gsc
Warn
Audited by Gen Agent Trust Hub on Feb 14, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- CREDENTIALS_UNSAFE (MEDIUM): The skill hardcodes absolute paths to a specific user directory (/Users/admin/clawd/skills/gsc/.env) and uses 'source' to load credentials into the environment. This exposes local system structure and relies on insecure plaintext storage of secrets.\n- COMMAND_EXECUTION (LOW): The skill relies on shell execution of a local Python script. While these scripts are part of the skill package, the hardcoded paths and use of 'source' for environment management are insecure practices.\n- PROMPT_INJECTION (LOW): Category 8 (Indirect Prompt Injection): The skill ingests untrusted data from the Google Search Console API.\n
- Ingestion points: Search queries and URL data are fetched from the GSC API via 'gsc_query.py'.\n
- Boundary markers: No markers or delimiters are present to separate untrusted API data from the agent's instruction context.\n
- Capability inventory: Executes Python scripts to query and display search analytics.\n
- Sanitization: No sanitization or filtering of API-returned content is defined in the skill documentation.
Audit Metadata