codeguard-graphql
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads content from .graphql and .gql files, which are untrusted data ingestion points. It lacks explicit boundary markers or sanitization instructions to prevent the agent from following malicious instructions embedded in code comments. 1. Ingestion points: Scans all .graphql and .gql files in the repository. 2. Boundary markers: Absent. 3. Capability inventory: Executes shell commands via npx and modifies local files. 4. Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill executes shell commands using the npx utility to run eslint. While it uses the --no-install flag for detection, the fix command (npx eslint --ext .graphql . --fix) allows for package downloads to ensure tool availability if the package is missing locally.
Audit Metadata