{"slug": "wikikb-a-fast-repository-native-knowledge-base-for-people-and-agents", "title": "WikiKB: a fast, repository-native knowledge base for people and agents", "summary": "GitHub Next released WikiKB, a knowledge base layer built on the GitHub repository wiki that enables humans and agents to ingest, search, and query curated knowledge efficiently. WikiKB leverages Self-Organizing Memory for Agents (SOMA) from Microsoft Research's GraphRAG Zero project, using a small, static, local CPU embedding model for indexing and retrieval without requiring an LLM provider, though it can be configured to use one for RAG operations. The tool supports ingestion from files, URLs, and GitHub issues, and is designed to be readable, reviewable, portable, efficient, agent-friendly, and provider-optional.", "body_md": "# WikiKB: a fast, repository-native knowledge base for people and agents\n\nWikiKB is an efficient, transparent, and explicitly curated knowledge base layer for humans and agents, built on the GitHub repository wiki.\n\nAgentic memory is incredibly useful. We want our agents to grow their understanding of a domain over the life of a project. But the knowledge an agent accumulates passively can be haphazard and opaque. Projects and topics bleed over into each other, amplifying and muddying the model’s context.\n\n[WikiKB](https://github.com/githubnext/wikikb) is an efficient, transparent and explicitly curated knowledge base layer, for humans and agents, built on the GitHub repository wiki. Yes, every repo comes with a wiki! Most of them go unused, making it the perfect landing place for a human-readable, agent-addressable memory. WikiKB content can be ingested and queried from the command line, through GitHub issues, or agentically using the provided skill definition.\n\nBecause the knowledge base lives in the repository’s wiki, everything ingested can be viewed in the rendered wiki or in the raw Markdown files behind it. You can install WikiKB alongside an existing project, or create a dedicated repository to hold knowledge about any topic.\n\n## Try it!\n\nTell your agent:\n\n```\n Install WikiKB using https://github.com/githubnext/wikikb/blob/main/INSTALL.md\n```\n\n## Implementation\n\nWikiKB is fast and efficient, leveraging Self-Organizing Memory for Agents (SOMA) from the [GraphRAG Zero project](https://www.microsoft.com/en-us/research/project/graphrag/) at Microsoft Research. SOMA uses a small, static, local embedding model running on the CPU for efficient indexing and retrieval. No LLM provider is required to add documents to the knowledge base or search it. For RAG operations such as summarization and question answering, you can configure WikiKB to use an LLM provider.\n\nFor example:\n\n```\nwkb ai-research query \\\n  \"How does graph-based retrieval differ from vector search?\"\n```\n\nThis retrieves relevant entries from the knowledge base and sends them to the configured AI provider to produce a synthesized, source-backed answer.\n\nRun the same command with `--no-ai`\n\nto return the relevant entries without generating new content:\n\n```\nwkb ai-research query \\\n  \"How does graph-based retrieval differ from vector search?\" \\\n  --no-ai\n```\n\nYou can also use `search`\n\n, which is always retrieval-only:\n\n```\nwkb ai-research search \"hybrid retrieval methods\" --top 5\n```\n\n## Add knowledge from files, URLs, and GitHub issues\n\nWikiKB can ingest a local document or a public URL:\n\n```\nwkb ai-research ingest ./notes/retrieval-design.md --tag retrieval,design\nwkb ai-research ingest https://example.com/research-paper --tag research\n```\n\nIt can also archive GitHub issues, optionally including their comments:\n\n```\nwkb ai-research.sources.discussions ingest-issues \\\n  owner/repository \\\n  --state all \\\n  --limit 50 \\\n  --comments\n```\n\nIngested material is normalized into Markdown pages with provenance, metadata, and tags. Namespaces let you organize larger knowledge bases and search only the relevant section:\n\n```\nwkb ai-research.papers.transformers search \"positional encoding\"\n```\n\n## WikiKB design principles\n\n**Readable:** every entry is ordinary Markdown.**Reviewable:** changes are preserved in Git history.**Portable:** the wiki remains useful without WikiKB.**Efficient:** indexing and retrieval run locally on the CPU.**Agent-friendly:** agents can recall and store durable project knowledge through a defined interface.**Provider-optional:** retrieval works without an LLM account or API request.\n\n## Install WikiKB\n\nTell your agent:\n\n```\n Install WikiKB using https://github.com/githubnext/wikikb/blob/main/INSTALL.md\n```\n\n## Keep in touch\n\nPlease reach out on the [GitHub Next Discord](https://gh.io/next-discord) and let us know how WikiKB is working for you.", "url": "https://wpnews.pro/news/wikikb-a-fast-repository-native-knowledge-base-for-people-and-agents", "canonical_source": "https://githubnext.com/posts/wikikb/", "published_at": "2026-08-23 00:00:00+00:00", "updated_at": "2026-09-03 09:51:59.964059+00:00", "lang": "en", "topics": ["ai-tools", "ai-agents", "ai-infrastructure"], "entities": ["GitHub Next", "WikiKB", "Microsoft Research", "GraphRAG Zero", "SOMA"], "alternates": {"html": "https://wpnews.pro/news/wikikb-a-fast-repository-native-knowledge-base-for-people-and-agents", "markdown": "https://wpnews.pro/news/wikikb-a-fast-repository-native-knowledge-base-for-people-and-agents.md", "text": "https://wpnews.pro/news/wikikb-a-fast-repository-native-knowledge-base-for-people-and-agents.txt", "jsonld": "https://wpnews.pro/news/wikikb-a-fast-repository-native-knowledge-base-for-people-and-agents.jsonld"}}