Skip to content
MCP server

Signalint

Grouped, prioritized JS/TS diagnostics for coding agents, run locally over stdio

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
1.2.1
Listing updated
2 days ago

About Signalint

Signalint is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Instead of pasting thousands of raw compiler lines into an agent's context, Signalint reads the output of the engines your project already uses (Oxlint, tsc, Biome, ESLint) and returns bounded clusters ranked from P1 to P5. Repeated rules are grouped by scope, and full issue detail is fetched only when the agent asks for it.

The workflow is built around baselines. An agent saves a check, makes an edit, then compares the new result to see which diagnostics were introduced, resolved or left unchanged. A loop tracker flags fixes that keep oscillating, and the README says diagnostic checks are read-only by default, with patch application as a separate explicit operation.

The README describes 13 typed tools, monorepo-aware planning for pnpm workspaces, host hooks for Claude Code, Cursor, Codex and VS Code, and ingestion of diagnostics from LSP or other compilers. It needs Node 22.12 or later and does not bundle any engines, so your project must have its own installed.

Install

Run in a terminal

claude mcp add signalint -- npx -y signalint-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "signalint": {
      "command": "npx",
      "args": [
        "-y",
        "signalint-mcp"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "signalint": {
      "command": "npx",
      "args": [
        "-y",
        "signalint-mcp"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "signalint": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "signalint-mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "signalint": {
      "command": "npx",
      "args": [
        "-y",
        "signalint-mcp"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "signalint": {
      "command": "npx",
      "args": [
        "-y",
        "signalint-mcp"
      ]
    }
  }
}

Tools this server provides (9)

Taken from the project documentation; the running server may differ.

ping
Verifies the server responds and reports the canonical project root.
check_project
Full project or workspace scan across all enabled engines.
check_files
Incremental check of specific files using content-hash caching.
get_issue_detail
Returns full issue records anchored to a check ID.
get_loop_status
Reports oscillating fix signatures and rule churn in the session.
get_diagnostic_snapshot
Retrieves state and metrics for an immutable check snapshot.
compare_diagnostics
Computes the delta between a baseline check and a current check.
after_edit_check
Checks edited files and diffs them against a baseline.
ingest_diagnostics
Ingests external LSP or compiler diagnostics.

What it does

Priority-ranked clusters
Groups repeated rule hits and orders them from P1 to P5 so the agent starts with systemic errors.
Verification deltas
Compares a baseline check with a later one and lists introduced, resolved and unchanged diagnostics.
Cached incremental checks
Reuses results when the engine's cache key matches and checks only changed files.
Fix-loop detection
Reports oscillating fix signatures and rule churn across a session.
Workspace planning
Orders pnpm workspace packages topologically using manifests and TypeScript project references.
Staged patch apply
Previews replacements in memory, then applies them with drift checks and best-effort rollback.

Good for

  • Shrink noisy lint and type-check output before it reaches a coding agent
  • Confirm an edit fixed an error without introducing new ones
  • Catch an agent repeatedly flipping the same lint rule
  • Run verify-after-change hooks in Claude Code or Cursor
  • Plan checks across a pnpm monorepo

Pricing

Open source

MIT licensed; free to run locally

Prices change. Check tranqui004.github.io before you buy.

Signalint FAQ

Does it ship its own linters?

No. It detects the TypeScript, Oxlint, ESLint or Biome installs in your project and ships none of them.

Which Node version does it need?

Node 22.12 or later, because it uses the built-in node:sqlite module.

Does a check change my source files?

Diagnostic checks are read-only. Applying patches is a separate explicit operation.

How do I add it to a client?

Add a stdio entry that runs npx -y signalint-mcp, or run npx signalint init to configure installed clients.

Is Signalint free?

Yes. Signalint is open source, so you can use and self-host it for free.

How do I add Signalint to Claude Code?

Run this command in your terminal: claude mcp add signalint -- npx -y signalint-mcp

What are the best alternatives to Signalint?

The closest alternatives to Signalint are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Signalint

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools