Knowledge Base MCP Server

Enterprise knowledge base for ADK-Rust Enterprise agents. 9 MCP tools for articles, TF-IDF search boosted by helpfulness, feedback loops, gap detection, versioning, and draft/publish workflow.

Architecture

What It Does

Your agent resolves issues from knowledge before creating tickets. The KB gets smarter over time — helpful articles rank higher, and gaps (queries with no results) tell you what articles to write next.

Tools (9)

Tool	What It Does	Use Case
search_articles	TF-IDF search boosted by helpfulness	"Find articles about VPN"
get_article	Full article with body, stats, version	"Show me KB-001"
list_related_articles	Related by tags/category	"What else is related?"
create_article_draft	Create new draft	"Write an article about X"
publish_article	Make draft searchable	"Publish this article"
suggest_article_update	Update existing (new version)	"Update this article"
record_article_feedback	Helpful/not helpful + comment	"This was helpful"
list_articles	Browse by category/status	"Show all Network articles"
get_article_gaps	Queries with no results	"What articles are missing?"

How It Gets Smarter

1. Feedback boosts ranking — articles marked "helpful" score higher in search
2. Gap detection — every failed search is tracked. Gaps show what to write next.
3. View counting — popular articles surface naturally
4. Versioning — every update creates a new version

Verified Output

> search_articles(query: "password reset")
  1 result: "How to reset your password" (score: 8.3, helpful: 1)

> search_articles(query: "printer not working")
  0 results (gap tracked)

> get_article_gaps()
  1 gap: "printer not working" (searched 1x)

> record_article_feedback(article_id: "KB-001", helpful: true)
  recorded: true (boosts future search ranking)

Installation

cargo install mcp-knowledge-base

MCP client config

{ "mcpServers": { "kb": { "command": "/path/to/mcp-knowledge-base" } } }

Integration with ITSM

The ITSM MCP's handle_support_request can use this KB for resolution:

1. User reports issue → ITSM searches KB
2. KB returns relevant article → issue resolved without ticket
3. No result → gap tracked → team writes new article

Contributors

James Karanja Maina

License

Apache-2.0 — Part of the ADK-Rust Enterprise MCP server ecosystem.

Registry Compliance

This server implements the ADK MCP SDK contract:

• HealthCheck — async health probe for registry monitoring
• mcp-server.toml — manifest declaring tools, risk classes, and credentials
• Structured tracing — RUST_LOG env-filter for observability

rmcp and MCP compatibility

This server is built with rmcp 3.1.2 and requires Rust 1.94.1 or newer. The rmcp 3 rollout retains legacy MCP initialization compatibility and targets MCP protocol revisions 2025-11-25 and 2026-07-28.

MCP 2026-07-28 rollout (P4 workflow/business)

This server uses rmcp 3.1.2 and adk-mcp-sdk 0.2 with a minimum supported Rust version of 1.94.1. It accepts stateless MCP 2026 requests with per-request protocol, client identity, and capability metadata while retaining the legacy MCP 2025-11-25 initialize flow for ordinary tools.

• Tasks: None; this server's operations are short-lived and execute directly.
• MRTR approvals: publish_article
• Discovery and routing: rmcp serves on-demand discovery and validates the per-request protocol envelope; HTTP deployments can route with Mcp-Method and Mcp-Name. The packaged binary currently uses stdio.
• Caching: tools/list returns a public ttlMs of 60,000 for MCP 2026; rmcp omits the cache fields for legacy clients.
• Deprecated extensions: this server does not add new Roots, Sampling, or dynamic client-registration dependencies.

Protected tools require MCP_REQUEST_STATE_KEY with at least 32 high-entropy bytes. All replicas must share that key so sealed approval state can resume on another instance. Approval state is bound to the client identity, tool, and arguments and expires after two minutes. Missing identity, invalid state, rejection, or legacy protocol use fails closed. Task records are process-local for the current stdio runtime; use a durable task store before deploying the server behind scale-to-zero HTTP infrastructure.