Skip to content
MCP server

Agnes

Read frontend code-health reports and rank fixes by points gained, via Agnes

Status and history

Answered our latest tool-list check 16 hours ago with 6 tools. It will be marked verified after 3 successful checks in 7 days.

Install
Pricing
Freemium
Platforms
API, Web
Runs
Remotely
Version
1.0.0
Listing updated
3 days ago

About Agnes

Agnes is a freemium MCP server for developer tools. It runs as a hosted remote server, so there is nothing to install locally.

Agnes scores frontend repositories, and this server lets an AI agent read those scores for your team. The agent can list projects, open a report, see which checks did not run and pull findings 50 per page, filtered by rule, category, severity or folder, each with file, line and a link to the rule.

The most useful call for planning is rank_fixes, which orders rules by the points their findings cost and shows how many points fixing each would gain. The agent can also fetch get_scoring so it can explain how findings turn into scores, and get_sbom for a CycloneDX software bill of materials with every installed package, its licence and known vulnerabilities.

Access is read-only and limited to the team a token was created in. Claude and ChatGPT can connect as a custom connector and sign in, or you can create a token under Settings, API and agents. An optional skill file walks the agent through reading the report, checking each finding in your code and writing a plan for you to approve before it changes anything.

Install

Run in a terminal

claude mcp add --transport http agnes https://app.agnes.dev/api/mcp

Add to Settings → Connectors → Add custom connector

https://app.agnes.dev/api/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "agnes": {
      "url": "https://app.agnes.dev/api/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "agnes": {
      "type": "http",
      "url": "https://app.agnes.dev/api/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "agnes": {
      "serverUrl": "https://app.agnes.dev/api/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "agnes": {
      "type": "http",
      "url": "https://app.agnes.dev/api/mcp"
    }
  }
}

Tools this server provides (6)

Read directly from the server.

list_projects
The projects of your team: each repository branch Agnes analyses, with its latest score.
get_report
The overall score, the score per category, the findings per category and severity, and which checks did not run, for a project's latest successful analysis.
rank_fixes
The rules whose findings cost the project the most points. Each entry says how many points the overall score and its category would gain if every finding of that rule were fixed (the report rescored without them), with one example finding.
get_findings
A project's findings, most severe first, 50 per page, with file, line, message and a documentation link where there is one. Filter by rule, category, severity or file (or folder).
get_sbom
The software bill of materials of a project's latest successful analysis (or the given run) as CycloneDX 1.6 JSON: every installed package with its purl, licence where known and dependencies, and the known vulnerabilities that affect them. Large for big projects; save it to a file rather than reading it all.
get_scoring
How Agnes turns findings into scores: the categories, their weights in the overall score, and how severity, confidence and codebase size count.

What it does

Ranked fixes
Lists rules by the points their findings cost and how many points each fix would gain.
Paged findings
Returns 50 findings per page, filterable by rule, category, severity or folder, with file and line.
Per-category report
Shows the overall score, scores per category and which checks did not run.
SBOM export
Returns a CycloneDX bill of materials with packages, licences and known vulnerabilities.
Read-only access
Tokens only read reports for the team they were created in and can be revoked at any time.
Plan-first skill
An optional skill asks the agent to write a plan and wait for approval before changing code.

Good for

  • Ask an agent to plan how to raise a repository's health score
  • Find which rule violations cost the most points
  • Review a project's dependency licences and vulnerabilities
  • Check which findings sit in a particular folder
  • Explain to a team how scores are calculated

Pricing

Freemium, from $20 a month

Free plan; Essentials $20/mo, Pro $50/mo, Premium $100/mo

Prices change. Check agnes.dev before you buy.

Agnes FAQ

Does the agent change my code or Agnes settings?

No. The tools only read reports, and the optional workflow asks the agent to wait for your approval before changing code.

How do I connect?

Add the address https://app.agnes.dev/api/mcp as a custom connector and sign in, or use a token with a Bearer header in Claude Code, Cursor or VS Code.

What does Agnes cost?

There is a free plan; paid plans are Essentials at $20, Pro at $50 and Premium at $100 per month.

Is open source analysed for free?

Yes. On every plan up to 10 public repositories of up to 100,000 lines each are analysed free.

Is Agnes free?

Agnes has a free plan, with paid plans for more usage or features. Paid plans start at $20 a month.

How do I add Agnes to Claude Code?

Run this command in your terminal: claude mcp add --transport http agnes https://app.agnes.dev/api/mcp

What are the best alternatives to Agnes?

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

Alternatives to Agnes

  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