Agent Claude Code
Architecture planning and architecture review for the declarative agent platform. Use to design or evaluate changes that affect layers, lifecycles, or contracts. Read-only — does not implement code unless explicitly asked.
Agent Claude Code
Performs senior-level code review on diffs/PRs for the agent platform, focused on architecture boundaries, lifecycle/state, security, and tests. Read-only; outputs a structured review.
Agent Claude Code
Updates README, docs/, ADRs, and skill files for the agent platform. Keeps documentation honest and synchronized; never claims unimplemented features work.
Agent Claude Code
Plans and writes pytest tests for the agent platform. Behavior-focused, mocks all external systems, adds negative/security tests. Never calls real external services.
Skill Claude CodeCodex
Guard the project's architecture invariants. Use to evaluate any change affecting layers, boundaries, lifecycles, or contracts, and to decide whether an ADR is required.
Skill Claude CodeCodex
Senior-level code review for the agent platform — architecture and boundaries first, then correctness, security, testability, and simplicity. Outputs a structured review.
Skill Claude CodeCodex
Keep the repository's documentation accurate, honest, and synchronized with the code. Use when editing docs, README, AGENTS.md, or ADRs.
Skill Claude CodeCodex
Work safely with Git — keep changes scoped and reviewable, never overwrite the user's work, and produce clear history. Use before editing, branching, staging, or committing.
Skill Claude CodeCodex
Working on Python plugin tools, MCP server connections, tool invocation, or tool tracing. Primary task is tasks/0007-mcp-tools-and-permissions.md.
Skill Claude CodeCodex
Foundational rules and layout for the declarative agent platform. Read this before any change, regardless of area; it anchors every other skill.
Skill Claude CodeCodex
Working on prompt templates, template loading/caching, or per-request rendering. Primary task is tasks/0005-prompt-rendering.md.
Skill Claude CodeCodex
Change the shape of code without changing behavior — safely, in small steps, preserving the project's layers and public contracts.
Skill Claude CodeCodex
Working on RuntimeEngine, ExecutionContext, agent execution, or the request lifecycle. Primary tasks are tasks/0003-compiled-agent-graph.md and tasks/0004-runtime-engine.md.
Skill Claude CodeCodex
The standard for writing Python here — small, typed, explicit, testable modules with side effects pushed to the edges. Use when implementing or designing any Python code.
Skill Claude CodeCodex
Working on resolver plugins, the fixed access plugin contract, mapping resolver outputs into ExecutionContext, or protected-node filtering. Primary task is tasks/0006-sidecar-auth-context.md.
Skill Claude CodeCodex
How to create or restructure a skill so the .ai/skills/ system stays consistent, operational, and trustworthy. A skill is a short, repeatable playbook — not an essay.
Skill Claude CodeCodex
How to write and run fast, deterministic, behavior-focused pytest tests that never touch real external systems. Use whenever adding or changing behavior, or fixing a bug.
Skill Claude CodeCodex
Loading agent.yml, defining schema models, and validating the specification (definitions, hierarchy, security). Primary task is tasks/0002-yaml-schema-and-validation.md.
Instructions file CodexOpenCode
Instructions for extra-org/extra, covering agents.md, 1. project mission, 2. core architecture pipeline, 3. non-negotiable architecture rules and 4. repository structure.
Instructions file
Instructions for extra-org/extra, covering claude.md, naming, skills, commands and guardrails.