mel-framework-citation
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to ensure that technical outputs for Monitoring, Evaluation, and Learning (MEL) use current, authoritative framework versions (e.g., OECD-DAC, WHO/UNFPA). It defines specific citation rules based on audience tiers and focuses on professional rigour.
- [INDIRECT_PROMPT_INJECTION]: The skill employs
WebSearchandWebFetchto verify the recency and canonical URLs of cited academic and grey literature. - Ingestion points: Data enters the agent's context from external web pages during the verification of publication dates and URLs.
- Boundary markers: The instructions do not specify explicit delimiters or "ignore" headers for the content fetched from external sources.
- Capability inventory: The skill uses web search and retrieval tools to validate data but does not contain instructions for persistent file writing, subprocess execution, or arbitrary command execution.
- Sanitization: No explicit sanitization or filtering logic is defined for the content returned by web searches, relying on the underlying agent platform's safety guardrails.
Audit Metadata