Knowledge Search
MCP connector that searches across internal documentation repositories, wikis, and knowledge bases to surface relevant answers in context.
Overview
Knowledge Search connects your AI assistant to internal documentation so it can answer questions with actual company context instead of generic responses. It indexes Confluence spaces, Notion databases, and internal wikis, returning the most relevant passages alongside source links.
Capabilities
| Capability | Description |
|---|---|
| Multi-source Search | Queries Confluence, Notion, and custom markdown repositories |
| Semantic Ranking | Returns results ranked by relevance, not just keyword match |
| Source Attribution | Every answer includes a direct link to the source document |
| Context Injection | Automatically injects relevant docs into the AI’s context window |
| Access Control | Respects document-level permissions from the source system |
Install Steps
1. Add to Claude Desktop config
{ "mcpServers": { "knowledge-search": { "command": "npx", "args": ["-y", "@qiscus/knowledge-search-mcp"], "env": { "KNOWLEDGE_BASE_URL": "https://your-confluence.atlassian.net", "API_TOKEN": "your_api_token" } } }}2. Restart Claude Desktop
3. Verify the connection
Ask Claude: “Search internal docs for our onboarding process”
Need help? Contact the Integration Team or check the internal wiki.