knowledge-base-builder

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external documents (PDFs, Markdown, text) into an agent's knowledge base. This creates a surface for indirect prompt injection if the source documents contain adversarial instructions.
  • Ingestion points: Phase 1 and Phase 3 describe document ingestion from external sources.
  • Boundary markers: No specific boundary markers or 'ignore' instructions are included in the provided markdown templates.
  • Capability inventory: The skill utilizes Write, Read, Edit, Grep, Glob, and Bash tools to manage the knowledge filesystem.
  • Sanitization: No explicit sanitization logic for ingested text is provided in the code snippets.
  • [COMMAND_EXECUTION]: The skill requests Bash and filesystem tool access. This is aligned with its primary purpose of creating directory structures and managing knowledge files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 06:20 PM