harness-codebase-reverse-spec

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local shell script (scripts/scan_leakage.sh) to identify technical jargon in the generated specification. The script performs local text processing using standard tools and does not interact with the network or access sensitive system files.
  • [EXTERNAL_DOWNLOADS]: No remote scripts, external packages, or dependencies are downloaded or executed. All logic and resources are contained within the skill's local files.
  • [DATA_EXFILTRATION]: The skill accesses the local codebase for analysis purposes. There are no patterns suggesting unauthorized data transmission, network exfiltration, or hardcoded credentials.
  • [PROMPT_INJECTION]: The instructions focus on structured analysis and business logic extraction. No patterns intended to bypass safety guardrails or override system prompts were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 08:17 PM
Security Audit — agent-trust-hub — harness-codebase-reverse-spec