graphql-architect

Fail

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: CRITICALEXTERNAL_DOWNLOADSMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill contains references to a domain that is blacklisted by automated reputation scanners.\n
  • Evidence: The URL 'https://jeffallan.github.io/claude-skills/skills/api-architecture/graphql-architect/' is flagged as malicious by URLite.\n- [METADATA_POISONING]: The skill file itself has been flagged for containing malware signatures.\n
  • Evidence: Automated scan result 'FileRepMalware' for SKILL.md.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides architecture for processing external data and schemas, creating a surface for indirect prompt injection attacks.\n
  • Ingestion points: Field resolvers in 'references/resolvers.md' and subgraph introspection in 'references/federation.md'.\n
  • Boundary markers: The skill suggests implementing query depth limiting and complexity rules to mitigate resource exhaustion.\n
  • Capability inventory: Includes 'readFileSync' for schema access and network operations for gateway-to-subgraph communication.\n
  • Sanitization: Recommends using 'zod' for input validation and strict GraphQL naming conventions.
Recommendations
  • CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
  • AI detected serious security threats
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Sep 5, 2026, 11:39 AM
Security Audit — agent-trust-hub — graphql-architect