The Agent Skills Directory

[COMMAND_EXECUTION]: The maintenance script scripts/search_rank_audit.py utilizes subprocess.run to execute the GitHub CLI (gh) for checking search rankings. This operation is restricted to public repository data from a well-known service (GitHub) and uses controlled inputs within a developer-facing utility.\n- [PROMPT_INJECTION]: The skill processes user-supplied text, which represents a potential surface for indirect prompt injection. This was evaluated for risk factors.\n * Ingestion points: User-provided Chinese text drafts (processed via SKILL.md).\n * Boundary markers: None identified.\n * Capability inventory: Command execution via gh in maintenance scripts.\n * Sanitization: No explicit input sanitization or boundary delimiters were identified for user-provided content.\n- [DATA_EXFILTRATION]: No sensitive data exposure or exfiltration patterns were found. The skill does not access environment secrets or private system files, and its network activity is confined to querying public data via established tools.

remove-ai-flavor