The Agent Skills Directory

[SAFE]: The skill provides an purely instructional and architectural framework for integrity checking. It lacks any active code that performs network operations, sensitive file access, or remote code execution.\n- [SAFE]: The provided Python script scripts/check-output.py is a simple utility for validating markdown report structures. It uses the sys and pathlib modules safely to read a file provided via command-line argument and check for required substrings.\n- [SAFE]: The skill mentions external tools like python3 scripts/audit-artifacts.py and python3 scripts/generate-skills-lock.py but these are not included in the skill package itself and are described as project-specific maintenance tools. The documentation correctly identifies the security boundary, noting that SHA-256 hashes prove integrity (prevention of drift/tampering) rather than code safety or trust.\n- [SAFE]: External links in the documentation point to official resources for the NIST FIPS 180-4 standard and reproducible-builds.org, both of which are trusted information sources.

skills-lock