geo-state-report
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- COMMAND_EXECUTION (SAFE): The skill executes a local Python script (
geo-state-report/scripts/brightdata-geo.py) to perform its core data collection task. This behavior is consistent with its stated purpose.\n- PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection as it ingests untrusted data from an external API (viaresults.json) and uses it to generate analysis and a customized HTML report.\n - Ingestion points:
results.json(read after script execution)\n - Boundary markers: None specified in the instructions.\n
- Capability inventory: Executes local Python scripts and writes HTML files to the filesystem.\n
- Sanitization: No sanitization or validation of the ingested JSON data is mentioned before it is processed by the AI.\n- CREDENTIALS_UNSAFE (SAFE): The skill requires an API key but includes explicit instructions to help the user avoid leaking the key in the conversation, specifically recommending environment variables.
Audit Metadata