yyyyy
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to interact with the OpenSpec tool and verify project health.
- Evidence: Use of
openspec listto check for active changes and the requirement to run project test suites. - [INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted content from code and documentation files, which constitutes a vulnerability surface for embedded instructions.
- Ingestion points: Files within the project root and the
openspec/directory are read and analyzed. - Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within the data being reviewed.
- Capability inventory: The skill has the capability to read files, execute shell commands (listing changes, running tests), and modify files (attempting to fix bugs).
- Sanitization: There is no evidence of content sanitization or validation for the code and documentation processed.
Audit Metadata