verification

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute various npm scripts and local binaries to verify code quality. Commands include npm run compile, npm run test, npm run check:actions, and direct execution of Jest via node 'node_modules/.bin/jest'. These are routine development operations for managing a Node.js-based repository.
  • [INDIRECT_PROMPT_INJECTION]: As the skill is designed to verify arbitrary code changes, it presents a surface for indirect prompt injection if the code being verified contains malicious instructions intended to influence the agent's behavior during the verification process.
  • Ingestion points: Code changes in the repository, GitHub workflow files, and test files.
  • Boundary markers: No specific delimiters or warnings are used to separate untrusted code content from agent instructions.
  • Capability inventory: The agent has the ability to execute shell commands and read/write files as part of the verification process.
  • Sanitization: The skill does not specify any sanitization or validation steps for the code content being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 10:06 AM
Security Audit — agent-trust-hub — verification