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: annotated config file + routing injection tests gstack-config now writes a documented header on first config creation with every supported key explained (proactive, telemetry, auto_upgrade, skill_prefix, routing_declined, codex_reviews, skip_eng_review, etc.). Users can edit ~/.gstack/config.yaml directly, anytime. Also fixes grep to use ^KEY: anchoring so commented header lines don't shadow real config values. Tests added: - 7 new gstack-config tests (annotated header, no duplication, comment safety, routing_declined get/set/reset) - 6 new gen-skill-docs tests (preamble routing injection: bash checks, config reads, AskUserQuestion, decline persistence, routing rules) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump to v0.13.9.0, separate CHANGELOG from main's releases Split our branch's changes into a new 0.13.9.0 entry instead of jamming them into 0.13.7.0 which already landed on main as "Community Wave." Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: clarify branch-scoped VERSION/CHANGELOG after merging main Add explicit rules: merging main doesn't mean adopting main's version. Branch always gets its own entry on top with a higher version number. Three-point checklist after every merge. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: put our 0.13.9.0 entry on top of CHANGELOG Newest version goes on top. Our branch lands next, so our entry must be above main's 0.13.8.0. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: restore missing 0.13.7.0 Community Wave entry Accidentally dropped the 0.13.7.0 entry when reordering. All entries now present: 0.13.9.0 > 0.13.8.0 > 0.13.7.0 > 0.13.6.0. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add CHANGELOG integrity check rule After any edit that moves/adds/removes entries, grep for version headers and verify no gaps or duplicates before committing. Prevents accidentally dropping entries during reordering. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
#652) * feat: rotating founder resources in /office-hours closing Add Beat 3.5 with 34 curated resources (5 Garry Tan videos, 2 YC Backstory, 9 Lightcone Podcast, 8 Startup School, 10 PG essays) that rotate contextually each session. Includes dedup log to avoid repeats, analytics logging, and browser-open offers. Also adds chmod +x safety net to build script. * chore: bump version and changelog (v0.13.10.0) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
…(v0.14.0.0) (#653) * feat: /design-html skill — Pretext-native HTML from approved mockups New skill that takes approved design-shotgun mockups and generates production-quality HTML with Pretext for computed text layout. Text reflows on resize, heights adjust to content, zero hardcoded CSS. Includes vendored Pretext bundle (30KB), smart API routing per design type, AskUserQuestion refinement loop, framework detection, and 3-viewport verification screenshots. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate /design-html into design skill pipeline - design-shotgun: Step 6 option B now chains to /design-html - design-consultation: suggests /design-html after shipping DESIGN.md (conditional on screen-level output, not tokens-only) - plan-design-review: expanded chaining to include /design-shotgun and /design-html alongside review skills Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * test: update plan-design-review chaining test for design skills plan-design-review now chains to /design-shotgun and /design-html in addition to review skills. Update the assertion to match. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: add gstack keyword to design-html description for validation Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v0.14.0.0) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.