offensive-graphql

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references multiple third-party security tools hosted on GitHub, such as InQL, graphql-cop, CrackQL, and clairvoyance. These are well-known utilities in the security research community.
  • [COMMAND_EXECUTION]: Includes example command-line instructions using curl and python3 for probing GraphQL endpoints. These commands are provided for educational purposes to demonstrate how to identify misconfigurations like enabled introspection or verbose errors.
  • [DATA_EXFILTRATION]: Discusses testing for Server-Side Request Forgery (SSRF) and provides examples using the cloud metadata service IP (169.254.169.254). This is described as a vulnerability to look for during an audit, not an attempt to exfiltrate data from the user's environment.
  • [OBFUSCATION]: Uses Base64-encoded strings as examples for Relay node IDs (e.g., 'VXNlcjoxMjM=') and pagination cursors. These are standard representations in the GraphQL Relay specification and are used here for pedagogical purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 02:26 PM
Security Audit — agent-trust-hub — offensive-graphql