understand-anything-knowledge-graph

Warn

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to add and install a plugin from the repository Lum1104/Understand-Anything.
  • [REMOTE_CODE_EXECUTION]: The development installation steps involve cloning a remote repository and executing build commands (pnpm install, pnpm build) which run code from a third-party source.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes all files within a project's codebase to build a knowledge graph. Evidence: (1) Ingestion points: All files, functions, and classes within the project directory being analyzed. (2) Boundary markers: No explicit boundary markers or instructions to ignore embedded prompts in the analyzed codebase are defined. (3) Capability inventory: Reads files, orchestrates multiple analysis agents, writes analysis results to .understand-anything/knowledge-graph.json, and opens a web dashboard. (4) Sanitization: No sanitization or validation of the ingested code content is specified before it is processed by the LLM agents.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 12, 2026, 11:45 PM
Security Audit — agent-trust-hub — understand-anything-knowledge-graph