Contextual AI
This MCP server provides AI assistants with enterprise search and document navigation capabilities through Contextual AI's platform, featuring three distinct implementations: single-agent RAG for specific knowledge bases, multi-agent routing with automatic reranking to select the best agent from available options, and an advanced document navigator that leverages Contextual AI's parse API to enable intelligent navigation of complex documents through hierarchical structure analysis and page-based content retrieval. The document navigator implementation stands out by providing agentic RAG tools that go beyond naive retrieval, offering interpretable attributions via tool call traces as agents navigate documents using inferred metadata like table of contents and document hierarchy. Built with FastMCP framework and featuring token counting for context management, tiktoken integration, and support for both local files and URLs, it serves enterprises needing conversational access to internal knowledge bases, teams requiring intelligent document comprehension for large reports and manuals, and organizations where natural language interfaces enhance information discovery across multiple specialized knowledge domains.
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
View full trust & usage report →Metadata
- Version
- 1.0.0
- Skill type
- atomic
- Execution layer
- mcp-remote
- Category
- social-media
- Source
- PulseMCP
- Author type
- human
- Updated
- 2026-05-25
Use via MCP
Resolve Contextual AI from your agent
Streamable HTTP transport at https://api.skillsregistry.net/mcp. No auth for read tools. Discovery: .well-known/mcp.json.
One command in your shell — Claude Code wires it up and verifies the connection. Run /mcp in any session to confirm.
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp--scope user for --scope project to commit it to .mcp.json.