sitrep
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project metadata which can influence the agent's recommended actions.\n
- Ingestion points: Project state files located in
.lev/pm/workstreams/*/trace/*.yaml.\n - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within these YAML files.\n
- Capability inventory: The skill produces an 'Exec Menu' and 'Inline shortcuts' (e.g.,
expand <n>) that drive the agent's next steps and command execution.\n - Sanitization: The skill does not specify any sanitization or validation of the data retrieved from the project files before rendering it to the user or agent.\n- [COMMAND_EXECUTION]: The skill dynamically generates command suggestions (e.g.,
/exec,/propose) and protocol-based actions (e.g.,expand://) based on the contents of files it reads from the workspace.
Audit Metadata