skills/kntnt/skills/brief/Gen Agent Trust Hub

brief

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a stylistic utility that provides guidelines for response conciseness. It does not execute code, perform network requests, or access sensitive files. The instructions in references/mode.md include a specific rule (Rule 8) to preserve the full content of security warnings and error reports, ensuring safety information is not truncated.\n- [INDIRECT_PROMPT_INJECTION]: The skill implements an 'Invocation Envelope' pattern that allows contextual instructions after a separator (--). While this provides an entry point for user-supplied data, the impact is negligible due to the lack of skill capabilities.\n
  • Ingestion points: User-provided text appearing after the -- separator in commands like /brief on -- instruction.\n
  • Boundary markers: The skill uses a reserved separator (--) to clearly delimit formal commands from contextual guidance.\n
  • Capability inventory: The skill has no tool access and explicitly sets disable-model-invocation: true.\n
  • Sanitization: Instructions are processed as standard conversational context by the agent's underlying safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:18 PM
Security Audit — agent-trust-hub — brief