graphql-mcp-server
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npxto download and execute thegraphql2mcppackage and instructs users to install@graphql2mcp/libfrom NPM. These are established delivery methods for the vendor's own tools. - [INDIRECT_PROMPT_INJECTION]: The skill creates an interface between the agent and external GraphQL APIs, establishing an attack surface where responses from those APIs could contain instructions designed to influence the agent.
- Ingestion points: Untrusted data enters the agent context via responses from the configured GraphQL endpoint during tool execution (specified in
SKILL.mdandreferences/cli.md). - Boundary markers: The provided instructions do not include the use of explicit delimiters or instructions for the agent to ignore content within API responses.
- Capability inventory: The agent can perform queries and mutations (if enabled) against the GraphQL API, which could include data retrieval or modification.
- Sanitization: While input arguments are validated using Zod schemas based on the GraphQL schema, the output data from the API is not sanitized for potential prompt injection content.
Audit Metadata