wechat-draft-writer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, including 'reference materials' and 'voice transcript drafts', which could contain embedded malicious instructions designed to influence the agent's output behavior during the drafting phase.
- Ingestion points: The skill explicitly requires external inputs in
SKILL.md(confirmed outline, reference materials, voice transcript draft). - Boundary markers: There are no explicit instructions or delimiters defined to isolate user-provided content from the system prompts, increasing the risk of the agent following instructions found within the source material.
- Capability inventory: The skill's primary function is text generation. It does not possess high-risk capabilities such as network access, system-level file writes, or arbitrary command execution in the analyzed files.
- Sanitization: There is no evidence of input validation, sanitization, or filtering applied to the ingested source materials before they are processed by the LLM.
Audit Metadata