graphql-schema

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute project-level build and synchronization commands, specifically npm run gql:typegen and npm run sync-types. These are standard development tasks for GraphQL type generation and schema synchronization.
  • [INDIRECT_PROMPT_INJECTION]: The skill involves reading and writing project source files (GraphQL schemas and TypeScript components) to generate and modify code. This represents a standard attack surface for indirect prompt injection if the project files being processed contain malicious instructions.
  • Ingestion points: Processes .gql and .tsx files within the user's project workspace.
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are provided for the agent when processing these files.
  • Capability inventory: Includes shell command execution through the npm CLI.
  • Sanitization: No explicit sanitization or validation of the input file content is mentioned, relying on standard development tool behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 02:56 PM
Security Audit — agent-trust-hub — graphql-schema