The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill consists entirely of Markdown documentation providing organizational structure and best practices.
[NO_CODE]: The skill does not include any executable scripts, source code, binaries, or platform-specific command injections (!command).
[CREDENTIALS_SAFE]: The skill includes a security guardrail (references/guardrails.md) that explicitly instructs the agent not to expose secrets or API keys and to use appropriate secret management like .env files.
[INDIRECT_PROMPT_INJECTION]: While the skill functions as an orchestrator that processes user intent to select other skills, it emphasizes that user instructions take priority and includes boundaries requiring manual verification of all outputs before task completion.

claukit