skills/dmytri/shipshape/crew/Gen Agent Trust Hub

crew

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions require the agent to execute shell commands to reproduce failing tests and verify implementation changes. This is essential for the stated purpose of code implementation and testing.- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests externally provided test targets and Gherkin specifications to guide its modifications.
  • Ingestion points: Reads user or Quartermaster-supplied verification targets and feature files in SKILL.md.
  • Boundary markers: Includes negative constraints (e.g., 'Do not change specs', 'Do not broaden scope') but lacks technical delimiters for input data.
  • Capability inventory: Can modify production source code and execute testing commands.
  • Sanitization: No sanitization of ingested test/spec content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 06:56 PM
Security Audit — agent-trust-hub — crew