dln-network

Pass

Audited by Gen Agent Trust Hub on Mar 21, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill coordinates session data synchronization with Notion to maintain a persistent record of the learner's progress, including their stated mental models and session summaries. This external data transfer is a core component of the skill's distributed revision cycle and utilizes a well-known service.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface in the model retrieval step. 1. Ingestion points: Untrusted data enters the context when the learner explains their understanding of a domain in SKILL.md. 2. Boundary markers: The instructions do not employ explicit delimiters or 'ignore' warnings for the captured user input. 3. Capability inventory: The skill has the capability to write and modify data in Notion via a defined merge protocol. 4. Sanitization: There is no evidence of validation or filtering for the learner's verbal responses before they are processed and stored.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 21, 2026, 12:33 AM
Security Audit — agent-trust-hub — dln-network