principle-prove-it-works

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill outlines best practices for task verification, focusing on inspecting actual artifacts such as git diffs, file contents, and runtime behavior rather than relying on summaries or proxies.\n- [INDIRECT_PROMPT_INJECTION]: The skill highlights a surface where the agent processes external data (code, diffs, outputs) for verification. This ingestion of untrusted data is a standard part of the agent's workflow but is noted as a potential surface for indirect prompt injection.\n
  • Ingestion points: Inspecting git diffs, file contents, and runtime behavior as specified in SKILL.md.\n
  • Boundary markers: None specified in the instructions.\n
  • Capability inventory: Relies on the agent's existing capabilities to read files and execute shell commands.\n
  • Sanitization: No specific sanitization or filtering logic is provided within the skill instructions.\n- [COMMAND_EXECUTION]: The skill encourages the use of deterministic scripts to verify work, leveraging the agent's inherent capabilities to execute local commands for validation purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 03:44 PM
Security Audit — agent-trust-hub — principle-prove-it-works