mcpField
M
mushi-mushi

mushi-mushi

Remote
kensaur.usDeveloper Tools3Updated today

About mushi-mushi

mushi-mushi is an MCP server that brings bug reports into coding editors for diagnosis and repair workflows. It uses the editor's own model, and can provide report context or dispatch and merge fixes without requiring a second LLM key.

Key features

  1. 01Diagnose reports with evidence and repro steps
  2. 02Surface live triage queue
  3. 03Dispatch and merge agent fixes
  4. 04Search project docs
  5. 05Filter tools by feature group

Use cases

  1. 01Investigate a production bug from an editor
  2. 02Give an agent context for a report fix
  3. 03Review and merge a draft fix PR

MCP support

Connects MCP clients such as Cursor, Claude Code, and VS Code to Mushi bug reports, documentation, and fix actions. It exposes tools, resources, and prompts, with read-only configuration available for hosted connections.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://kensaur.us/mushi-mushi/hosted-mcp/
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Packages
npm@mushi-mushi/mcp0.30.0

Tools

Details

From the project README

@mushi-mushi/mcp

Your AI wrote it. Mushi tells you why it broke.

Add to Cursor Add to VS Code Try the demo — no signup npm

The Model Context Protocol server for Mushi Mushi. It lets Cursor, Claude Code, VS Code or any MCP client read your app's bug reports, get a plain-English diagnosis with a fix to start from, and dispatch or merge fixes. Your editor's own model does the work, so you need no second LLM key.

Docs: MCP guide · tool reference

flowchart LR
    A["Bug reports<br/>in Mushi"] --> B["This MCP server"] --> C["Your editor<br/>'what's broken in prod?'"] --> D["Fix: you, or a<br/>draft PR from an agent"]

Quick start

npx mushi-mushi setup --ide cursor   # or: claude, continue, zed

Restart the editor and ask "what's broken in prod?". For Cursor and Claude, setup writes the hosted server entry: the editor opens a browser sign-in on first use, and no API key is stored on disk. Pass --stdio for a local server, or --with-rules to also write the lesson-library rules file. You can also use the one-click Add to Cursor button on the connect page.

Manual setup (any MCP client)
{
  "mcpServers": {
    "mushi-mushi": {
      "command": "npx",
      "args": ["-y", "@mushi-mushi/[email protected]"],
      "env": {
        "MUSHI_API_KEY": "mushi_xxxxxxxxxxxxxxxxxxxx",
        "MUSHI_PROJECT_ID": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
      }
    }
  }
}
VariableWhere to find it
MUSHI_API_KEYConsole → Connect or MCP: mint an mcp:read or mcp:write key. Not the SDK key from Setup → Verify, and not an AI key from Settings
MUSHI_PROJECT_IDConsole → Projects → the UUID under your project's name
MUSHI_API_ENDPOINTOptional. Only for self-hosting, e.g. http://localhost:54321/functions/v1/api
MUSHI_FEATURESOptional. Tool groups to expose (see below)

Claude Desktop does not expand ${VAR} references in this file; Cursor and VS Code expand ${env:MUSHI_API_KEY} and Claude Code expands ${MUSHI_API_KEY}. Keep the key out of any mcp.json you commit.

What your editor gets

Start hereWhat it does
triage_next_stepsWhat needs attention right now, from the live queue
get_fix_contextOne brief for a report: root cause, repro, the smallest set of files
triage_issueEverything about one report: evidence, similar bugs, blast radius, logs, next actions
dispatch_fix / merge_fixHave an agent open a draft PR, then merge it and close the report
diagnose_setupExplains a broken install and the one next step
search_mushi_docsAnswers setup and API questions from the docs

The full catalog has 129 tools, plus 8 resources (project://dashboard, project://stats, …) and 4 prompts (summarize_report_for_fix, triage_next_steps, …). Every tool carries read-only and destructive hints, so your editor can ask before it changes anything. Full list: MCP tool reference.

Tool groups. New installs get a lean default: triage, fixes, inventory, setup, docs. Turn on more with MUSHI_FEATURES=triage,qa,codebase (stdio) or ?features=… on the hosted URL, or everything with all. Groups: qa, skills, codebase, rewards, admin, audit, usage. Add ?read_only=1 to the hosted URL to hide every write tool.

Keys and safety

ScopeCan doGive it to
report:writeSend reports only. No admin accessYour app's SDK. Never an MCP client
mcp:readEvery read tool and resourceMost agent loops
mcp:writeReads plus dispatch_fix, merge_fix, transition_status and other changesAgents that should act on bugs

A key without the needed scope gets a clear INSUFFICIENT_SCOPE error, never a silent failure. Your SDK key ships in your app's bundle, so it must never carry mcp:write; the console mints MCP keys separately. Never use a Supabase service-role key. If a laptop is lost, rotate the key in the console.

Troubleshooting

  • No tools showing? Check the MCP panel for a green dot, then fully restart the editor after any config change.
  • Keep one config. Put the Mushi entry in your global ~/.cursor/mcp.json only; a second copy in a project file can cause connection storms.
  • Windows paths in JSON need forward slashes (C:/Users/...) or escaped backslashes.
  • Still stuck? Ask your editor to run diagnose_setup, or run mushi doctor --mcp.

Programmatic use

ImportExports
@mushi-mushi/mcp/servercreateMushiServer() for a custom host
@mushi-mushi/mcp/catalogTOOL_CATALOG, scopes and tool specs
@mushi-mushi/mcp/feature-groupsFeature-group filtering
@mushi-mushi/mcp/clientsThe supported editors and how each one installs
@mushi-mushi/mcp/brandingIcons and server metadata

Learn more

MCP guide · MCP quickstart · The incident loop · Fix orchestrator

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.