The Agent Skills Directory

[SAFE]: The skill is primarily composed of markdown playbooks and structured routing data (CSV) used to guide the agent's behavior during design reviews. No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected.
[NO_CODE]: The skill does not include any executable scripts for the agent's runtime environment. It contains a developer-facing shell script for static validation (evals/run-static-checks.sh) that uses standard local utilities like grep, wc, and head to ensure file integrity and structure.
[EXTERNAL_DOWNLOADS]: All external references in the skill metadata and playbooks point to reputable sources, including official documentation and research papers from trusted organizations like Anthropic, Cloudflare, and the Linux Foundation. No downloads from untrusted or unverified sources were found.

agent-experience