# Claude Code Instructions

## Task Master AI Instructions
**Import Task Master's development workflow commands and guidelines, treat as if import is in the main CLAUDE.md file.**
@./.taskmaster/CLAUDE.md

## Skill routing

When the user's request matches an available skill, ALWAYS invoke it using the Skill
tool as your FIRST action. Do NOT answer directly, do NOT use other tools first.
The skill has specialized workflows that produce better results than ad-hoc answers.

Key routing rules:
- Product ideas, "is this worth building", brainstorming → invoke office-hours
- Bugs, errors, "why is this broken", 500 errors → invoke investigate
- Ship, deploy, push, create PR → invoke ship
- QA, test the site, find bugs → invoke qa
- Code review, check my diff → invoke review
- Update docs after shipping → invoke document-release
- Weekly retro → invoke retro
- Design system, brand → invoke design-consultation
- Visual audit, design polish → invoke design-review
- Architecture review → invoke plan-eng-review
- Save progress, checkpoint, resume → invoke checkpoint
- Code quality, health check → invoke health

## Resource types

`mcpctl` resource cheatsheet:

- `server` — MCP server definition
- `instance` — running container (immutable, replicas-managed)
- `secret` / `secretbackend` — credentials
- `template` — reusable server blueprint
- `project` — workspace grouping servers, prompts, agents
- `llm` — server-managed LLM provider (api key + endpoint)
- `agent` — LLM persona pinned to one Llm; project attach surfaces project Prompts as system context, project MCP servers as tools, and exposes the agent itself as an MCP virtual server (`agent-<name>/chat`). See `docs/agents.md`, `docs/chat.md`.
  Tool access is bounded by **whoever invoked the run**, not by the agent's
  memberships — knowledge travels, action is gated. See `docs/agent-authority.md`.
- `prompt` / `promptrequest` — curated content / legacy pending proposal (use `proposal` for new work).
- `skill` — Claude Code skill bundle (SKILL.md + files + typed metadata). Materialised onto disk by `mcpctl skills sync`. See `docs/skills.md`.
- `proposal` — generic pending proposal queue, replaces `promptrequest`. Covers both prompts and skills. See `docs/proposals.md`. Triage via `mcpctl review`.
- `revision` — append-only audit + diff log shared by prompts and skills. Auto-bumps semver on save. See `docs/revisions.md`.
- `epic` / `epictask` / `task` — the work tracker: vision → roadmap node → work item.
  Tasks carry a risk level that decides which review gates they must pass, and
  accepting one **merges its PR**. See `docs/epics.md`.
  **`tasks` = work items**; the inference queue is `inference-tasks` / `itasks`.
- `gitprovider` / `repo` — git forge connection + attached repositories. Discovery
  and import via `mcpctl repos discover|import`; a forge connection is what makes
  merge-on-accept possible. See `docs/gitproviders.md`.
- `rbac` — access control bindings
- `mcptoken` — bearer credentials for HTTP-mode mcplocal