mushi-mushi
About
Your AI shipped it. Mushi tells you why it broke — plain diagnosis and a paste-ready fix prompt in Cursor. No second LLM key.
Details
- Transport
- SSE
Explore
- Turns user bugs into diagnosis and fix prompt
- Works from Cursor, Claude Code, and VS Code
- Tools: triage, get_fix_context, QA runs, codebase search
- No second LLM key required
- Free tier available for triage and fix context
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
mushi-mushiCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
get_recent_reports
List recent bug reports with optional filters (status / category / severity). Use this to survey what the triage queue looks like right now.
get_report_detail
Full payload for a single report — description, console logs, network requests, screenshot URL, classification, fix history.
search_reports
Semantic + keyword search over reports. Uses pgvector similarity server-side — falls back to substring match when embeddings are unavailable.
get_fix_context
Bundle the full context an agent needs to fix a bug: report detail, reproduction steps, component, root cause, ontology tags, AND the inventory expected_outcome contract (whitepaper §2.10) when one is declared.
get_fix_timeline
Ordered timeline of a fix attempt — dispatched → started → branch → commit → PR opened → CI → completed/failed.
inventory_get
Current inventory.yaml snapshot for a project (latest ingest, validation errors, per-action status summary).
inventory_findings
Latest gate runs + findings (dead-handler, mock-leak, crawl, status-claim, agentic-failure). Filter by gate name or severity.
graph_node_status
Fetch a single graph node row (label, type, metadata — includes derived status).
get_blast_radius
Return the blast radius for a codebase node: all downstream dependents and their weighted impact scores. Use this to understand how far a change (or bug fix) propagates before deciding to dispatch a fix.
get_knowledge_graph
Traverse the codebase knowledge graph from a seed node. Returns nodes and edges within the given depth. Depth 1 = direct dependents; depth 2 = transitive. Max depth 3.
run_nl_query
Natural-language question → SQL query run against your project data. Read-only, rate-limited, no privileged schemas.
dispatch_fix
Dispatch the Mushi agentic fix orchestrator for a classified report. Returns a dispatch id; subscribe to /v1/admin/fixes/dispatch/:id/stream for live progress.
transition_status
Move a report between workflow states (new → classified → grouped → fixing → fixed → dismissed).
query_lessons
Token-budget retrieval of relevant learning rules (lessons) for a given code diff or PR context. Returns ranked lessons packed within max_tokens using bi-encoder retrieval + severity-weighted scoring. Use this before opening a PR, writing a fix, or asking "what mistakes should I avoid in this area of code?"
list_lessons
List promoted learning rules (lessons) for the current project. Each lesson represents a named pattern of mistakes that has been encoded from bug reports. Use this to understand what systemic issues have been identified and encoded as heuristics.
setup_check
Run the 4 **dispatch-readiness** checks for a project and return their pass/fail status (GitHub repo connected, codebase indexed, Anthropic BYOK key present, autofix enabled). Also returns the target repo URL when GitHub is connected. Use this before calling dispatch_fix to understand why a dispatch might fail. For SDK ingest health (API key → heartbeat → first report), call ingest_setup_check instead.
ingest_setup_check
Run the 4 **required ingest** checks for the project tied to this API key: project exists, active API key, SDK heartbeat (or real report), and at least one ingested report. Returns per-step pass/fail plus last_sdk_seen_at and endpoint host diagnostics. Use after wiring env vars or pasting the SDK snippet to confirm the banner will work.
diagnose_setup
Single entry point for setup health. mode=full (default) runs ingest + dispatch checks; mode=ingest runs SDK ingest checks only; mode=dispatch runs fix-dispatch preflight only. Prefer this over setup_check, ingest_setup_check, or diagnose_connection alone.
search_mushi_docs
Search official Mushi docs (guides, MCP setup, inventory, QA, skills) by keyword. Returns ranked page titles, URLs, and excerpts.
list_projects
List the Mushi projects accessible to this API key. Returns project id, name, and created date. For multi-project tokens this lists all accessible projects; for single-project keys it returns only the bound project.
get_project_context
Return a rich context snapshot for a project: ingest health, SDK heartbeat, autofix readiness, open-report counts, and active integrations. Combine with get_recent_reports before triaging.
get_pipeline_logs
Pull recent log entries from the Mushi pipeline services (fix-worker, pipeline, qa-story-runner). Accepts project_id, service, since (ISO timestamp), limit (max 200), level (all/info/warn/error) filters.
get_report_evidence
Return the focused evidence package for a single bug report: screenshot URL, console logs, network excerpts, environment info, user comments, and browser/OS data. Lighter than get_report_detail — skips the full classification/fix history.
triage_issue
Read-only orchestration tool that combines report detail, evidence, similar bugs, fix context, blast radius, recent pipeline logs, and recommended next actions into a single triage packet. This is the primary entry point for agent-driven bug investigation.
get_usage
Read-only diagnoses quota and billing summary for the current project: diagnoses used / limit / percentage, spend cap, period start/end, plan name, and whether the project is approaching or over its quota. Use this to answer "how many diagnoses do I have left?" or "am I close to my spend cap?".
get_report_timeline
Ordered timeline for a report (comments, fixes, QA, status).
get_similar_bugs
Find similar bugs via pgvector search.
inventory_diff
Diff two inventory commits.
graph_neighborhood
BFS graph neighborhood around a node.
fix_suggest
Stage-2 suggested fix slice for a report.
diagnose_connection
Validate MCP credentials and ingest/dispatch readiness.
submit_fix_result
Record a fix outcome from an external agent.
trigger_judge
Run Sonnet-as-Judge batch.
test_gen_from_report
Generate Playwright test from report.
merge_fix
Squash-merge a fix PR.
refresh_ci
Refresh GitHub CI status for a fix.
reopen_report
Reopen a report for regression triage.
list_top_contributors
Top contributors leaderboard.
award_bonus_points
Award bonus points to a contributor.
set_tier
Override contributor tier.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mushi-mushi": {
"mushi-mushi": {
"command": "npx",
"args": [
"-y",
"@mushi-mushi/mcp@latest"
],
"env": {
"MUSHI_API_KEY": "<YOUR_MUSHI_API_KEY>",
"MUSHI_PROJECT_ID": "<YOUR_PROJECT_UUID>",
"MUSHI_API_ENDPOINT": "https://dxptnwrhwsqckaftyymj.supabase.co/functions/v1/api",
"MUSHI_FEATURES": "triage,fixes,inventory,setup,docs"
}
},
"mushi-mushi-hosted": {
"type": "http",
"url": "https://dxptnwrhwsqckaftyymj.supabase.co/functions/v1/mcp?features=triage,fixes,inventory,setup,docs",
"headers": {
"Authorization": "Bearer <YOUR_MUSHI_API_KEY>",
"X-Mushi-Project-Id": "<YOUR_PROJECT_UUID>"
}
}
}
}
}McpServers
{
"mushi-mushi": {
"command": "npx",
"args": [
"-y",
"@mushi-mushi/mcp@latest"
],
"env": {
"MUSHI_API_KEY": "<YOUR_MUSHI_API_KEY>",
"MUSHI_PROJECT_ID": "<YOUR_PROJECT_UUID>",
"MUSHI_API_ENDPOINT": "https://dxptnwrhwsqckaftyymj.supabase.co/functions/v1/api",
"MUSHI_FEATURES": "triage,fixes,inventory,setup,docs"
}
},
"mushi-mushi-hosted": {
"type": "http",
"url": "https://dxptnwrhwsqckaftyymj.supabase.co/functions/v1/mcp?features=triage,fixes,inventory,setup,docs",
"headers": {
"Authorization": "Bearer <YOUR_MUSHI_API_KEY>",
"X-Mushi-Project-Id": "<YOUR_PROJECT_UUID>"
}
}
}Why it broke — in your editor
Mushi turns user-felt bugs into a plain diagnosis and a paste-ready fix prompt. Works from Cursor, Claude Code, or VS Code over MCP. Install: https://kensaur.us/mushi-mushi/docs/connect · npx mushi-mushi setup --ide cursor Tools: triage reports, get_fix_context, QA story runs, codebase search (when indexed). No second LLM key. Free tier covers triage + fix context.Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
