add-experience

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data to update professional materials, creating a potential attack surface.
  • Ingestion points: User input provided via "自由文本" (free text) or "结构化描述" (structured descriptions) is parsed in Step 1 and used to populate project experience files.
  • Boundary markers: The instructions do not define boundary markers or explicit instructions for the agent to ignore or sanitize embedded commands within the user's project descriptions.
  • Capability inventory: The skill has capabilities to read, write, and create files and directories within the materials/ path, including timeline.md, core-projects.md, and base-resume.md (via Step 4).
  • Sanitization: There is no mention of sanitizing or escaping the user-provided text before it is interpolated into the markdown files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 05:04 PM
Security Audit — agent-trust-hub — add-experience