The Agent Skills Directory

[SAFE]: The skill consists of Markdown documentation (SKILL.md), a Python script for internal consistency checks (contract_check.py), a JSON file containing evaluation scenarios (held-out-scenarios.json), and a shell script to run evaluations (run-eval.sh).
[SAFE]: The Python script contract_check.py performs a simple string-matching check against SKILL.md to ensure required phrasing is present. It does not use any dangerous functions like eval() or exec(), nor does it perform network or sensitive file operations.
[SAFE]: The shell script run-eval.sh executes local validation scripts and does not perform any remote downloads or command execution from external sources.
[SAFE]: All content in the files matches the stated purpose of defining SEO research gates and content planning workflows. There are no signs of obfuscation, exfiltration, or persistence mechanisms.

01-audience-topic-research