diagnose

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill contains standard software development and debugging instructions. It does not include executable code, external dependencies, or network operations. No indicators of prompt injection, obfuscation, or persistence were found.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external data like issue reports and repository instructions. However, this is inherent to the primary purpose of a diagnostic tool and no exploitable pattern was detected. 1. Ingestion points: Reads CONTEXT.md, glossary, ADRs, issue/spec context, and repository instructions. 2. Boundary markers: Absent. 3. Capability inventory: File system write access for fixes and regression tests, and command execution for running experiments. 4. Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 08:50 AM
Security Audit — agent-trust-hub — diagnose