Searchable codebase wikis and context for coding agents

Point it at any repo, get a searchable Wiki and an MCP server for your coding agent.

Quickstart  ·  Website  ·  Discord  ·  Documentation  ·  MCP  ·  Agent Integrations  ·  Languages

python -m pip install "codenib[mcp,semantic]==0.2.0"
codenib wiki /path/to/your/repo

Local, open source, and no cloud required. CodeNib combines BM25 and dense code search, adds optional SCIP symbol graphs, and incrementally rebuilds repository indexes as commits change. The same index powers the Wiki, Dependency Map, Ask, and MCP tools instead of making every agent rediscover the codebase.

News

• 2026-08-05 — CodeNib 0.2.0. Build a static Wiki and reusable context artifact once, then serve it through Pages or the official MCP package. Hybrid retrieval and managed SCIP/LSP providers ship in the same CLI. Release notes
• 2026-08-05 — SweRank recipe. Run SweRank retrieval and reranking over a local checkout. Example
• 2026-08-04 — Native repository explorer. CodeNib's planner now targets the SWE-Explore source-region protocol. Validation
• 2026-08-03 — Native LocAgent policy. LocAgent runs directly on CodeNib views without LocAgent, LiteLLM, or LlamaIndex dependencies. Support matrix
• 2026-08-02 — OrcaLoca SearchAgent. OrcaLoca's six-tool search loop now reuses CodeNib's symbol graph. Support matrix

System Architecture

repository change
  -> materialize or repair affected views
  -> publish a capability-bearing manifest
  -> plan repository queries
  -> deliver bounded, source-linked context

On a later commit, CodeNib can reuse unchanged vector content and patch supported graph transitions at file or symbol granularity. Unsupported, inconsistent, or unverified transitions fall back to a fresh build instead of publishing a partially updated view.

Quickstart

Requires Python 3.10+ and Git. The recommended local path includes the pinned CodeRankEmbed model and serves hybrid BM25+dense retrieval:

python -m pip install "codenib[semantic]==0.2.0"
codenib doctor --require core --require wiki
codenib wiki /path/to/repository

codenib wiki selects the semantic route because the installed environment contains its dependencies. A smaller python -m pip install codenib==0.2.0 installation selects the deterministic BM25 fallback and downloads no model. The 0.2 release notes record the upgrade boundary and verification evidence.

CodeNib detects the repository languages, builds a reusable index under ~/.codenib/repositories, launches the local Wiki, and opens http://localhost:3000. The wheel includes the production Wiki frontend, so normal use does not require Node.js or npm and the target repository stays untouched. This command exercises the same compiler and serving runtime used by agents. Set CODENIB_HOME to relocate state.

Check the environment or index without opening the Wiki:

codenib doctor --require core --require wiki
codenib index /path/to/repository

For a structural view, CodeNib detects the repository languages and manages only their package-level providers; operating-system and project prerequisites remain explicit:

python -m pip install "codenib[graph]==0.2.0"
codenib toolchain install /path/to/repository --scope graph
codenib doctor /path/to/repository --require graph

Export that indexed commit as a serverless Wiki when a live Ask backend is not needed:

codenib export /path/to/repository --output /tmp/repository-wiki

The export contains a versioned provenance manifest, precomputed Wiki pages, source citations, and available page-level dependency data. It contains no provider credential; interactive Ask and runtime graph exploration remain on the local or MCP serving path.

For a repository-hosted Wiki, CodeNib also ships a reusable GitHub workflow that incrementally builds the same manifest, deploys the static site to Pages, and uploads the matching commit-addressed context artifact. Its default semantic route builds BM25 and vector views with a cached local Hugging Face model and needs no API key. An explicit fast route avoids the model download; a BYO OpenAI-compatible endpoint can replace local embedding. Query-time search remains in the local or MCP runtime. See GitHub Pages. The published BM25/vector artifact can then be verified against an exact local checkout and served through MCP without rebuilding the repository views.

See the Quickstart for ports, advanced indexing, and troubleshooting.

Serve An Agent

Install the MCP extra, build once, and serve the same repository manifest over stdio:

python -m pip install "codenib[mcp,semantic]==0.2.0"
codenib index /path/to/repository
codenib mcp /path/to/repository

The MCP server advertises search_context as its default ranked entry point, then exposes the underlying BM25, vector, graph, and navigation operations for explicit control. It uses the compiled manifest to decide which calls have a fresh backing view. An agent can therefore reuse available repository work instead of rebuilding context through unbounded grep and read loops, while unavailable searches fail explicitly. BM25, semantic, regex, Zoekt, dependency, and static-navigation results retain source locations for follow-up reads and citations. Large search bodies are projected under one aggregate budget without changing rank; the bounded read_source tool recovers exact source windows from the verified checkout. See MCP Server for client configuration and tool contracts.

The same planner is available directly to Python agents:

from codenib.agent import RepositoryContextExplorer

with RepositoryContextExplorer.from_repository(
    "/path/to/repository", policy="auto"
) as explorer:
    result = explorer.explore("where is request retry behavior implemented?", top_k=10)

Each result includes source-validated evidence plus the selected plan, capabilities, loaded views, fusion, graph, and reranking trace.

What CodeNib Provides

Language support varies by surface. The generated capability matrix records chunking, graph, incremental, and C++ decoder support.

Documentation

• Quickstart
• GitHub Pages
• MCP Server
• Agent Integrations
• RAG Models and Planner
• Benchmarks and Evaluation
• Web UI
• Language Capabilities
• Concepts and development guides

Build the documentation site locally with:

python -m pip install -e ".[dev]"
mkdocs serve

Development

git clone https://github.com/sysevol-ai/CodeNib.git
cd CodeNib
make dev
make test

The test suite is split into unit, integration, serial integration, core, graph-consumer, and slow tiers. See CI/CD before running the credential- or toolchain-dependent tiers.

Status

CodeNib 0.1.0 is a developer preview. The CLI and manifest format are usable, but public interfaces may still change before a stable release. Historical research artifacts retain their published dataset identifiers; the maintained package, import namespace, commands, and repository use CodeNib. See Naming.

Citation

If you use CodeNib in your research, please cite our arXiv paper:

@misc{yu2026codenibmultiviewdataserving,
      title={CodeNib: A Multi-View Data System for Serving Repository Context to Coding Agents},
      author={Zhongming Yu and Hengjia Yu and Boqin Yuan and Shuting Zhao and Yizhao Chen and Aryan Dokania and Mihir Jagtap and Jiayu Chang and Yitong Ma and Yash Jayswal and Wentao Ni and Hejia Zhang and Zhaoling Chen and Gangda Deng and Jishen Zhao},
      year={2026},
      eprint={2607.25431},
      archivePrefix={arXiv},
      primaryClass={cs.SE},
      url={https://arxiv.org/abs/2607.25431},
}

Project

Website  ·  Changelog  ·  Contributing

License

CodeNib is licensed under the Apache License, Version 2.0.