Files

vitya 05ea501683 meta(tasks): create [bootstrap-skill-deps-check]; supersede [bootstrap-recommend-projects-meta]

New ⚪ ready task to refactor project-bootstrap Step 5.6 from a per-skill
mirror shape (superpowers detector + deferred projects-meta detector) into
a single generic "skill dependencies check" that walks the canonical
CLAUDE.md template, maps each trigger line to its fulfiller (skill or
plugin), and prints one chat-only recommendation block with install
commands. Hard rule "never auto-install" carries over verbatim.

Decision driven by the [interns-skills-mvp] commit (0eb7dd1): adding
`delegate to interns when allowed` made it visible that every new
canonical trigger would otherwise need its own Step 5.X mirror — doesn't
scale.

[bootstrap-recommend-projects-meta] marked superseded — that task was the
seed of this generalization; it stays in STATUS.md for archaeology and
will be closed once [bootstrap-skill-deps-check] ships.

.tasks/bootstrap-skill-deps-check.md carries the full spec: scope,
key files, decisions log (5 entries), open questions, initial
trigger→fulfiller seed table.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-05 19:41:04 +03:00

16 KiB

Raw Blame History

Task Board

Updated: 2026-05-05

🟢 [shared-wiki-skill-catalog] — публикация каталога claude-skills в shared projects-wiki

Status: done Where I stopped: packages/claude-skills опубликован в projects-wiki через knowledge_ingest (3 коммита Gitea: ae2cc9a3 page + 14bdd863 index + 001cdd0e log); каталог всех 20 скиллов сгруппирован: bootstrap & discipline / wiki+tasks / MCP integrations / caveman cluster / discovery+platform; cross-link с concepts/setup-using-skill-pair (паттерн setup-X/using-X) и packages/projects-meta-mcp; local ~/projects/projects-wiki/ подтянут ff-only до 001cdd0e; запись в local .wiki/log.md добавлена Next action: (none — kept until merged) Branch: master

🟢 [project-discipline-skill] — project-discipline skill + project-bootstrap 1.5.0 canonical trigger

Status: done Where I stopped: skills/project-discipline/ shipped at v0.1.0 (four rules: convention-override, master-only, semver-bumping with first-edit-unversioned clause, session-scoped ask-before-push with grant/revoke and force/delete/non-ff exceptions); assets/CLAUDE.md.template gains follow project discipline after pull remote before work; project-bootstrap 1.4.0→1.5.0 with new commentary paragraph in Step 5 + manifest table row in Step 5.5; .wiki/concepts/skill-versioning.md extended (Rule 3 expands version: requirement to ALL skills, not just infra subset); root CLAUDE.md dogfood line added; both archives rebuilt in dist/; installed via install.sh; description length verified 855 chars (≤900); smoke scenarios 1 (skill listing with full description, not H1 fallback) + 9 (length regression check) verified inline post-install; scenarios 2-8 (Rule 1-4 in fresh sessions, bootstrap upgrade) deferred — require multi-session orchestration; design rationale at .wiki/concepts/project-discipline-design.md Next action: (none — kept until merged) Branch: master

🟢 [pulling-before-work-skill] — pulling-before-work skill + project-bootstrap 1.4.0 canonical trigger

Status: done Where I stopped: skills/pulling-before-work/ shipped at v1.0.0 (mode-3 activation: session start + on-demand re-sync; git pull --ff-only; skips on no-remote / no-upstream / dirty / detached); assets/CLAUDE.md.template gains pull remote before work; project-bootstrap 1.3.0→1.4.0 with new commentary paragraph in Step 5; root CLAUDE.md dogfood line added; both archives rebuilt in dist/; installed via install.sh; smoke scenarios 1/2/3/4/5/8/9 + inline bootstrap-merge check verified by controller, scenarios 6 (diverged remote) and 7 (re-sync trigger across a fresh session) deferred — both require multi-session orchestration not yet tooled; design rationale at .wiki/concepts/pulling-before-work-design.md Next action: (none — kept until merged) Branch: master

🟢 [bootstrap-claude-md-merge] — project-bootstrap upgrade: idempotent merge of canonical CLAUDE.md triggers

Status: done Where I stopped: Step 5 in skills/project-bootstrap/SKILL.md split into Init / Upgrade-merge — upgrade reads existing CLAUDE.md, substring-diffs vs template (case-insensitive trim), platform line preserved if user pinned a non-host platform, asks confirm with diff, appends only missing canonical lines; version bump 1.2.0 → 1.3.0; skill row in skills/project-bootstrap/README.md updated to note idempotent merge; root README.md + README.ru.md upgrade-mode paragraph extended; reinstalled to ~/.claude/skills/project-bootstrap/, dist/project-bootstrap.skill rebuilt; concept page .wiki/concepts/bootstrap-claude-md-merge.md written; index + log entries added Next action: (none — kept until merged) Branch: master

🟢 [bootstrap-projects-meta-trigger] — project-bootstrap CLAUDE.md template gains `check across all projects`

Status: done Where I stopped: template + inline SKILL.md copy patched, version bumped 1.1.0→1.2.0; both READMEs and local CLAUDE.md synced; skill reinstalled to ~/.claude/skills/project-bootstrap/, dist/project-bootstrap.skill rebuilt; concept extension added to .wiki/concepts/projects-meta-skills.md; log entry written; verified installed version: 1.2.0 and template carries the new line Next action: (none — kept until merged) Branch: master

Status: done Where I stopped: Step 5.6 added to skills/project-bootstrap/SKILL.md (read ~/.claude/plugins/installed_plugins.json, print install command + upstream link if superpowers@claude-plugins-official key missing); version bumped 1.0.0 → 1.1.0; README workflow numbering updated; skill reinstalled to ~/.claude/skills/project-bootstrap/ and dist/project-bootstrap.skill rebuilt; log entry added Next action: (none — kept until merged) Branch: master

🟢 [mac-support-scripts] — fix install.sh / build.sh for stock macOS

Status: done Where I stopped: patched both scripts to drop mapfile (bash 4+) and find -printf (GNU find) — replaced with portable shell glob + basename + sort; verified on git-bash (16 skills discovered, install dry-run + build smoke-test passed); wiki concept page install-portability.md added; index + log updated Next action: (none — kept until merged) Branch: master

🟡 [skill-readmes] — write English README.md for every skill + translate root README

Status: paused Where I stopped: infra-skill cluster done — READMEs for project-bootstrap, setup-wiki, setup-tasks, using-wiki, using-tasks; root README translated; cross-links between all five skills wired up Next action: continue with the next batch — suggest the caveman cluster (caveman, caveman-commit, caveman-review, caveman-help, caveman-compress, compress) since they form a coherent group; or jump to active-platform, find-skills, setup-context7, using-context7, using-markitdown if the caveman cluster needs deduplication first (see compress-dedup task) Branch: master

⚪ [compress-dedup] — resolve compress vs caveman-compress duplication

Status: ready Where I stopped: (not started) — noticed during initial migration that both skills ship identical scripts/ subtrees (compress.py, cli.py, detect.py, validate.py, benchmark.py) Next action: diff -r skills/compress skills/caveman-compress; decide if one is canonical and the other should be removed (or aliased) Branch: (not started)

⚪ [install-ps1] — PowerShell-native install.ps1

Status: ready Where I stopped: (not started) — install.sh already works in git-bash on Windows, so this is convenience for PS-only users Next action: mirror scripts/install.sh logic in PowerShell (copy skills/<name>/ → ~/.claude/skills/<name>/, support $env:CLAUDE_SKILLS_DIR override) Branch: (not started)

⚪ [archive-roundtrip-test] — smoke-test for .skill archive shape

Status: ready Where I stopped: (not started) — caught the PowerShell-Compress-Archive backslash bug manually; a smoke-test would catch the next regression automatically Next action: add a script that builds dist/<name>.skill, unzips into a temp dir, and diff -r against skills/<name>/. Fail on any difference. Run from CI or pre-commit if we add one Branch: (not started)

⚪ [active-platform-tuning] — tune active-platform triggers based on real usage

Status: ready Where I stopped: (not started) — current triggers are best-guess; haven't observed real misfires/missed-fires yet Next action: keep a tally of times the skill should have triggered but didn't (or vice versa); revise the description / trigger list after ~5 real signals. Also add WSL handling explicitly if it comes up Branch: (not started)

⚪ [active-platform-eval] — run formal skill-creator eval loop on active-platform

Status: ready Where I stopped: (not started) — depends on having ~5 real interactions logged for tuning context Next action: invoke skill-creator to set up an eval set for active-platform; benchmark current vs revised description Branch: (not started) Blocker: waits on active-platform-tuning to gather signals first

⚪ [skills-grouping-revisit] — revisit flat vs grouped skills/ layout if count grows past ~30

Status: ready Where I stopped: (not started) — current 14 skills fit fine in flat skills/; threshold for revisiting is ~30 Next action: when triggered (skill count crosses threshold), evaluate variant B (grouped by family) vs variant C (flat + manifest) from the original brainstorm in .wiki/concepts/repo-layout.md Branch: (not started)

⚪ [bootstrap-recommend-projects-meta] — Step 5.7 in project-bootstrap: recommend `setup-projects-meta` if MCP tools missing

Status: superseded Where I stopped: (not started) — opened as follow-up to [bootstrap-projects-meta-trigger] (commit b120397). The check across all projects trigger is now in every CLAUDE.md template, but on a fresh machine without projects-meta-mcp registered the trigger is silently no-op until the user verbally asks for cross-project work Next action: superseded by [bootstrap-skill-deps-check] (2026-05-05) — that task absorbs this one into a generic Step 5.6 that walks the whole canonical trigger list, not just projects-meta. Kept here for git-log archaeology; close once [bootstrap-skill-deps-check] ships Branch: (not started) Blocker: waits on observing a real fresh-machine bootstrap that silently misses the dependency — defer until we see the failure mode in practice

⚪ [bootstrap-skill-deps-check] — refactor project-bootstrap Step 5.6 into a generic skill-dependencies check (replaces per-skill `Step 5.X` mirror shape)

Status: ready Where I stopped: (not started) — design agreed in chat 2026-05-05 after [interns-skills-mvp] shipped: adding delegate to interns when allowed to the canonical CLAUDE.md template made it visible that every new trigger would need its own Step 5.X detector mirror, which doesn't scale. Decision: collapse Step 5.6 (superpowers plugin) and the deferred [bootstrap-recommend-projects-meta] into a single generic Step 5.6 that reads the canonical trigger list and maps each line to a fulfiller (skill or plugin) for missing-deps detection. Hard rule "never auto-install" carries over verbatim — chat-only recommendation block with copy-pasteable install commands Next action: Read .tasks/bootstrap-skill-deps-check.md for the full spec — initial trigger→fulfiller seed table is in the Notes section. Write .wiki/concepts/bootstrap-skill-deps-check.md design page first (rationale: why generic over per-skill mirrors, kind: skill vs plugin distinction, MCP server caveat). Then edit skills/project-bootstrap/SKILL.md Step 5.6 in place, bump version: 1.6.0 → 1.7.0. Update skills/project-bootstrap/README.md Workflow Step 5.6 description. Rebuild + install. Close [bootstrap-recommend-projects-meta] with a note in commit message Branch: (not started)

🟢 [using-projects-meta-freshness-gate] — добавить Step 0 freshness check в using-projects-meta

Status: done Where I stopped: Step 0 добавлен в using-projects-meta/SKILL.md (cache_age>10мин или errors>0 → dist/sync.js для read; shared-wiki write — безусловный git pull --ff-only; 401/403 → auth.toml rotation); зеркалирован в README; bump 1.0.0→1.1.0; rebuild+install OK; description trimmed до 924 chars после регрессии (полное Step 0 в description ~1450 chars сломало парсинг — harness зафолбэкнулся на H1, скилл потерял все триггеры); Common mistakes + Red flags расширены; concept page + log обновлены; new feedback memory skill_description_length_limit Next action: (none — kept until merged) Branch: master

🟢 [projects-meta-skills] — setup-projects-meta + using-projects-meta — пара скиллов для projects-meta-mcp

Status: done Where I stopped: оба скилла v1.0.0 написаны (8-phase setup + using policy с local-first и two-step mutation), README обоим, собраны в dist/, установлены в ~/.claude/skills/, видны в Skill tool listing; concept-страница projects-meta-skills.md добавлена в .wiki/; path-canon ~/projects/projects-wiki/ починен в скиллах после wiki-path-mismatch fix (upstream commit 621a69f); якоря MCP подтянулись после ручного pull Next action: (none — kept until merged); требуется restart Claude Code чтобы новые скиллы стали активными в новой сессии (build/install уже сделаны) Branch: master

⚪ [refresh-project-bootstrap] — Run the `project-bootstrap` skill in this repo to refresh its layout to the canonical state (git, .gitignore, README.md, .wiki/, .tasks/, CLAUDE.md). The skill handles both greenfield bootstrap and refresh of existing repos.

Status: ready Where I stopped: (not started) Next action: Invoke the project-bootstrap skill at the repo root, confirm each file write, push to main. Branch: n/a

🟢 [interns-skills-mvp] — Ship `setup-interns` + `using-interns` skills (v0.1.0) and bump `project-bootstrap` 1.5.0 → 1.6.0 with `delegate to interns when allowed` CLAUDE.md trigger, manifest entry, Step 5 commentary paragraph, idempotent merge into existing CLAUDE.md files. Two-skill pair following the setup-X / using-X prior art (setup-context7/using-context7, setup-projects-meta/using-projects-meta). Per-session permission grant policy mirroring project-discipline Rule 4 (ask-mode default, conversational grant/revoke, always-ask paths for secrets even with active grant, session-end reset).

Status: done Where I stopped: skills/setup-interns/ v0.1.0 shipped (8-phase install: detect .common/lib/interns-mcp/ source, pip install -e, write .common/secrets/interns.env with endpoint API keys + ensure .gitignore covers it, register mcpServers.interns in ~/.claude.json with absolute Python interpreter path + cwd to project root); skills/using-interns/ v0.1.0 shipped (runtime policy mirroring project-discipline Rule 4: ask-mode default, conversational grant/revoke, always-ask paths for .env / **/secrets/** / **/*.key / **/.ssh/** / **/.aws/credentials etc with transitive rule + cost-cap >$0.10, session-end reset; routing hints for bulk_text_read + transcript_distill); project-bootstrap 1.5.0→1.6.0 with canonical CLAUDE.md trigger delegate to interns when allowed inserted between follow project discipline and we're on Windows, new Step 5 commentary paragraph, Step 5.5 manifest table extended with setup-interns + using-interns + project-discipline rows; root CLAUDE.md dogfood line added; both READMEs written; descriptions verified within budget (setup-interns 899 / using-interns 814 / ≤900); both archives rebuilt + installed; harness listing confirms full descriptions (no H1 fallback) for all three skills Next action: (none — kept until merged) Branch: master

16 KiB Raw Blame History Unescape Escape