safety-management

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted data from project-specific files such as specs-quality.json and directory.json to generate safety alerts, creating an indirect prompt injection surface.
  • Ingestion points: Data enters the agent context from safety-log.json, specs-quality.json, plans-spatial.json, and directory.json as described in the integration section of SKILL.md.
  • Boundary markers: There are no explicit delimiters or instructions to the model to ignore embedded commands when interpolating this external data into prompts.
  • Capability inventory: The skill is designed to perform file-write operations to safety-log.json but does not contain capabilities for network requests or arbitrary command execution.
  • Sanitization: The skill documentation does not outline any procedures for validating or sanitizing the content retrieved from the external data files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 06:53 AM
Security Audit — agent-trust-hub — safety-management