harness-spec-anatomy

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No security issues detected. The skill serves as a reference guide for authoring Hedera Harness recipes and provides a shell script for local validation.
  • [COMMAND_EXECUTION]: The skill includes a shell script (scripts/check-spec.sh) that validates project configuration files. It uses standard system utilities like grep, awk, and sed, and may execute the hedera-harness CLI tool if it is available in the user's environment. These operations are conducted locally and are transparently documented as part of the skill's linting functionality.
  • [INDIRECT_PROMPT_INJECTION]: The validation script reads and analyzes user-authored project files (such as PRDs and recipe YAMLs) to ensure cross-file coherence. This analysis is limited to text-matching and basic parsing, with no evidence of dangerous interpolation or execution of instructions contained within the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 01:06 PM
Security Audit — agent-trust-hub — harness-spec-anatomy