narrative-baseline-mapper
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill proactively addresses the risk of indirect prompt injection by including explicit defensive instructions to treat all scraped and pasted content as untrusted and to never follow instructions embedded within that data.
- [COMMAND_EXECUTION]: The skill utilizes local helper scripts located in the project's
scripts/connectors/directory (firecrawl.py,wayback.py). These scripts are used for intended core functionality such as web scraping and historical data retrieval. - [EXTERNAL_DOWNLOADS]: The skill interacts with external web surfaces and the Wayback Machine to fetch current and historical messaging. These operations target public data and use established services as part of its documented purpose.
Audit Metadata