Project RAG wiki
Allows agents to search, read, list, and write Markdown wiki notes indexed with vector embeddings, supporting frontmatter and semantic sections for structured knowledge retrieval.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Project RAG wikisearch the wiki for 'deployment steps'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
local-rag-wiki
A repo-local RAG wiki for AI agents: a per-repository MCP knowledge service (governed search / read / write of Markdown wiki notes) plus an installer that wires it into your agent tooling. Docker is required to run the service.
Quick start (recommended)
Pre-pull the service image before your first agent session to avoid cold-start delays:
docker pull ihorleleka/project-rag-wiki:latestRelated MCP server: hw_rag_mcp
Install
Run in your repository root:
npx github:ihorleleka/Local-Rag-Wiki install .This scaffolds the managed .agents/ runner, the AGENTS.md policy section,
MCP config for DeepSeek Harness (.dsh/mcp.servers.yml), Claude Code, Codex,
VS Code, and OpenCode, plus a wiki/ folder. These integrations are additive:
installing DSH support does not remove or replace the other harness configs.
The MCP service then starts automatically per repository when your agent client
connects. Pre-pulling the image (above) is recommended before running the agent
in a fresh environment.
DeepSeek Harness
Install the native DSH bundle once into the profile you use (usually web):
dsh plugin --profile web add github:ihorleleka/Local-Rag-WikiRestart that DSH profile, open an installed repository as its workspace, and
start a session. On agent/created, the bundle uses the agent's actual
workspace (including a custom --agents-dir) to mount DSH's official
@deepseek-ai/dsh-mcp-client, launch the managed MCP runner, reconnect it,
discover its tools, and register predictable
mcp__wiki-manager__* tools. The client remains agent-scoped even though the
namespace is stable. The runner starts or attaches to that
repository's Docker service exactly as it does for the other harnesses.
It also mounts a marker-gated Cordis recall coordinator for that workspace:
agent/session-start seeds recurring-topic hints; agent/pre-step uses local
state, then performs bounded wiki retrieval only for a material prompt—first
wiki_search depth="abstract" (L0), then matching depth="packet" context
(L1). It labels the evidence and leaves full L2 wiki_read decisions to the
model. session/event records an assistant summary, while agent/turn-stopping
and session/flush trim and persist state. Recall is per-workspace cached and
in-flight deduplicated; L1 packets are rate-limited, cold starts have a bounded
90-second allowance, and state updates are queued with atomic replacement. No
state or wiki process is used outside a repository with a wiki-kit install marker.
The generated .dsh/mcp.servers.yml remains a workspace-config alternative;
prefer the profile bundle rather than running both bridges, which would create
two MCP connections to the same repository service. AGENTS.md remains the
portable policy layer.
The .agents/hooks/ and .claude/settings.local.json hooks are Claude
Code-specific. DSH lifecycle interception belongs in trusted Host/Profile
Cordis plugins, not a project-local hook manifest. See
.dsh/README.md for the workspace-config
boundary.
Update
After install, refresh the managed files from inside the repo — no npx needed:
.agents\update-wiki-kit.cmd --forcesh .agents/update-wiki-kit.sh --forceLifecycle
npx github:ihorleleka/Local-Rag-Wiki status .
npx github:ihorleleka/Local-Rag-Wiki restart .
npx github:ihorleleka/Local-Rag-Wiki doctor . --livestart ., stop ., and pull . are available the same way. The repository
container persists independently of individual agent clients.
How it works
This package is the installer: it scaffolds and maintains the
.agents/runner and MCP configuration in a consumer repository.kb-service/ is the Dockerized MCP knowledge service the runner launches per repository. It serves
wiki_search,wiki_read,wiki_list,wiki_tree,wiki_schema_report,wiki_write,wiki_capture,wiki_delete, andwiki_renameoverPOST /mcp/(loopback by default), with hash-protected writes.wiki_searchsupports tiered retrieval (depth=abstract|packet) pluspath_prefixdirectory scoping.
Versioning
The installer and kb-service ride the same Git tag (X.Y.Z or vX.Y.Z).
Release automation:
License
MIT (see kb-service/LICENSE).
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseAqualityDmaintenanceAn MCP server that enables saving, retrieving, and managing research content using ChromaDB vector storage and semantic search.5MIT
- FlicenseNot gradedqualityCmaintenanceMCP server that indexes a knowledge base into Chroma and provides search tools for retrieving document fragments via vector embeddings.
- AlicenseNot gradedqualityCmaintenanceMCP server to query a Git-hosted Wiki, offering tools to list, search, and read Markdown documents via Codex or VS Code.MIT
- AlicenseNot gradedqualityBmaintenanceMCP server that indexes Markdown files into a local SQLite vector store and provides semantic search tools for Claude Code.3MIT
Related MCP Connectors
Serve a folder of Markdown notes as an MCP server: hybrid search, reading, and sourced answers.
Remote ChromaDB vector database MCP server with streamable HTTP transport
Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/ihorleleka/Local-Rag-Wiki'
If you have feedback or need assistance with the MCP directory API, please join our Discord server