skills/thaitype/chief/chief-rule/Gen Agent Trust Hub

chief-rule

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by reading and writing files in the .chief/_rules/ directory that are intended to govern agent behavior. Malicious content in these files could override agent constraints in future interactions.
  • Ingestion points: User responses during the interview process and existing rule files read during collision handling in SKILL.md.
  • Boundary markers: No specific delimiters are used to wrap user-provided rule statements or motivation to distinguish them from agent instructions.
  • Capability inventory: File system operations including directory creation and markdown file writing.
  • Sanitization: User content is drafted into markdown files without escaping or content validation.
  • [COMMAND_EXECUTION]: The skill instructs the agent to perform file system management tasks, specifically creating the .chief/ directory structure and writing configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 12:35 AM
Security Audit — agent-trust-hub — chief-rule