Mzizi MCP Server

The WCAG 2.x contrast ratio of a foreground on a background colour, with the AA-large, AA and AAA verdicts and whether it meets the Mzizi floor of 7:1. Computed locally — no network. It does not compute APCA Lc or simulate colour-vision deficiency; read the Mzizi accessibility doctrine (mzizi_get_tokens family=accessibility) for those rules.

Delegate a task to the fundi agent, or check on one. Submit-and-poll: `submit` returns a task id immediately and never waits for the run. Needs sign-in: fundi runs as your Mzizi console user.

The Mzizi DNA double helix — nodes, rungs and strands — with live counts. Pass `node` for one node's covenant, stakeholder and implementation rules.

Fetch one component in full, Rust first. When the component has a Mzizi Roots (Rust) implementation, `rust` comes first (Roots' crates are built on Dioxus 0.7 today and are moving to Mzizi's own renderer): its crate, the `cargo add` line, its CONTRACT clause block and the `.rs` source inline. `astro` follows when the component has a pure Astro implementation (mzizi-registry#397): its `mzizi add --target astro` line, the files it installs into src/components/mzizi/ and their dependencies; null otherwise. `react` follows: the React build (deprioritised; it keeps working, new work goes to Rust), shadcn-shaped with files[] inline, its `npx shadcn add` line, and `vendor`: the registry commit, each file's install target, sha256 and a raw URL pinned to that commit, each `@/` import with the dependency that installs it, and the live API item URL (labelled live: it can serve a different commit). Each sha256 is of the registry source at path@ref; a file `shadcn add` writes can differ (it strips "use client" under rsc: false and rewrites aliases). With no Rust, `rust` is null and `lead` is "react". Add docs (use cases, variants, a11y: the documented contract) and version history (from the changelog) with `include`. Former nyuchi-* names resolve to their mzizi-* successors.

How to build here: Ubuntu pillars and principles, mzizi conventions, and the AI instruction sets.

List the published Mzizi agent skills, or fetch one in full. Skills are what an agent loads first. Each skill has a category: `dev` (how every agent works in every repo: digital-hygiene and progress-report — load these before any dev work), `design`, `language` or `build`. Pass `category` to get every skill in it in full (e.g. category=dev), or `name` for one skill. A skill's answer lists the files it ships beside its SKILL.md (scripts such as hygiene.sh, token stylesheets); pass `name` and `file` to get one file's content. Each skill is also an MCP resource (mzizi://skills/<name>, files at mzizi://skills/<name>/<path>) and a prompt (skill-<name>). Served from the skill files bundled into this server at build time, so they are exactly the text in mzizi-dev/agent-tools at the deployed commit.

The Mzizi brand system: 21 colour families (7 minerals, 7 heritage, 7 experimental), semantic colours, typography, spacing, radii and the ecosystem brands. Ask for `colors` to get every colour family in one call. For text, use each family's `textLight` / `textDark` on the base surface, and `onContainerLight` / `onContainerDark` on its container; never use `hex`, `lightHex` or `darkHex` as a text colour (they are accents).

The registry index, optionally filtered by node, owner, collection, type or rust. Returns a lean row per component (name, type, node, owner, collection, and rustCrate when it has a Mzizi Roots (Rust) implementation, which is the lead) — use get_component for one component in full, or search when you only have words. Paged: up to 400 rows and at most 25000 characters per call; total, has_more and next_offset say how to get the rest.

What this MCP server exposes: its tools, which are free and which need sign-in, what each replaces, and where its data comes from.

Report a problem with a component or the registry. It goes to your own issue desk — a Durable Object holding your reports — which reproduces it via fundi, drafts an issue body including what the reproduction actually returned, and logs it to fundi. fundi files the GitHub issue with its proposed fix in it and tracks that issue afterwards, so action 'get' shows where yours got to: filed, picked up, closed. Use 'list' to see your earlier reports and 'retry' to re-attempt one that drafted but could not be logged. Your history is included in each new issue, because a repeated complaint from one reporter is often one bug. Needs sign-in: the desk belongs to your Mzizi console user.

Search the registry by words: components, conventions, AI instruction sets and releases. Start here when you do not know the exact name of what you want; use mzizi_get_component once you have a component's name. A component hit with `rustCrate` has a Mzizi Roots (Rust) implementation, which is the lead; at an equal score it ranks first. Documentation pages are searched by the docs_* tools. Paged: up to 50 hits per call and at most 25000 characters; meta.has_more and meta.next_offset say how to get the rest.

[docs.mzizi.dev] Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Mzizi documentation pages. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

[docs.mzizi.dev] Search across the Mzizi knowledge base to find relevant information, code examples, API references, and guides. Use this tool when you need to answer questions about Mzizi, find specific documentation, understand how features work, or locate implementation details. The search returns contextual content with titles and direct links to the documentation pages. If you need the full content of a specific page, use the query_docs_filesystem tool to `head` or `cat` the page path (append `.mdx` to the path returned from search — e.g. a result at `/some/page` is read with `head -200 /some/page.mdx`).

[docs.mzizi.dev] Report a problem with this documentation site so the docs team can fix it. Use when a documentation page is incorrect, outdated, confusing, incomplete, or has a broken example. This is for feedback about the documentation content itself — not for product support requests or feedback about this tool or assistant.