mdrgnb-daily-push

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill is configured to transmit data to external platforms (Feishu and QQBot). While this is the intended functionality, it establishes a network pathway for data exfiltration.
  • [DATA_EXFILTRATION]: Absolute system file paths are hardcoded (/home/chenyechao/.openclaw/workspace/mdrgnb_proj), which exposes the local user's directory structure and environment details.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it processes untrusted external data from PubMed (Step 4 in SKILL.md).
  • Ingestion points: External scientific literature (abstracts and full text) retrieved from PubMed.
  • Boundary markers: None provided; the instructions do not include delimiters or warnings to ignore embedded instructions within the papers.
  • Capability inventory: The skill has the ability to write to the local file system (summary.md, scores.md, evaluated_papers.json) and perform network operations (pushing to Feishu and QQBot).
  • Sanitization: There is no evidence of sanitization or validation of the content retrieved from external sources before it is processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 08:18 AM