verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to identify and run shell commands (e.g., tests, builds, linters) to verify work status. This is the core functionality and uses the agent's existing shell tools.
- [INDIRECT_PROMPT_INJECTION]: The agent is instructed to read the full output of these verification commands. This creates an attack surface where malicious data within a codebase (e.g., crafted test failures or linter messages) could be used to inject instructions into the agent's context.
- Ingestion points: Command outputs from tests, builds, and linters referenced in SKILL.md.
- Boundary markers: None defined; the agent is told to read the 'FULL output'.
- Capability inventory: Execution of shell commands identified as verification steps.
- Sanitization: The skill does not provide mechanisms for sanitizing or escaping the command output before processing.
Audit Metadata