commet-cli
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill determines its operational capabilities by reading local manifest files and parsing the output of the installed CLI tool. This ingestion of external data creates a surface for indirect prompt injection if the local environment or vendor tools are compromised.
- Ingestion points: Output of commet --output agent, commet doctor, and the node_modules/@commet/node/docs/manifest.json file.
- Boundary markers: The instructions explicitly require the agent to state exact targets and wait for user authorization before performing any local file writes or remote resource mutations.
- Capability inventory: The skill can perform repository writes, link organizations, push configuration, and create remote resources via the CLI.
- Sanitization: The skill mandates that the agent never print secret values or credentials and warns the user before authentication flows.
Audit Metadata