# FSoft-AI4Code/CodeWiki

[ACL 2026] Open-source framework for holistic, structured repository-level documentation across multilingual codebases

Repository: https://github.com/FSoft-AI4Code/CodeWiki
Canonical: https://ross.abutalabs.com/products/codewiki
Homepage: https://fsoft-ai4code.github.io/CodeWiki/
Language: Python
License Family: other
Topics: automated-documentation, code-documentation, code-understanding, documentation-generator, repository-documentation, open-source-deepwiki
Last push: 2026-08-24T02:30:37+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 26, longevity 31
- inputs: {"age_days": 434, "days_push": 10, "days_rel": 283, "gap_med": null, "n_releases_24m": 1}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1644, forks 250 (observed 2026-08-28T04:05:16.040874+00:00)

## What it is
CodeWiki is an open-source Python framework that uses LLM-powered multi-agent processing to generate holistic, structured documentation for large, multilingual codebases. It decomposes repositories hierarchically, captures cross-module architecture, and produces docs with diagrams, supporting multiple LLM providers via a CLI.

## Use cases
- generate documentation for an entire repository automatically
- open-source DeepWiki alternative for repo-level docs
- understand a large unfamiliar codebase quickly
- create architecture diagrams and docs for multilingual projects
- document a 1M+ line codebase with LLMs
- auto-generate wiki pages from a GitHub repo

## When to choose
- you need repository-level, architecture-aware documentation rather than per-function docstrings
- your codebase spans multiple programming languages
- you want an open-source, self-hostable DeepWiki alternative with provider flexibility

## When to avoid
- you only need simple docstring or API reference generation from source comments
- you cannot send proprietary code to external LLM providers and lack local models
- you need guaranteed up-to-date docs on every commit without review

## Facets
- artifact type: framework
- maturity: active
- function: documentation, llm-inference, agent-framework, cli, developer-tools
- domain: developer-tools, documentation, large-language-models
- platform: python, cli, cross-platform
- tags: repository-documentation, documentation-generator, code-understanding, deepwiki-alternative, multilingual-codebases, architecture-diagrams, acl-2026, ai-agents

## Member repositories
- FSoft-AI4Code/CodeWiki (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:16.040874+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:45:47.667634+00:00, confidence not recorded.
  - readme: https://github.com/FSoft-AI4Code/CodeWiki (fetched 2026-08-28T04:05:16.040874+00:00, sha f10cb25dd69d)
  - homepage: https://fsoft-ai4code.github.io/CodeWiki/ (fetched 2026-08-29T11:19:01.599092+00:00, sha 8fc90dc39896)
- Data as of 2026-08-30T08:39:29.467469+00:00.