The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified during the analysis of the skill's instructions and scripts.
[COMMAND_EXECUTION]: The skill executes git commands via local Python scripts (analyze_changes.py and audit_claudemd.py). These executions are for legitimate purposes such as analyzing commit history and listing tracked files. The implementations use subprocess.run with argument lists (shell=False), which is the standard security practice to prevent command injection.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from CLAUDE.md and git diffs to determine documentation updates. While this represents an ingestion point for untrusted data, the skill's primary function is to summarize and organize technical documentation, and no patterns were found that would allow this data to compromise the agent's core instructions.

continuous-claudemd-updates