As A Judge MCP Server
MCP as a Judge is a behavioral MCP that strengthens AI coding assistants by requiring explicit LLM evaluations for: Research, system design, and planning Code changes, testing, and task-completion verification. Its 8 documented tools cover coding, judge, task, raise. It runs locally over stdio via the published package.
People who work with coding, judge and task and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Python.
VERIFIED ACTIVE
LAST COMMIT 2025-12-15 · ★ 17 · #69 OF 106 MAINTAINED AI & ML · VERIFIED 2026-09-18
MIT · Python servers · how we verify → /methodology
01 · Install As A Judge
before you install - you'll need
LLM_API_KEY, LLM_MODEL_NAME are optional or environment-specific per the README.
Claude Code
claude mcp add othervibes-mcp-as-a-judge -- uvx mcp-as-a-judge Claude Desktop / Cursor / VS Code - add to config
{
"mcpServers": {
"othervibes-mcp-as-a-judge": {
"command": "uvx",
"args": [
"mcp-as-a-judge"
]
}
}
} Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.
Using another client? Same JSON, different key
Claude Desktop · mcpServers
Cursor · mcpServers
VS Code · servers
Windsurf · mcpServers
Zed · context_servers
Cline · mcpServers
Roo Code · mcpServers
Continue · mcpServers
LibreChat · mcpServers
Gemini CLI · mcpServers
Codex CLI · mcp_servers
Full setup guides: every client.
02 · Evidence
Security posture
What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.
runs as local process (stdio) - runs on your machine with your user's permissions
license MIT - declared in the repository
pypi package mcp-as-a-judge - check the name against the project README before installing (PyPI has no namespace ownership)
registry namespace io.github.OtherVibes is GitHub-verified and matches the repo owner
03 · What As A Judge can do
Prose above is summarized from the project's README and registry record - no invented capabilities.
What you can build
With this server connected, an agent can validate plan/design, and validate tests using real runner output and optional coverage.
Capability map
Tools grouped from the project's README - what As A Judge lets an agent do.
Judge
4 tools - e.g. Validates plan/design; requires library selection and internal reuse maps; flags risks; Reviews unified Git diffs for correctness, reuse, security, and code quality; Validates tests using real runner…
judge_coding_plan · judge_code_change · judge_testing_implementation · judge_coding_task_completion
Raise
2 tools - e.g. Elicits missing details and decisions to unblock progress; Engages the user on trade‑offs, constraints, and enforced changes
raise_missing_requirements · raise_obstacle
update
1 tool - e.g. Creates/updates task metadata; classifies task_size; returns next-step workflow guidance
set_coding_task
read & search
1 tool - e.g. Recovers the latest task_id and metadata to resume work safely
get_current_coding_task
Latest releases
v0.5.0 · 2025-09-29
bump version to 0.5.0 · implement human plan approval workflow and enhanced judge response handling (#31) · implement human plan approval workflow and enhanced judge response handling
v0.4.0 · 2025-09-24
bump version to 0.4.0 · bind workflow guidance to plan system prompt with structured field specs (#28) · bump actions/download-artifact from 4 to 5 (#30) · bump actions/create-github-app-token from 1 to 2 (#29) · bump…
v0.3.20 · 2025-09-20
bump version to 0.3.20
04 · Who maintains As A Judge
mcp-as-a-judge is maintained by othervibes. It's the only MCP server we track from this author; the repo dates to Aug 2025.
05 · Facts
- repository
- github.com/othervibes/mcp-as-a-judge
- category
- AI & ML - ranked #69 of 106 actively-maintained AI & ML servers as of 2026-09-18.
- registry
- io.github.OtherVibes/mcp-as-a-judge (active, first published 2025-09-18 · 6 versions)
- packages
- pypi:mcp-as-a-judge
06 · As A Judge FAQ
Is As A Judge still maintained?
Yes - as of 2026-09-18, its last commit was 2025-12-15. We re-verify nightly.
What can As A Judge do?
With this server connected, an agent can validate plan/design, and validate tests using real runner output and optional coverage.
How do I install As A Judge?
Run `uvx mcp-as-a-judge`. The README documents 2 environment variables (LLM_API_KEY, LLM_MODEL_NAME) to set first. LLM_API_KEY, LLM_MODEL_NAME are optional or environment-specific per the README. You can also paste the ready-made client config above.
Does As A Judge run locally?
Yes - it's a stdio server: it runs on your machine (via uvx) with your user's permissions. Your data stays local unless the server itself calls external APIs.
07 · Alternatives to As A Judge
Alternatives to As A Judge
Maintained AI & ML servers if As A Judge isn't the fit.
- Fun ASR Transcribe local audio with FunASR and SenseVoice using private, on-device inference. ★ 20,411 · 2026-09-18
- Auto TS Automated time series forecasting with model search, anomaly detection, and event risk analysis ★ 1,428 · 2026-08-25
- Haiku RAG Local-first agentic RAG with citations - hybrid search, reranking, multimodal document retrieval ★ 609 · 2026-09-17
- Plan Exe MCP server for generating rough-draft project plans from natural-language prompts. ★ 401 · 2026-09-13
- Local RAG (shinpr) Easy-to-setup local RAG server with minimal configuration ★ 398 · 2026-09-18
- PageIndex MCP Reasoning-based RAG system for chatting with long PDFs. Supports local and online files. ★ 388 · 2026-07-25
Pairs well with
Servers that cover what As A Judge doesn't - only shown when the pairing reason fits the companion.
More AI & ML MCP servers · Pipes Hub · Ncp · Prompt Branch · Promptibus MCP
More Python MCP servers · Ouestcharlie Woof · Cloak MCP · Watch Skill · Global Agent Memory · Northdata MCP · see all