> For the complete documentation index, see [llms.txt](https://docs.visdom.virtuslab.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/wiki-module.md).

# Wiki module

The wiki module maintains a [Karpathy-style LLM wiki](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f) over a configured corpus, stored as markdown pages. The source material is typically the project's documentation, both internal to a repository and external.

Wiki is a project-scoped module. After enabling it, the `Wiki` page lists the concepts it has derived so far, and leads to the corpus behind them.

![The Wiki page: concept pages on the left, recently changed ones on the right](https://1398155133-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FED9nLcXxDQR0x7nt8zMS%2Fuploads%2FaOQH3cbDzzoK67HZyMI8%2Fwiki-concepts.png?alt=media)

The corpus can be:

* directly uploaded files (`.pdf`, `.md`, `.txt`, `.html` and similar)
* websites, refreshed manually
* paths in a repository, such as its ADRs or generated plans

![The Corpus page, where files, URLs and repository paths are added as sources](https://1398155133-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FED9nLcXxDQR0x7nt8zMS%2Fuploads%2FG8OTqlxcxmYUXoLCePgI%2Fwiki-corpus.png?alt=media)

A Confluence connector is planned. After adding a corpus item, allow time for ingestion.

## How the content is organised

The data is organised into concepts, following the emerging [Open Knowledge Format](https://cloud.google.com/blog/products/data-analytics/how-the-open-knowledge-format-can-improve-data-sharing). The goal is short pages, one concept each, with long sources split across several. An agent then looks information up by searching for the relevant keywords.

By design the source is not repeated verbatim; it is grouped and summarised. Each concept page links to the sources that contributed to it, so an agent can fetch the original when it needs to, and the wiki engine can keep the page current when a source changes or disappears.

Each page carries a title, a description, tags, aliases for synonyms, and the content itself.

## How agents read it

The wiki is browsable in the UI, but that is not the intended consumer. An agent reads it either:

* through a shared disk mount, by navigating and grepping markdown files, which coding agents are very good at, or
* through a dedicated set of MCP tools, see `wiki_*`, where no mount is available

{% hint style="info" %}
With the Orchestrator, the shared disk holding the wiki is mounted automatically for every agent in a matching project.
{% endhint %}

The wiki is self-healing. Lint warnings such as missing links or over-long pages often resolve themselves as concept pages are refined and new sources are incorporated. Relationships between pages can also be viewed as an Obsidian-style knowledge graph.

![The wiki as a knowledge graph, each node a concept page and each edge a link between them](https://1398155133-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FED9nLcXxDQR0x7nt8zMS%2Fuploads%2FfuHBD0Gmo9QymNLCTOyt%2Fwiki-knowledge-graph.png?alt=media)

{% hint style="warning" %}
**Every wiki operation is LLM-based, so watch the cost.** Indexing uses cheaper models, incremental updates and caching, but a codebase whose docs churn heavily will accumulate a daily bill.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/wiki-module.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.