196 Commits

Author SHA1 Message Date
Jeff Smolinski 882ef55e37 fix(docs): update skill count in README banner; teach update-docs.py to manage it
The capsule-render banner embeds counts URL-encoded in the image URL,
where no marker comment can live, so it stayed at 66 skills through the
v0.4.16 release. update-docs.py now rewrites the banner's desc=
parameter from computed counts, matching the existing marker-less
version-badge treatment.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 15:19:14 -05:00
Jeffallan d5d2dd8ae2 chore: release v0.4.16 (#227)
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
v0.4.16
2026-08-07 14:25:06 -05:00
Jeff Smolinski fb67815707 fix(devops-engineer): operationalize production-deploy approval gate in core workflow
The constraints said never to deploy to production without explicit
approval, but the workflow's deploy step proceeded straight to rollout.
The workflow now determines the target environment and, for production
or customer-facing targets, requires presenting the deployment summary
and rollback plan and receiving explicit approval first. Mirrors the
terraform-engineer gate added in #213. Closes #196, closes #212.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 14:14:54 -05:00
Jeff Smolinski 38979c718b docs(changelog): record #213 terraform approval gate under Unreleased
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 13:34:48 -05:00
Sergiu Lupaiescu d0e7f4e890 fix(terraform-engineer): require user approval between plan and apply (#213) 2026-08-07 13:34:27 -05:00
Jeff Smolinski 10fe40c827 docs(changelog): record #216 Cohere key fix under Unreleased
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 13:28:42 -05:00
Haoqian d759b157fa fix(rag-architect): avoid hard-coded Cohere API key in reranking example (#216) 2026-08-07 13:28:21 -05:00
Jeff Smolinski 2bb82611ee feat(skills): add django-storages-s3 skill (#218)
Production S3 file storage for Django: STORAGES dict (4.2+), custom
public/private backends, presigned URLs, CloudFront, IAM, and testing
with moto/InMemoryStorage. Squash-merged locally to resolve count and
changelog conflicts with today's main; counts recomputed via
update-docs.py (67 skills, 371 references).

Co-authored-by: Awais Qureshi <awais.qureshi@arbisoft.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 13:26:34 -05:00
Jeff Smolinski 1508683aad fix(devops-engineer): replace archived kaniko with BuildKit rootless in gitlab-ci reference
Post-merge patch for #219: kaniko is archived and unmaintained, so the
flagship build example now uses moby/buildkit:rootless with registry
cache, and principle 8 notes the migration. Also syncs reference-file
counts (366 -> 367) via update-docs.py, fixing the CI failure on the
merge commit. Social preview PNG regen deferred to next release.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 13:17:17 -05:00
AKA 04d49bb11b feat(devops-engineer): add GitLab CI/CD reference (#219)
Co-authored-by: Азамат Касымов <azkasymov@dc.kompanion.kg>
2026-08-07 13:15:02 -05:00
Jeff Smolinski dff05dcd9e fix(site): clear post-July dependency advisories via fresh npm audit fix
Follow-up to #220: reduces audit findings from 9 to 5. The remaining
five require a semver-major Astro 7 upgrade and are tracked separately.
Site builds clean (98 pages) with the updated lockfile.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 13:10:10 -05:00
Jeff Smolinski 30b8137dfa docs(changelog): record #220 docs-site dependency patch under Unreleased
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 13:08:33 -05:00
chgreer1070 0f2b80cbf1 fix(site): patch docs-site dependency vulnerabilities via npm audit fix (#220)
Co-authored-by: chris greer <chgreer1079@gmail.com>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-08-07 13:08:15 -05:00
Jeff Smolinski 7f3f3bbc83 fix(skills): remove hardcoded local path in nestjs-expert; reject absolute refs in ReferencePathChecker
CI caught a contributor-machine absolute path in nestjs-expert that the
local run missed: joining an absolute ref onto a base dir returns it
unchanged, so it "resolved" against a stale clone outside the repo.
The checker now flags absolute paths unconditionally, keeping the check
hermetic across environments.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 12:51:02 -05:00
Jeff Smolinski d9101ef77c fix(skills): correct legacy-modernizer path in fastapi-expert; add ReferencePathChecker
A strict path audit following #225 found one more broken reference:
fastapi-expert's migration-from-django.md cited an absolute-style path
that resolves nowhere. To guard this recurring bug class, add
ReferencePathChecker to validate-skills.py: every backtick or
markdown-link .md path in skill files must resolve relative to the
containing file or the skill root. Runs by default, so CI and
make validate exercise it on every push.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01E6LA4sndtVGyvoXqwYeHoB
2026-08-07 12:48:51 -05:00
Vasu Garg cfce89847a fix(skills): broken relative reference paths in vue-expert-js and react-expert (#225)
- vue-expert-js/SKILL.md pointed to vue-expert/references/*.md, which does
  not exist relative to the skill directory; the shared references live in
  the sibling skill, so the paths need a ../ prefix.
- react-expert/references/migration-class-to-modern.md pointed to
  react-expert/references/server-components.md from inside that same
  directory; corrected to references/server-components.md to match the
  path convention used elsewhere in the skill.

Without these, an agent following the 'defer to vue-expert' instruction
fails to Read the referenced files.
2026-08-07 12:45:44 -05:00
Jeff Smolinski e8be415bc9 chore: release v0.4.15
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v0.4.15
2026-05-20 08:33:47 -05:00
Jeff Smolinski d3a08a374e fix(commands): correct YAML frontmatter in complete-ticket (#193)
The frontmatter in commands/project/execution/complete-ticket.md had two
strict-YAML errors that broke parsing in oh-my-pi and the GitHub renderer:

- Line 3: unquoted double quotes mid-scalar (`"In Review"`) were
  interpreted as a quoted-string opener with trailing junk.
- Line 4: `argument-hint: [ticket-key] (optional...)` led with `[`,
  which opens a flow-style sequence and then failed on the prose tail.

Wrap both values in matching outer quotes.

Also add CommandFrontmatterChecker to scripts/validate-skills.py. It
strict-parses the YAML frontmatter of every commands/**/*.md with
PyYAML (no fallback to the lenient simple_yaml_parse, which had been
masking these bugs), and runs as part of the workflow validation pass.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 08:30:56 -05:00
Jeff Smolinski 5e8b6b8ff4 docs(site): add privacy policy
Adds a privacy policy at /privacy/ on the docs site, required for the
Anthropic plugin marketplace submission. Discloses the docs site's
third-party services (GA4, jsDelivr, GitHub Pages), clarifies that the
plugin itself collects nothing and runs locally, and notes that user-
configured MCP server integrations (e.g., Atlassian) follow the vendor's
own privacy policy. Contact channel is the maintainer's existing
Formspree contact form on jeffallan.github.io.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-01 10:43:38 -05:00
Jeff Smolinski da77cf5ef8 chore: release v0.4.14
Patch release crediting @sobstel for fixing invalid pg_stat_* column
references in postgres-pro reference docs (#191).

Also clarifies the release checklist: only regenerate the social preview
when skill/workflow/reference counts change (the image embeds counts, not
the version), and use `npx -p puppeteer` instead of `npm install --no-save`.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v0.4.14
2026-05-01 09:10:30 -05:00
sobstel 21cde6f6e1 fix(postgres-pro): correct pg_stat query examples (#191)
Made-with: Cursor
2026-05-01 08:52:20 -05:00
Jeff Smolinski efebc44c90 feat(skills): add canonical Documentation backlink to every SKILL.md
Append a [Documentation](docs-site-url) line as the last content line of
every SKILL.md (66 files). Skill aggregators that consume raw source
markdown (skills.sh, etc.) will render this as a real <a href> backlink
to the docs site, providing free SEO link equity from each aggregator.

The strip rule in syncSkillPages (commit 3385184) suppresses the line in
the docs site's own rendered output, so the docs site, public markdown
mirrors, llms.txt, and llms-full.txt remain free of self-references.

URL formula: https://jeffallan.github.io/claude-skills/skills/{domain}/{skill-name}/
where {domain} comes from metadata.domain in the frontmatter (defaulting
to "specialized" if absent, matching the sync-content.mjs fallback) and
{skill-name} is the skill's directory name.

Documented the convention in CLAUDE.md so future skill additions follow
the same pattern.

Release: v0.4.13 (patch). No skill count changes (66 skills, 9 workflows,
366 reference files).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v0.4.13
2026-04-28 22:32:32 -05:00
Jeff Smolinski 3385184ef5 feat(docs-site): strip canonical Documentation backlink during sync
The bulk SKILL.md update appends a [Documentation](docs-site-url) line so
that aggregators (skills.sh, etc.) consuming raw source markdown render
a real <a href> backlink to the docs site for SEO. On the docs site itself
that URL points at the page rendering it — a self-link with no value and
visual redundancy. syncSkillPages now strips that exact line at build time
so the docs site, public markdown mirrors, llms.txt, and llms-full.txt
all render clean while the source files retain the line for downstream
syndication.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-28 22:24:45 -05:00
Jeff Smolinski 3d95bb16b0 chore: regenerate social preview for v0.4.12
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v0.4.12
2026-04-21 09:48:33 -05:00
Jeff Smolinski 055f25edf4 chore: release v0.4.12
Cuts the Unreleased section into v0.4.12, covering hub skill
cross-references (#68), Prettier markdown enforcement (#158),
the README docs link fix (#180), and workflow command frontmatter
additions (#182-#186).

Counts unchanged: 66 skills, 9 workflows, 366 reference files.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 09:48:24 -05:00
Jeff Smolinski 701e8037ca docs: log PRs #180, #182-#186 and credit contributors
Moves the merged README link fix and command frontmatter additions
into the Unreleased section: #180 and #186 under Fixed, #182-#185
under Changed (explicit-over-implicit, since filename-derived naming
was already working). Credits @modir and @xiaolai under Contributors.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 09:26:21 -05:00
xiaolai 1339f2f151 Fix: add missing name field to execution command frontmatter (#185)
Commands execute-ticket and complete-ticket were missing the required
name field in their YAML frontmatter, preventing registration by name
in Claude Code.

Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
2026-04-21 09:24:57 -05:00
xiaolai f83ad12243 Fix: add missing name field to planning command frontmatter (#184)
Commands create-epic-plan and create-implementation-plan were missing
the required name field in their YAML frontmatter, preventing registration
by name in Claude Code.

Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
2026-04-21 09:24:44 -05:00
xiaolai 04157247ad Fix: add missing name field to discovery command frontmatter (#183)
Commands approve-synthesis, synthesize-discovery, and create-epic-discovery
were missing the required name field in their YAML frontmatter, preventing
registration by name in Claude Code.

Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
2026-04-21 09:24:33 -05:00
xiaolai 30f4520602 Fix: add missing name field to retrospectives command frontmatter (#182)
Commands complete-sprint and complete-epic were missing the required
name field in their YAML frontmatter, preventing registration by name
in Claude Code.

Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
2026-04-21 09:24:22 -05:00
xiaolai 8c1f8d64c8 Fix: add missing name field to common-ground command frontmatter (#186)
COMMAND.md was missing the required name field in its YAML frontmatter,
preventing registration as the common-ground command in Claude Code.
The name common-ground is confirmed by common-ground.yaml which declares
command: "common-ground" for this file.

Co-authored-by: claude[bot] <claude[bot]@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
2026-04-21 09:24:06 -05:00
Raffael 12acfd885e Link to "docs" is not working in README.md (#180)
A small hack because the link to the workflow commands docs is not correctly rendered because the beginning of the line started with a comment sign. Hence I added the word "The" on the front.

Signed-off-by: Raffael <r.luthiger@huanga.com>
2026-04-21 09:23:42 -05:00
Jeff Smolinski 5b761018ce docs: fold #54 and #55 into v0.5.0 Chunk 2 command rewrites
Add performance/security testing (#55) and regression test suite (#54)
requirements to planning and retrospective command rewrites. Update
GitHub Issues table to reflect current status.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 17:39:26 -05:00
Jeff Smolinski 6590b9c7f0 feat: add markdown formatting enforcement and document conventions (closes #158)
Add prettier-markdown pre-commit hook for skills/, docs/, and root .md
files (excludes commands/ prompt templates). Add markdown check to CI
validate workflow and Makefile lint/format/validate targets. Document
all linting and formatting conventions in CONTRIBUTING.md.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 17:26:17 -05:00
Jeff Smolinski 173614e309 docs: add hub skill cross-reference work to unreleased changelog (#68)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 17:06:00 -05:00
Jeff Smolinski 3ba746ec95 feat: add curated cross-references to 7 hub skills (closes #68)
Add 22 back-references across devops-engineer, fullstack-guardian,
test-master, database-optimizer, security-reviewer, architecture-designer,
and kubernetes-specialist. Each back-reference was evaluated for genuine
agent routing value — framework/language skills that merely consume hub
capabilities were excluded (15% acceptance rate from 145 candidates).

Bump skill versions 1.1.0 → 1.1.1.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 17:02:07 -05:00
Jeff Smolinski 40dbf7e0e8 chore: regenerate social preview for v0.4.11
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
v0.4.11
2026-03-23 16:04:12 -05:00
Jeff Smolinski 4c167527f5 release v0.4.11: add context management reference to prompt-engineer
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:00:51 -05:00
Jeff Smolinski 791c7e41d2 feat: integrate context management content into prompt-engineer (#168)
Add context-management.md reference covering attention budgets,
lost-in-the-middle mitigation, four-bucket context tiering, KV-cache
optimization, observation masking, and degradation metrics. Content
adapted from Genius-apple's context-engineer skill submission.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 15:37:02 -05:00
Jeff Smolinski 45f1489ef1 add Serena gitignore and onboarding memories
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 11:24:09 -06:00
Jeff Smolinski 3bf9a24b76 add markdown as primary language for Serena MCP
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
v0.4.10
2026-03-06 15:28:38 -06:00
Jeff Smolinski c7afefbc1b release v0.4.10: skill quality improvements and Description Trap refinement
- Bump 65 skill versions from 1.0.0 to 1.1.0 (PR #172 changes)
- Update changelog for v0.4.10
- Update version.json and docs to 0.4.10

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-06 15:27:59 -06:00
Jeffallan 99b028f9a3 feat: improve skill quality across 65 skills (#172)
feat: improve skill quality across 65 skills
2026-03-06 15:19:33 -06:00
Jeff Smolinski b4da351dbf refine Description Trap: permit capability verbs, prohibit process steps
The original Description Trap (from obra/superpowers research) warned
specifically about process steps leaking into descriptions. The project
over-generalized this into a trigger-only rule that also excluded
capability verbs. This aligns the guidance with the AgentSkills.io spec
which recommends describing both what a skill does and when to use it.

- CLAUDE.md: narrow Description Trap, update format template and examples
- MODELCLAUDE.md: matching update to Description Trap section
- validate-skills.py: change startswith to contains check for "Use when"
- CHANGELOG.md: document changes under [Unreleased]

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-06 15:18:21 -06:00
Alan Pope cba32731d2 feat: improve skill quality across 65 skills via tessl review and optimization
Hullo @Jeffallan 👋

I ran your skills through `tessl skill review` at work and found some targeted improvements. Here's the before/after:

| Skill | Before | After | Change |
|-------|--------|-------|--------|
| ml-pipeline | 45% | 100% | +55% |
| rag-architect | 49% | 100% | +51% |
| test-master | 52% | 100% | +48% |
| javascript-pro | 54% | 100% | +46% |
| legacy-modernizer | 54% | 100% | +46% |
| fine-tuning-expert | 55% | 100% | +45% |
| wordpress-pro | 56% | 100% | +44% |
| angular-architect | 57% | 100% | +43% |
| atlassian-mcp | 57% | 100% | +43% |
| chaos-engineer | 57% | 100% | +43% |
| cloud-architect | 57% | 100% | +43% |
| kotlin-specialist | 57% | 100% | +43% |
| microservices-architect | 57% | 100% | +43% |
| monitoring-expert | 57% | 100% | +43% |
| react-native-expert | 57% | 100% | +43% |
| rust-engineer | 57% | 100% | +43% |
| salesforce-developer | 57% | 100% | +43% |
| spring-boot-engineer | 57% | 100% | +43% |
| sre-engineer | 57% | 100% | +43% |
| swift-expert | 57% | 100% | +43% |
| typescript-pro | 57% | 100% | +43% |
| vue-expert-js | 57% | 100% | +43% |
| code-reviewer | 61% | 100% | +39% |
| prompt-engineer | 61% | 100% | +39% |
| fullstack-guardian | 62% | 100% | +38% |
| nextjs-developer | 62% | 100% | +38% |
| spec-miner | 62% | 100% | +38% |
| database-optimizer | 63% | 100% | +37% |
| code-documenter | 65% | 100% | +35% |
| csharp-developer | 65% | 100% | +35% |
| golang-pro | 65% | 100% | +35% |
| security-reviewer | 65% | 100% | +35% |
| vue-expert | 65% | 100% | +35% |
| cli-developer | 66% | 100% | +34% |
| django-expert | 66% | 100% | +34% |
| embedded-systems | 66% | 100% | +34% |
| fastapi-expert | 66% | 100% | +34% |
| java-architect | 66% | 100% | +34% |
| kubernetes-specialist | 66% | 100% | +34% |
| laravel-specialist | 66% | 100% | +34% |
| mcp-developer | 66% | 100% | +34% |
| nestjs-expert | 66% | 100% | +34% |
| pandas-pro | 66% | 100% | +34% |
| php-pro | 66% | 100% | +34% |
| playwright-expert | 66% | 100% | +34% |
| rails-expert | 66% | 100% | +34% |
| shopify-expert | 66% | 100% | +34% |
| spark-engineer | 66% | 100% | +34% |
| terraform-engineer | 66% | 100% | +34% |
| debugging-wizard | 68% | 100% | +32% |
| architecture-designer | 70% | 100% | +30% |
| python-pro | 70% | 96% | +26% |
| react-expert | 70% | 100% | +30% |
| secure-code-guardian | 70% | 100% | +30% |
| api-designer | 71% | 100% | +29% |
| dotnet-core-expert | 71% | 100% | +29% |
| flutter-expert | 71% | 100% | +29% |
| graphql-architect | 71% | 100% | +29% |
| postgres-pro | 71% | 100% | +29% |
| websocket-engineer | 71% | 100% | +29% |
| devops-engineer | 73% | 100% | +27% |
| cpp-pro | 74% | 100% | +26% |
| game-developer | 74% | 100% | +26% |
| sql-pro | 74% | 100% | +26% |
| feature-forge | 79% | 100% | +21% |

**Average improvement: +35% across 65 skills** (from ~64% average to ~100% average)

<details>
<summary>Changes made across skills</summary>

The optimizations followed consistent patterns across all 65 skills:

- **Expanded frontmatter descriptions** — Added concrete action verbs (creates, configures, validates, generates) describing what each skill actually does, improving trigger specificity and completeness
- **Removed redundant sections** — Eliminated "Role Definition" and "When to Use This Skill" sections that duplicated information already in the frontmatter description
- **Added structured workflows with validation checkpoints** — Enhanced Core Workflow steps with explicit feedback loops (e.g., "if tests fail, fix and re-run") and verification commands
- **Added inline code examples** — Included concrete, executable code snippets demonstrating key patterns for each technology, making skills immediately actionable
- **Tightened constraints** — Made MUST DO/MUST NOT DO lists more specific with brief explanations of why each constraint matters, removing generic best practices Claude would already follow
- **Improved progressive disclosure** — Better organized reference tables with clearer "Load When" guidance

</details>

Honest disclosure — I work at @tesslio where we build tooling around skills like these. Not a pitch - just saw room for improvement and wanted to contribute.

If you want to run evals yourself, just `npm install @tessl/cli` then run `tessl skill review path/to/your/SKILL.md`, and click [here](https://tessl.io/registry/skills/submit) to find out more.

Thanks in advance 🙏

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-06 16:51:18 +00:00
Jeff Smolinski be4e1e265c update changelog for PR #166
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 07:52:05 -06:00
Jeff Smolinski 7ed3d62166 Merge branch 'origin/main' into clean-readme
Resolve README.md conflict: accept PR #166 formatting changes.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 07:50:05 -06:00
Jeff Smolinski e54ffc7486 fix stale counts on docs landing page, add astro files to release automation
The docs site landing page still showed 65 skills and 357 references.
Updated to 66/365 and added both astro site files to update-docs.py
so they stay in sync on future releases.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 07:45:15 -06:00
hasan613 408c037c31 docs: standardize links and improve Quick Start readability 2026-02-27 03:02:20 +03:00
Jeff Smolinski ad70aa3c40 add formatting commit to .git-blame-ignore-revs
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-24 22:10:01 -06:00