constructive-storage

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references several external packages required for functionality in SKILL.md and references/server-setup.md, including graphile-presigned-url-plugin, graphile-bucket-provisioner-plugin, graphile-upload-plugin, and @aws-sdk/client-s3. These are standard dependencies for the described storage architecture.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data through GraphQL mutations.
  • Ingestion points: The requestUploadUrl mutation in references/graphql-mutations.md accepts filename and contentType as user-supplied input.
  • Boundary markers: No explicit boundary markers or "ignore embedded instructions" warnings are specified for the agent when handling these fields.
  • Capability inventory: The system creates database records and S3 objects, and generates presigned URLs for client-side uploads and downloads.
  • Sanitization: The skill documents validation rules including size constraints, MIME type restrictions (allowed_mime_types), and filename length limits, which provide basic sanitization against malformed input.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 01:06 PM
Security Audit — agent-trust-hub — constructive-storage