commet
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the local
commetCLI tool (viacommet doctor) to gather diagnostic evidence about the project integration. This is an intended vendor-provided utility for project verification and status reporting. - [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to ingest data from local repository files, including dependency manifests, lockfiles, documentation artifacts within installed SDK packages, and generated type definitions.
- Ingestion points: Artifacts located in
node_modules,vendor/, or JAR resources (e.g.,docs/manifest.json). - Boundary markers: The skill lacks explicit prompt delimiters for this content but instructs the agent to treat the installed SDK as the definitive contract.
- Capability inventory: The agent can read files, execute the
commetCLI, and perform code edits. - Sanitization: No specific sanitization is mentioned for the ingested local documentation, representing a standard ingestion surface for development-focused skills.
Audit Metadata