1inch-mcp-server
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to use
npx -y supergatewayto bridge HTTP transport to stdio for specific clients. This fetches thesupergatewayutility from the official npm registry. - [COMMAND_EXECUTION]: The documentation provides shell commands using
npxto facilitate protocol bridging, which is a standard setup procedure for MCP (Model Context Protocol) servers in specific environments. - [INDIRECT_PROMPT_INJECTION]: The skill implements a
searchtool and aget_exampletool that ingest external documentation and code examples from 1inch's servers, creating a potential surface for indirect prompt injection. - Ingestion points: Documentation search results and SDK example packages fetched from 1inch API endpoints (e.g.,
api.1inch.com). - Boundary markers: None explicitly defined in the skill instructions to separate external data from system instructions.
- Capability inventory: The skill includes tools for financial operations (
swap,orderbook) and a generalproduct_apitool capable of performing network writes (POST requests). - Sanitization: The instructions do not specify validation or sanitization of the documentation content before it is presented to the agent.
Audit Metadata