verification-before-completion

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it requires the agent to ingest and act upon data from external sources without sufficient safety boundaries. \n
  • Ingestion points: The agent is instructed to read output from tests, build commands, and GitHub Actions logs via the Bash tool and github MCP in SKILL.md. \n
  • Boundary markers: The instructions lack requirements for delimiters or warnings to ignore instructions that might be embedded in these external outputs. \n
  • Capability inventory: The skill utilizes the Bash and Read tools, providing the agent with the ability to execute shell commands and read project files. \n
  • Sanitization: No sanitization or validation of the external content is mentioned before the agent processes it. \n- [COMMAND_EXECUTION]: The skill frequently uses the Bash tool to run standard development operations such as mvn test, npm test, and gh CLI commands. These are context-appropriate for a verification specialist and do not include high-risk patterns like unauthorized remote code execution or privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 01:54 PM
Security Audit — agent-trust-hub — verification-before-completion