twitch-category-opportunity-research
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local bash script (
scripts/crawlora.sh) to perform network requests. The script implements strong security controls, including using temporary configuration files for API keys to prevent exposure in process lists and whitelisting permitted API paths. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted metadata from Twitch (such as stream titles and tags) that could potentially contain adversarial instructions.
- Ingestion points: Data is fetched via the
crawlora.shscript from the/twitch/streamsand/twitch/searchendpoints. - Boundary markers: The instructions do not specify any delimiters or warnings for the agent to ignore instructions embedded in the external data.
- Capability inventory: The skill's capabilities are limited to making GET requests via the provided script and analyzing the returned data; it does not have the ability to write to the filesystem or execute arbitrary code.
- Sanitization: The helper script ensures that all query parameters are URL-encoded before they are included in the API request.
Audit Metadata