developer-oss-research
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches GitHub and Chrome Web Store data from the official Crawlora API at
api.crawlora.net. This is the intended behavior for researching developer resources and uses the vendor's own API service. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (GitHub and Chrome Web Store) via the Crawlora API. This third-party content, such as repository descriptions or extension reviews, could contain instructions intended to influence the agent's behavior.
- Ingestion points: API responses from
https://api.crawlora.net/api/v1fetched viascripts/crawlora.sh. - Boundary markers: Data is returned as structured JSON, but there are no explicit instructions to the agent to disregard potential commands embedded within that data.
- Capability inventory: The skill is restricted to making
GETrequests to a specific, validated list of API endpoints; it lacks arbitrary command execution or file-write capabilities. - Sanitization: The helper script performs thorough input validation and URL encoding for parameters, though it does not sanitize the content of the data returned by the API.
Audit Metadata