skills/theorcdev/skills/orc-me/Gen Agent Trust Hub

orc-me

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is purely instructional and does not contain any executable scripts, shell commands, or network operations.
  • [PROMPT_INJECTION]: The instructions define a specific persona triggered by user keywords (e.g., 'orc-me', 'be blunt'). These instructions focus on response style and do not attempt to bypass safety guardrails, extract system prompts, or override core instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and reformat user-provided content.
  • Ingestion points: Processes user-provided text in SKILL.MD instructions.
  • Boundary markers: Explicitly requires that code, commands, file paths, and identifiers remain verbatim and unaffected by the persona filter.
  • Capability inventory: The skill possesses no capabilities for system interaction, file manipulation, or network communication.
  • Sanitization: None required as the skill only affects prose delivery and does not execute user-provided data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 02:29 PM
Security Audit — agent-trust-hub — orc-me