The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests external data by running git diff to provide code reviews.\n- Ingestion points: git diff and git diff --stat output processed in the 'Check-ins' section of SKILL.md.\n- Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are defined within the skill.\n- Capability inventory: The skill is capable of running git commands (diff, commit), spawning a subagent for review, and communicating via AskUserQuestion.\n- Sanitization: No explicit sanitization, validation, or filtering of the ingested code content is performed.\n- [SAFE]: The skill implements standard collaborative programming patterns such as ping-pong and driver/navigator loops. No malicious code execution, data exfiltration, or obfuscation techniques were identified.

pair-with-me