github-repo-browser-read-only
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not contain any executable code (e.g., scripts, binaries); it only defines action schemas and instructions for remote tool calls hosted by AgentPMT.
- [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection as it facilitates the ingestion of external data from GitHub repositories (e.g., code files and READMEs).
- Ingestion points: The
get_fileandget_readmeactions ingest external text directly into the agent context. - Boundary markers: No specific delimiters or "ignore instructions" warnings are defined in the tool's schema to wrap external content.
- Capability inventory: The tool is explicitly read-only and does not possess capabilities to execute local commands, write files, or perform arbitrary network operations beyond its defined scope.
- Sanitization: There is no evidence of content sanitization or validation for the text retrieved from the repositories.
Audit Metadata