The Agent Skills Directory

[SAFE]: The skill primarily provides structural guidance and per-folder rules for managing engineering documents like RFCs, ADRs, and implementation specs. No unauthorized network operations, credential access, or malicious code execution patterns were found.
[INDIRECT_PROMPT_INJECTION]: The skill defines agent behaviors that respond to document states (e.g., scanning existing records for duplicates or surfacing stale drafts). This creates a surface where external content in markdown files is processed by the agent. However, these behaviors are standard for documentation management, and no exploitable capability chain was identified.

software-lifecycle