Menü
Use when creating a new ra recipe — a complete agent configuration with skills, middleware, and config.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
| name | write-recipe |
| description | Use when creating a new ra recipe — a complete agent configuration with skills, middleware, and config. |
A recipe is a self-contained agent configuration. Copy the directory, run it — everything works.
recipes/<name>/
ra.config.yaml # Agent configuration
README.md # What it does, how to run it
skills/ # Bundled skills (MUST be self-contained)
<skill-name>/
SKILL.md # YAML frontmatter + instructions
scripts/ # Optional: run at activation, stdout → context
references/ # Optional: injected as context
middleware/ # Optional: lifecycle hooks
<hook>.ts
Define the agent — one sentence: "A code reviewer that reads diffs and gives structured feedback."
Write skills — skills/<name>/SKILL.md:
name, descriptionscripts/ for dynamic context (git diff, env vars, etc.)Write config — ra.config.yaml:
agent:
provider: anthropic
model: claude-sonnet-4-6
skillDirs:
- ./skills # always relative to recipe dir
maxIterations: 10 # 10 for focused tasks, 50+ for exploratory
# Optional:
mcp:
servers:
- name: github
transport: stdio
command: npx
args: ["-y", "@modelcontextprotocol/server-github"]
middleware:
afterModelResponse:
- ./middleware/token-budget.ts
Add middleware (optional) — common hooks: token budget, tool filtering, logging
Write README — install, configure, run
Test — cd recipes/<name> && ra --config ra.config.yaml "test prompt"
skills/, use skillDirs: [./skills]../../skills is wrong)recipes/coding-agent/ and recipes/code-review-agent/ for referenceSystem design and architecture decisions. Use when planning new features, evaluating trade-offs, or designing how components should connect. Proposes 2-3 approaches and recommends one.
Smart commit workflow. Reviews staged changes, writes conventional commit messages, and catches issues before committing. Use when ready to commit work.
Quality review of completed work. Use after making changes and before claiming completion. Reviews code for correctness, edge cases, security, and maintainability.
Systematic debugging workflow. Use when diagnosing bugs, test failures, or unexpected behavior. Follows a rigorous reproduce → isolate → hypothesize → fix → verify cycle.
Systematic multi-agent research. Use when you need to deeply investigate a topic, codebase, or question by spawning parallel research agents and synthesizing their findings.
Deep code explanation. Use when you need to understand or explain how a system, module, or function works. Traces data flow, maps dependencies, and explains design decisions.