jerlin-weread-skill

Pass

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill uses a local shell script (scripts/weread.sh) to communicate with the official WeChat Reading API gateway at i.weread.qq.com using standard Bearer token authentication via environment variables.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data such as book reviews, notes, and titles from the WeRead API.
  • Ingestion points: Data flows into the agent context through the execution of scripts/weread.sh which fetches content from WeRead.
  • Boundary markers: No specific delimiters or boundary instructions are used to isolate API-provided data from agent instructions.
  • Capability inventory: The skill's capabilities are restricted to information retrieval and display; it lacks dangerous actions like arbitrary code execution or filesystem modifications.
  • Sanitization: Content retrieved from the WeRead service is displayed without specialized sanitization or filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
May 17, 2026, 09:07 PM
Security Audit — agent-trust-hub — jerlin-weread-skill