MiMoCode Self-Help Docs

What it does

When your agent uses it

• Asking how to switch between build, plan and compose modes
• Adding a custom OpenAI-compatible provider with a base URL and API key
• Finding where MiMoCode keeps its global or project configuration
• Diagnosing TUI lag, rendering glitches or SSH rendering problems
• Letting a script borrow this instance's models through its local /v1 endpoint

Example prompts

• “How do I leave plan mode so the agent can start editing files?”
• “Point MiMoCode at my company's OpenAI-compatible gateway and set the model name.”
• “Why is the terminal UI laggy over SSH, and what can I change?”
• “Which key cycles between the primary agents?”

Requirements

• MiMoCode (the `mimo` CLI)

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1Read the routed reference and identify the exact schema fields. Do not infer fields from another tool's config format.
2. 2Determine scope from the request. Treat model/provider setup as global unless the user says it is project-only; use project config for…
3. 3Inspect only the exact config candidates. Never recursively search the user's home directory. Prefer an existing higher-precedence .jsonc…
4. 4Keep secrets out of tool output and the final response. When inspecting a config, redact values for keys such as apiKey, token, secret…
5. 5Edit minimally. If the request says configure, use, or make default, also set the top-level model; if it only says add, leave the current…
6. 6When persisting a newly supplied API key and model, also put that exact provider/model at the front of the TUI recent-model state so the…
7. 7Validate the parsed configuration with the narrowest relevant command and report the config and recent-model state files changed, selected…

What it can do on your machine

Read from SKILL.md and the folder at commit 6babeb0. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are jsonc).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • mimo.xiaomi.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

MiMoCode Self-Help Docs loads about 2.7k tokens when it runs. Until then it costs about 256 tokens; SKILL.md has 1,198 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~256

When it runs · the whole SKILL.md, loaded when a task matches

~2.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from XiaomiMiMo/MiMo-Code at commit 6babeb0, republished under its MIT licence (© XiaomiMiMo). 1,198 words, ~2,668 tokens.

name

mimocode-docs

description

Use whenever the user asks about MiMoCode itself: features, TUI or CLI commands, keybindings, terminal compatibility, rendering glitches, TUI lag, SSH or remote rendering, agent modes (build / plan / compose) and how to switch between them, configuration, file locations, providers, models, authentication, or custom OpenAI-compatible or Anthropic-compatible API endpoints. Especially trigger when a prompt supplies or asks to configure a base URL/baseURL, API key/apiKey, model name or ID, provider, Anthropic Messages API, or global/project mimocode.json/jsonc, or when the user asks how to enter or leave plan mode. Also trigger when a skill, task, subprocess, or external client needs to borrow this instance's models — the OpenAI-compatible /v1 chat endpoints every MiMoCode server serves, `mimo llm-server` task tokens, or how to expose a listening port for them. Use this skill to inspect existing config safely, make minimal changes, and verify them without guessing schema fields or model capabilities.

{
  "$schema": "https://mimo.xiaomi.com/mimocode/config.json",
  "model": "provider/model",
  "permission": { "external_directory": { "/tmp/**": "allow" } }
}

© XiaomiMiMo, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 8 other files in packages/cli/src/skill/builtin/.bundle/mimocode-docs of XiaomiMiMo/MiMo-Code.

• SKILL.md
• reference/capability-api.md
• reference/commands.md
• reference/config.md
• reference/guide.md
• reference/mcp-sampling.md
• reference/permissions.md
• reference/providers.md
• reference/workflows.md

Open the folder on GitHubat commit 6babeb0