projectbluefin/website/docs

Claude Code and agent skill by projectbluefin/website

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Install

npx vskill@latest install projectbluefin/website/docs

SKILL.md

Website Skill Router

Agent entry point for projectbluefin/website. Find the skill that matches your task, load only that skill, then act.

Agents edit content. Agents never edit design.

Read order

  1. `AGENTS.md` — repo contract, boundaries, build commands.
  2. This file — task→skill mapping.
  3. The one skill file named in the table below.
  4. The source file that owns the requested content.

Do not load every skill. Load the smallest skill that answers the job.

Know what you are editing

/ is a website. /wolves/experience/ is a presentation performed to a live audience seated in a theater, projected and synchronized to music, with no interaction available to anyone. (/wolves/ itself is the teaser page that introduces it.) Web-page instincts — scroll it, shrink the type, add a control, add more words — are defects there. Read AGENTS.md under "/wolves/experience/ is a presentation" before touching that route.

Skill index

I need to...Load
Edit text, links, translations, data, or approved assets`content-maintenance/SKILL.md`
Edit SBOM extraction scripts, version comparison, or dakota-versions updater`sbom-version-extraction/SKILL.md`
Change layout, styling, a component, animation, or navigation`design-gate/SKILL.md`
Handle lore, fiction, quotes, or attribution`editorial-provenance/SKILL.md`
Run tests, builds, staging, deployment, or claim live status`validation/SKILL.md`
Route a session, set remotes, commit, signal a factory gate, or hand off production`agent-workflow/SKILL.md`
Work on Cloudflare DNS, Workers, Pages, domains, or Wrangler`cloudflare/SKILL.md`
Edit Wolves content or assets`wolves-content/SKILL.md`
Edit the /wolves/ teaser, Trailer 1 player, plates, bridge, or end card`wolves-teaser/SKILL.md`
Edit Wolves guardian cards, share pages, or lobby gallery data`guardian-character-cards/SKILL.md`
Do explicitly approved Wolves runtime engineering`wolves-runtime-engineering/SKILL.md`
Write back what I learned before finishing`skill-improvement/SKILL.md`
Create, restructure, or retire a skill`skill-authoring/SKILL.md`

This table is the single hand-curated router for this repository. `skills/INDEX.md` points here; do not maintain a second table.

How to load a skill

Read the skill's YAML front matter first. If description matches your task, read the body. If a fact belongs to another skill or reference, follow the link rather than duplicating it.

Factory context

This repository is part of the Project Bluefin factory. Local authority wins: AGENTS.md and this router are authoritative for paths, boundaries, and commands. projectbluefin/common attaches as a pinned shared-contract sidecar that supplies factory-wide rules; it never overrides local authority.

  • Factory model and cross-repo rules: projectbluefin/common

docs/factory/agentic-model.md

  • Onboarding contract this repo implements: projectbluefin/common

docs/skills/factory-onboarding.md

  • Four human gates and PR evidence rules: projectbluefin/common

docs/skills/human-gates.md

  • Cross-repo learning: open an issue in projectbluefin/common with the

learning, affected component, and evidence. Never edit ublue-os/*.

A missing or unreachable sidecar is degraded mode, not permission to substitute a stale sibling checkout. A sibling checkout on a fork or a feature branch is not the sidecar; fetch canonical projectbluefin/common main or proceed without it.

Every task loop

  1. Preflight — verify repository, remote, branch, dirty paths, and the

skill set you loaded. When the task arrives as a Hive assignment or GitHub issue, resolve it through the API and verify the repository, issue, branch target, and requested scope before editing.

  1. Detect — treat stale, contradictory, missing, or failed guidance as a

repair signal; do not silently fall back.

  1. Repair — make the smallest scoped change inside the approved boundary.
  2. Validate — run the smallest relevant check from

`skills/validation/SKILL.md`.

  1. Write back — update the skill that owns the area in the same commit.

See `skills/skill-improvement/SKILL.md`.

  1. Escalate — stop at the four factory gates (Design, Security, Breakage,

Merge) or a production claim. See `skills/agent-workflow/SKILL.md`.

Supporting references

Skills define procedures. References state current production facts. Keep those roles separate.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/projectbluefin/website/docs/badge)](https://verified-skill.com/skills/projectbluefin/website/docs)

Something wrong with this skill or its page? Tell us.