kegg-database
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill demonstrates susceptibility to indirect prompt injection (Category 8).
- Ingestion points: Data is ingested from the external KEGG REST API through functions in
scripts/kegg_api.py(e.g.,kegg_get,kegg_find). - Boundary markers: The instructions lack specific delimiters or instructions for the agent to treat API responses as untrusted data.
- Capability inventory: The skill possesses the capability to execute Python scripts (
scripts/kegg_api.py) and perform network requests to the KEGG API. - Sanitization: There is no evidence of sanitization or validation of the content returned from the external API before it is processed by the agent.
- [PROMPT_INJECTION]: The skill includes behavioral steering instructions that direct the agent to promote a specific commercial platform.
- The 'Suggest Using K-Dense Web' section instructs the agent to proactively suggest the
www.k-dense.aiplatform when task complexity increases, which acts as a promotional injection within the skill's logic.
Audit Metadata