empirical-proof

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill directs the agent to identify and execute shell commands found in project-specific documentation (e.g., CONTRIBUTING.md, AGENTS.md, or manifest scripts).
  • Ingestion points: Documentation files and manifest scripts within the current project directory (SKILL.md).
  • Boundary markers: The instructions lack specific guidance to inspect or sanitize discovered commands for malicious patterns before execution, though they suggest asking the user if a command is unclear.
  • Capability inventory: Shell execution is required to run the build, test, and linting commands described (SKILL.md).
  • Sanitization: No explicit sanitization or validation of the shell commands retrieved from project files is mentioned.
  • [COMMAND_EXECUTION]: The skill's fundamental requirement is the execution of shell commands (e.g., npm test, npx jest) to generate empirical evidence of task completion.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 09:03 AM
Security Audit — agent-trust-hub — empirical-proof