knowledge-synthesizer
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and synthesize findings gathered from multiple external agents, which introduces a surface for indirect prompt injection where malicious instructions could be embedded in the source data.
- Ingestion points: The skill ingests 'findings' and 'outputs' from multiple other agents as described in the working mode instructions in SKILL.md.
- Boundary markers: The instructions lack defined delimiters (e.g., XML tags or specific block markers) to separate the untrusted agent findings from the synthesizer's own instructions.
- Capability inventory: No dangerous tools, subprocess calls, or file-writing capabilities were detected within the skill; the metadata specifies a 'read-only' sandbox mode.
- Sanitization: There are no instructions provided for the agent to sanitize, escape, or filter the external content before synthesis.
Audit Metadata