nav-brief

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were identified in the provided skill files. The Python logic in functions/ambiguity_scorer.py is self-contained and uses safe regular expressions to calculate ambiguity scores.- [PROMPT_INJECTION]: The skill processes untrusted user prompts to generate an intent brief, which serves as an indirect prompt injection surface. This is a low-risk surface inherent to the skill's function of clarifying user intent.
  • Ingestion points: User prompts are parsed by the ambiguity_scorer.py script and interpolated into the Intent Brief template in SKILL.md.
  • Boundary markers: The skill is triggered by a specific NAV-BRIEF reminder block injected into the context.
  • Capability inventory: The skill documentation references the agent's ability to perform file modifications (Edit, Write) and shell commands (Bash).
  • Sanitization: No explicit sanitization or escaping of the user's prompt is performed before rendering the brief; however, the skill's mandatory rule to wait for user confirmation before any destructive action serves as a strong human-in-the-loop control.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 03:53 PM
Security Audit — agent-trust-hub — nav-brief