mission
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script located at
mission-spec/scripts/validate_spec.pyto validate specification files. This execution relies on a user-provided or contextually-derived file path argument.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data to determine routing logic, creating a surface for indirect instructions to influence agent behavior.\n - Ingestion points: The skill ingests and analyzes the headers of CSV files and the frontmatter/metadata of Markdown documents (e.g.,
mission: specandstatusfields) inSKILL.md.\n - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard instructions embedded within the processed CSV or Markdown content.\n
- Capability inventory: The agent has the capability to execute a local script (
validate_spec.py) and perform routing to other functional skills based on the content of the analyzed files.\n - Sanitization: The instructions do not specify any sanitization, escaping, or structural validation beyond checking for standard headers and specific metadata fields.
Audit Metadata