deep-agents

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation indices from LangChain's official domain (docs.langchain.com). This is a reference to a well-known technology service used for contextual retrieval.
  • [SAFE]: The reference file references/local-course-map.md contains absolute paths to the author's local directories (e.g., /Users/matheus/Documents/...). While this exposes local directory structure, it is categorized as safe within the context of a personal developer educational tool.
  • [SAFE]: Indirect Prompt Injection Surface: 1. Ingestion points: SKILL.md fetches from docs.langchain.com/llms.txt. 2. Boundary markers: None. 3. Capability inventory: Virtual filesystem, subagent delegation, and tool calling. 4. Sanitization: None. This surface is considered safe as the data source is a trusted technical documentation site.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 07:42 PM