agent-development
Guidance for creating Claude Code subagents: frontmatter, system prompts, triggering conditions, tools, and colors.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for agent-development, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
Part of the plugin-dev plugin, this skill guides authoring Claude Code agents/subagents, covering agent frontmatter, when-to-use descriptions, system-prompt design, triggering conditions, tool allowlists, and agent colors. It bundles example agent definitions, reference docs (system-prompt-design, triggering-examples), and a validate-agent.sh script run via bash to lint agent files.
Key features and capabilities
- Agent frontmatter and system-prompt design
- Triggering-condition examples
- validate-agent.sh linter script
Use cases
- Create a new Claude Code subagent
- Fix an agent that never triggers correctly