verify-against-spec
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for test runners, specifically mentioning
npx vitest run --reporter=verboseandpytest -vto gather evidence for the verification report. - [DYNAMIC_EXECUTION]: To address criteria not covered by existing tests, the skill directs the agent to generate and execute custom scripts (e.g.,
verify-gaps.*) within the.scratchpad/directory. This involves the runtime creation and execution of executable logic based on the agent's interpretation of the spec. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process implementation code and specification documents, which are untrusted external inputs that could contain malicious instructions.
- Ingestion points: Reading implementation source code and acceptance criteria from specifications provided in the workspace.
- Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from accidentally interpreting embedded instructions in the analyzed files.
- Capability inventory: The agent has the ability to execute shell commands, run test suites, and execute generated scripts, which could be leveraged if an injection occurs.
- Sanitization: No sanitization or validation steps are defined for the content of the specifications or implementation files being analyzed.
Audit Metadata