wechat-cover
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from an external API, creating a vulnerability surface for indirect prompt injection via article titles.\n
- Ingestion points: Data is retrieved from the
doubaoya.comAPI endpoint through thescripts/fetch_cover.pyscript and then presented to the agent context (SKILL.md, Step 3).\n - Boundary markers: No explicit delimiters or system instructions are provided to distinguish between the fetched data and the agent's core operating instructions.\n
- Capability inventory: The skill possesses the capability to execute local Python scripts and perform network requests to the
doubaoya.comdomain.\n - Sanitization: There is no evidence of sanitization or filtering applied to the article titles fetched from the API before they are processed by the LLM.
Audit Metadata