review-harness-spec

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a validation script (check-spec.sh) and the hedera-harness CLI to perform the audit and run the harness. These tools are local and provided by the same vendor as the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project configuration files, which could contain malicious instructions aimed at influencing the audit report or subsequent command execution.
  • Ingestion points: .harness/spec.yaml, .harness/prd.md, .harness/validators/static.json, and .harness/validators/yarn.json (SKILL.md).
  • Boundary markers: Absent.
  • Capability inventory: Shell script execution via bash and the hedera-harness CLI tool (SKILL.md).
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 01:07 PM
Security Audit — agent-trust-hub — review-harness-spec