pool-info
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches pool data from an external API and interpolates the resulting strings directly into the agent's context and output, which could be exploited if the API returns malicious instructions.
- Ingestion points: The skill performs network requests to
https://earn-service.kyberswap.com/api/v1/explorer/poolsto retrieve pool and token metadata. - Boundary markers: The instructions lack specific delimiters or "ignore instructions" markers to protect the agent from interpreting data fields (e.g., token symbols, DEX names) as commands.
- Capability inventory: The skill possesses network fetch capabilities (
WebFetch) and access to local reference files within the${CLAUDE_PLUGIN_ROOT}. - Sanitization: The skill does not implement validation or sanitization for the external content before rendering it in markdown tables or JSON blocks.
Audit Metadata