Commit graph

9 commits

Author SHA1 Message Date
Claude
a088c8ba77
fix: phantom-path sweep — 888 unresolvable references to zero
A directory reorg added a skills/ path segment; hundreds of references never
followed. This sweep repoints every path-like reference in SKILL.md, agents,
commands, orchestration, and templates to verified on-disk targets:

- 30 root commands + 19 root agents: missing skills/ segment inserted
- 7 c-level persona agents: 17 hallucinated reference filenames substituted
  with the real files (e.g. okr_execution.md -> process_frameworks.md)
- 5 research skills: phantom scripts/office/validate.py step replaced with a
  runnable stdlib zip-integrity check
- email agents: skills frontmatter corrected to productivity/email
- orchestration/ORCHESTRATION.md + templates: stale paths fixed;
  agent-template now requires trigger phrasing in descriptions (root cause)
- 76 more files across engineering, c-level-advisor, compliance-os,
  research-ops, ra-qm, marketing, productivity; dual-publish pairs mirrored
- dead refs dropped/replaced where no target ever existed (REGISTRY.md,
  trend_analyzer.py, cursor-microinteractions.md)

New: scripts/check_paths.py linter (CI gate G1) + narrow allowlist for
teaching examples. Verified: 540 files scanned, 0 unresolvable.

https://claude.ai/code/session_019AJddAL1NADWMXsy1qNPQF
2026-06-10 14:33:00 +00:00
Claude
0f88555485
chore(versioning,docs): normalize stale versions to 2.9.0 + refresh counts
Version normalization (scope: only plugins/skills older than 2.7.0):
- Bumped 52 plugin.json + 21 SKILL.md `version` fields from pre-2.7.0
  values (1.0.0 / 2.0.x / 2.2.x / 2.3.x / 2.4.x / 2.5.x) to 2.9.0.
  Left 2.7.0+ packages untouched. Tool mirrors (.codex/.gemini/.vibe/
  .hermes) excluded. All manifests still pass check_plugin_json --all.

Doc count refresh (recomputed raw figures: 338 skills, 16 domains,
62 plugins, 533 Python tools, 676 references):
- marketplace.json: both descriptions + metadata.version -> 2.9.0.
- Root README: headline, badges (Skills 338, Agents 51+, Commands 87+),
  intro counts, convert section (338 skills / 9 tools), and the full
  domain table rebuilt to 16 domains summing to 338 (adds research-ops,
  business-operations, commercial, compliance-os; corrects product 17,
  marketing 46, c-level 66, ra-qm 18, finance 4, engineering 51/78).
- Fixed stale per-skill README `Version:` lines left inconsistent by the
  bump (andreessen, c-level-agents, product-team, senior-qa).
- Fixed stale domain README footers (product-team 17/17, c-level 66/66,
  project-management 9/9).
- CLAUDE.md scope line, structure tree, highlight, and footer synced to
  the raw figures.

https://claude.ai/code/session_01PUNmQVE4WYvcrzpq2anC3D
2026-05-27 05:22:59 +00:00
Claude
8fd3155868
fix(plugin-schema): align with live Claude Code spec — migrate to "./"-prefixed skills field
Resolves #712 and #714.

Claude Code's plugin spec (code.claude.com/docs/en/plugins-reference) requires
"All paths must be relative to the plugin root and start with ./". CC 2.1.144+
rejects the bare "skills" form with `Validation errors: skills: Invalid input`,
breaking plugin installs from this repo.

Changes:
- 47 plugin.json manifests migrated from "skills": "skills" → ["./skills"]
  (canonical post-CC 2.1.144 form, also accepts string "./skills")
- scripts/check_plugin_json.py rewritten to enforce the live spec:
  * Hard-fails on non-"./"-prefixed strings (except the legacy "skills" literal)
  * Hard-fails on empty strings/arrays and non-string array entries
  * Emits WARN (non-fatal) on the legacy "skills" literal as a safety net
    for copied templates — the WARN tolerance can be removed later
  * Tested against all four canonical forms + 3 garbage forms
- CLAUDE.md §5 rewritten to document the live spec, including the historical
  context of the v2.1.107 → v2.1.144 regression window that's now closed
- 15 stale broken symlinks under .gemini/skills/ removed (orphans from older
  reorgs that the sync script didn't prune because its cleanup is disabled)

Verification:
- scripts/check_plugin_json.py --all → 0 FAIL, 0 WARN, exit 0 (69 files OK)
- pytest tests/ → 2103 passed
- find . -type l ! -exec test -e {} \; → 0 broken symlinks
- Live spec forms tested: ["./"], ["./skills"], "./skills", ["./a","./b"] all pass
- Garbage forms rejected: bare "./", empty array, non-string entries, plain strings
2026-05-21 07:03:11 +00:00
Claude
663bc8afa2
fix(plugin.json): drop "./" prefix from skills field to clear /doctor warning (#686)
Claude Code 2.1.133+ rejects "skills": "./skills" with a "Path escapes
plugin directory" warning, even though ./skills resolves to a valid
subdirectory inside the plugin root. Per the upstream docs example, the
correct value is "skills" (no ./ prefix), which works across all CC
versions and other tools (Codex, Gemini CLI, etc.).

Applied to all 47 plugin.json files that used "./skills"; updated
CLAUDE.md ClawHub publishing constraints to document the new convention
and reference the issue.
2026-05-19 03:43:34 +00:00
Reza Rezvani
c4801730f5 fix(plugins): restructure 21 single-skill plugins into ./skills/<name>/ layout
Claude Code v2.1.126 rejects "skills": ["./"] (array form) at runtime
with the same "Path escapes plugin directory: ./" error as the bare
string form, despite the official plugins reference docs explicitly
showing ["./"] as a valid example. `claude plugin validate` accepts it,
so the bug is in the runtime loader, not the manifest validator.

The proven workaround (already used for self-improving-agent in #536)
is to follow the canonical layout:

  <plugin>/
  ├── .claude-plugin/plugin.json    skills: "./skills"
  └── skills/<plugin-name>/
      ├── SKILL.md
      ├── scripts/
      ├── references/
      └── ...

Applied to 21 single-skill plugins:
- engineering-team/a11y-audit, google-workspace-cli, snowflake-development
- engineering/behuman, code-tour, data-quality-auditor, demo-video,
  docker-development, helm-chart-builder, karpathy-coder, llm-cost-optimizer,
  llm-wiki, prompt-governance, statistical-analyst, terraform-patterns
- finance/business-investment-advisor
- marketing-skill/video-content-strategist
- product-team/agile-product-owner, apple-hig-expert, code-to-prd,
  research-summarizer

Moved with each SKILL.md: scripts/, references/, assets/, templates/,
expected_outputs/. Kept at plugin root (per Claude Code requirements):
agents/, commands/, hooks/, settings.json, README.md.

Bumped patch version on each restructured plugin so /plugin update
picks up the new layout.

Refs #539
2026-05-02 22:33:59 +02:00
Reza Rezvani
2717c8f933 fix(plugins): bump versions to force /plugin update to pick up #587 fix
PR #587 fixed `"skills": "./"` in 35 plugin.json files but did not bump
versions. Claude Code uses version as the cache key — same version string
means /plugin update is a no-op and users keep loading the cached
plugin.json with the broken path. The /plugin UI shows "1 error" on
every plugin from this marketplace because the installed cache is stale.

Per the official version-management docs:
> If you set `version` in `plugin.json`, you must bump it every time
> you want users to receive changes. Pushing new commits alone is not
> enough.

Bump every plugin.json by one patch using max(plugin_version,
marketplace_version) as the base so no version moves backward. Sync the
new versions back into .claude-plugin/marketplace.json.

After this lands, users run `/plugin marketplace update claude-code-skills`
followed by `/plugin update --all` to pull the fixed manifests.
2026-05-02 22:14:12 +02:00
Reza Rezvani
55e7e45512 fix(plugins): repair skills path in 35 plugin.json files (#539)
Claude Code v2.1.107+ rejects bare "skills": "./" with "Path escapes
plugin directory: ./". This blocks every plugin in the marketplace from
loading. Replace with the layout-appropriate form per the official
plugins reference:

- Single-skill plugin (SKILL.md at root):
    "skills": ["./"]                  array form (per docs example)
- Plugin with skills/ subdir:
    "skills": "./skills"              standard subdir layout
- Multi-skill domain plugin (skills are subfolders at root):
    "skills": ["./sub1", "./sub2"]    explicit list of skill dirs

The multi-skill case omits "./" from the array so the index SKILL.md
at the plugin root does not register as a skill — that would create
a redundant `marketing-skills:marketing-skills` namespace.

Verified by running `claude plugin validate` against all 35 manifests.

Closes #539
2026-05-02 21:55:58 +02:00
Reza Rezvani
a28f83b190 fix(llm-wiki): pass 8-phase plugin audit
Auto-fixes from plugin-audit pipeline:

Security (Phase 5, HIGH → 0):
- Rename assets/.cursorrules.template → cursorrules.template to clear
  hidden-file warning from skill-security-auditor. init_vault.py updated
  to map the non-hidden template name to the vault's .cursorrules.

Quality (Phase 3, 58.7 → 67.2, above 60 threshold):
- Add --json output to init_vault, append_log, update_index, export_marp
  (4 scripts that were PARTIAL in Phase 4; now all 8 PASS)
- Add expected_outputs/ directory with 8 sample JSON fixtures covering
  every script's --json output, lifting Completeness 66.7 → 85.0
- Expand error handling (try/except on file I/O) in append_log, init_vault,
  update_index, export_marp
- Add more comprehensive argparse help text and epilogs

Scripts (Phase 4, 4/8 PARTIAL → 8/8 PASS):
- All 8 scripts now support --json
- End-to-end smoke test confirmed: init → append_log → update_index →
  export_marp → lint → wiki_search → graph_analyzer → ingest_source all
  round-trip cleanly in --json mode

Marketplace compliance (Phase 6):
- llm-wiki/.claude-plugin/plugin.json version: 1.0.0 → 2.3.0 (match repo)
- marketplace.json llm-wiki entry: version 1.0.0 → 2.3.0, cursorrules
  reference updated (no dot prefix)
- engineering/.claude-plugin/plugin.json: count 38 → 44, version 2.2.0 →
  2.3.0, added llm-wiki to description

Ecosystem integration (Phase 7):
- Ran sync-codex-skills.py → llm-wiki added to .codex/skills-index.json
  and .codex/skills/llm-wiki symlink
- Ran sync-gemini-skills.py → llm-wiki + cs-wiki-* agents + wiki-*
  commands copied into .gemini/skills/
- mkdocs.yml nav: added 5 /wiki-* command entries

Final audit scores:
- Phase 2 Structure:    79.2/100 GOOD (↑ from 77.1)
- Phase 3 Quality:      67.2/100     (↑ from 58.7)
- Phase 4 Scripts:      8/8 PASS     (↑ from 4/8 PASS)
- Phase 5 Security:     PASS 0c/0h   (↓ from 1 HIGH)
- Phase 6 Marketplace:  plugin.json valid, version aligned
- Phase 7 Ecosystem:    codex+gemini synced, mkdocs nav updated
- Phase 8 Code Review:  Karpathy attribution, iron rules, cross-tool docs

Verdict: PASS WITH WARNINGS (Documentation dimension 49.6 — acceptable
for a reference-style skill with rich separate reference docs).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-11 01:35:52 +02:00
Reza Rezvani
b6ca45ddec feat(engineering): add llm-wiki plugin — second brain for Claude Code + Obsidian
Implements Karpathy's LLM Wiki pattern as a production-grade plugin. The LLM
incrementally ingests sources into a persistent, interlinked Obsidian vault —
updating entity/concept/source pages, flagging contradictions, maintaining an
index and append-only log. Knowledge compounds instead of being re-derived by
RAG on every query.

Plugin contents (engineering/llm-wiki/):
- SKILL.md with `context: fork` frontmatter for skill chaining
- 3 sub-agents: wiki-ingestor, wiki-librarian, wiki-linter
- 5 slash commands: /wiki-init, /wiki-ingest, /wiki-query, /wiki-lint, /wiki-log
- 8 Python tools (stdlib only): init_vault, ingest_source, update_index,
  append_log, wiki_search (BM25), lint_wiki, graph_analyzer, export_marp
- 8 reference docs: schema, page-formats, ingest/query/lint workflows,
  obsidian-setup, cross-tool-setup, memex-principles
- Vault templates: CLAUDE.md, AGENTS.md, .cursorrules, index.md, log.md,
  5 page templates (entity, concept, source, comparison, synthesis)
- Worked example vault on "LLM interpretability"
- .claude-plugin/plugin.json manifest

Cross-tool compatibility: the scripts are pure Python stdlib. Only the schema
loader changes per tool (CLAUDE.md for Claude Code, AGENTS.md for Codex CLI /
Cursor / Antigravity / OpenCode / Gemini CLI, .cursorrules for legacy Cursor).
init_vault.py --tool all installs all three.

Repo-level registration:
- Commands mirrored to top-level commands/ for repo-wide discovery
- Agents mirrored to agents/engineering/ as cs-wiki-{ingestor,librarian,linter}
- .claude-plugin/marketplace.json: new llm-wiki entry + version bump to v2.3.0
- CLAUDE.md updated: 234 skills, 313 Python tools, 432 refs, 28 agents, 27 commands

Also saved (deferred): craighewitt-mattpocock reimplementation plan at
documentation/implementation/ — 4-pod proposal for building better versions
of selected skills from thecraighewitt-skills and mattpocock-skills
collections. Not executed; awaiting user confirmation on scope.

End-to-end smoke test passed: init_vault → ingest → update_index → append_log
→ wiki_search → lint → graph_analyzer → export_marp all run against a fresh
vault with real pages, wikilinks, and frontmatter.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-11 01:14:11 +02:00