The Agent Skills Directory

[SAFE]: No malicious patterns, exfiltration attempts, or obfuscation were detected. The skill behaves as described, providing a research-focused workflow.
[COMMAND_EXECUTION]: The skill utilizes the 'church' CLI utility to perform repository analysis, claim scanning, and ledger management. These commands operate on the local repository root provided by the user and are consistent with the skill's stated purpose.
[SAFE]: The skill processes project documentation and reports as untrusted data inputs for claim validation. While this presents an indirect prompt injection surface (Category 8), the structured use of specialized scanning tools and classification steps follows safe data handling practices for this context.

church-harden