The Agent Skills Directory

[COMMAND_EXECUTION]: Executes system commands and local Python scripts to manage the skill's operational environment. It uses openssl for directory naming, mkdir for temporary storage in /tmp/tunan/explain/, and executes a script outside its own directory (../../scripts/repo-profile-cache.py) to manage project metadata caching. It also utilizes cp to export generated artifacts to user-defined paths.\n- [DATA_EXFILTRATION]: Supports sending generated artifacts to external services. The skill provides destination sub-flows to publish markdown content to the 'Proof' web API and to share documents with 'Thinkroom' using detected skill or CLI capabilities.\n- [PROMPT_INJECTION]: Exhibits an indirect prompt injection surface due to the ingestion of untrusted repository data. \n
Ingestion points: Reads git history, pull request information, and project documentation (plans, solution docs) to generate explainers.\n
Boundary markers: Lacks explicit delimiters or 'ignore' instructions for the ingested data during the orientation and composition phases.\n
Capability inventory: Possesses command execution privileges (Python/shell) and the ability to perform network operations (API publishing).\n
Sanitization: No evidence of sanitization or filtering of external repository content before it is interpolated into the agent's context.

explain