graphql-schema-regression
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
@graphql-inspector/clipackage from the NPM registry. This is a standard and well-recognized tool within the GraphQL ecosystem. - [COMMAND_EXECUTION]: The skill employs various shell commands, including
graphql-inspector,npx, andgit, to perform tasks such as schema introspection, diffing, and version control operations. - [DYNAMIC_EXECUTION]: The skill documents the use of the
--ruleflag to dynamically load and execute local JavaScript modules. This feature allows for project-specific schema validation logic to be run at execution time. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external GraphQL schemas provided as files or via remote introspection endpoints. This introduces a surface for indirect prompt injection where malicious content in a schema could attempt to influence the agent.
- Ingestion points: GraphQL schema files (SDL), JSON introspection results, and remote GraphQL endpoints.
- Boundary markers: The skill does not implement specific delimiters or safety instructions to isolate the content of the schemas being processed.
- Capability inventory: The execution environment allows for subprocess command execution and file system access.
- Sanitization: Content is parsed by the GraphQL tool, but there is no specific sanitization of input data to prevent prompt injection at the agent level.
Audit Metadata