knowledge-synthesizer
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill content is informational and provides strategic guidance for knowledge engineering tasks without any executable components or tool definitions.
- [NO_CODE]: No scripts, configuration files, or external packages were provided or referenced in the skill definition. The analysis found no evidence of command execution or remote resource fetching.
- [INDIRECT_PROMPT_INJECTION]: The skill workflows are designed to process and synthesize information from external documents, which represents a surface for indirect injection attacks. 1. Ingestion points: Processes 'source documents', 'multiple sources', and 'unstructured text' for entity extraction (SKILL.md). 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified in the guidance. 3. Capability inventory: The skill has no access to sensitive tools, network operations, or file-writing capabilities. 4. Sanitization: Emphasizes fact verification, provenance tracking, and the use of confidence thresholds to mitigate potential hallucinations or malicious data influences.
Audit Metadata