51mee-position-parse

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by interpolating untrusted job description text into a model prompt.
  • Ingestion points: User-provided text is inserted into the {职位描述文本} placeholder within the prompt template in SKILL.md.
  • Boundary markers: The skill attempts to isolate user input using markdown code blocks (text ... ), which is a weak delimiter that can be escaped by adversarial content.
  • Capability inventory: No dangerous tools, subprocess calls, network operations, or file-system writing capabilities were identified in the skill definition or referenced files.
  • Sanitization: There is no evidence of input validation, filtering, or escaping logic to prevent malicious instructions within the job description from influencing the agent's behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 08:37 AM
Security Audit — agent-trust-hub — 51mee-position-parse