skills/itechmeat/llm-code/seaweedfs/Gen Agent Trust Hub

seaweedfs

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill provides instructions for managing SeaweedFS clusters which involves the agent processing untrusted data such as file paths, directory listings, and metadata, creating a surface for indirect prompt injection. Ingestion points: API responses in references/api-surfaces.md and directory listings in references/filer-core.md. Boundary markers: Absent. Capability inventory: Administrative shell access via weed shell and mutating API endpoints. Sanitization: Absent.
  • [EXTERNAL_DOWNLOADS]: The skill references official SeaweedFS project resources for documentation, releases, and container images from established sources. Evidence: Links to the official GitHub repository and the maintainer's Docker Hub profile.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 12:21 AM