muse-review
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local environment which could contain malicious instructions.
- Ingestion points: The skill reads the "current environment, including uncommitted work" and uses it to form a review brief.
- Boundary markers: No delimiters or instructions to ignore embedded instructions are specified for the ingested content.
- Capability inventory: The skill uses an orchestration CLI to perform network operations and shell-based tasks across all its functions.
- Sanitization: There is no mention of sanitizing or escaping the environment data before passing it to the external model.
- [COMMAND_EXECUTION]: The skill relies on an "orchestration CLI" to perform its primary logic, requiring the agent to execute shell commands to interact with providers, check model listings, and retrieve thread details.
Audit Metadata