needle-engine
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill integrates the
needle_searchMCP tool and a public search API (https://search.needle.tools/api/semantic-search) that retrieves content from documentation, forum posts, Discord, and source code. Ingesting this untrusted external data creates a potential surface for indirect prompt injection.\n - Ingestion points: Search results from
needle_searchand the search API (SKILL.md, references/mcp.md).\n - Boundary markers: No specific delimiters or safety instructions are defined for processing search content.\n
- Capability inventory: The agent can write project files, execute the included
lookup-api.mjsscript, and run standard build/deployment tools (npm,npx,vite).\n - Sanitization: The skill does not provide mechanisms for sanitizing or filtering search results before they are processed by the agent.\n- [COMMAND_EXECUTION]: The skill provides a Node.js script
scripts/lookup-api.mjsdesigned to be executed locally. The script reads and extracts snippets from TypeScript definition files (.d.ts) within the project's dependencies to provide accurate API information.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates the download of assets such as HDRIs fromcloud.needle.toolsand utilizes libraries fromcdn.jsdelivr.net. These are standard vendor resources and well-known CDNs.
Audit Metadata