diagnose

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill outlines a standard debugging process and includes a benign bash script for user interaction.
  • [COMMAND_EXECUTION]: The skill instructs the agent to create and execute various diagnostic tools, including shell commands, test suites, and browser automation scripts, which is consistent with its stated purpose.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from test outputs and logs. 1. Ingestion points: Test outputs, CLI diffs, and network logs mentioned in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Modifying files and executing shell commands (curl, git) and scripts (Playwright). 4. Sanitization: Absent. This is considered a low-risk surface inherent to debugging tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 03:01 AM
Security Audit — agent-trust-hub — diagnose