Skip to content
MCP server

Tently

Let coding agents look up team decisions and change impact before they write code

Runs locallyStatus and history
Install
Pricing
Freemium
Platforms
CLI, API
Runs
Locally
Version
0.1.4
Listing updated
2 days ago

About Tently

Tently is a freemium MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Tently's pull request reviewer checks each change against decisions a team already made and against the callers a diff never touched. This MCP server is the earlier of its two gates: it puts the same decision store inside Claude Code, Cursor or Codex while the agent is still writing.

Two tools do the work. One searches the decisions that apply to the code about to change, scoped to the repository in your checkout, and each result says whether breaking it is blocking or advisory. The other computes the blast radius of a symbol, meaning who calls it and what it depends on, locally on your working tree with CodeGraph, so source code stays on your machine. The README warns that impact results are evidence, not proof, because callers reached through interfaces, dynamic dispatch or re-exports can be missed.

Setup is one command, npx @tently/mcp init, which edits .mcp.json or .cursor/mcp.json, installs a small skill and backs up files first. It needs a Tently workspace API key. The service is in early access for GitHub teams. The Solo plan is free at 30 reviewed PRs a month, and the MCP gate is unlimited on every plan.

Install

Run in a terminal

claude mcp add tently-mcp --env TENTLY_API_KEY='<TENTLY_API_KEY>' -- npx -y @tently/mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "tently-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@tently/mcp"
      ],
      "env": {
        "TENTLY_API_KEY": "<TENTLY_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "tently-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@tently/mcp"
      ],
      "env": {
        "TENTLY_API_KEY": "<TENTLY_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "tently-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@tently/mcp"
      ],
      "env": {
        "TENTLY_API_KEY": "<TENTLY_API_KEY>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "tently-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@tently/mcp"
      ],
      "env": {
        "TENTLY_API_KEY": "<TENTLY_API_KEY>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "tently-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@tently/mcp"
      ],
      "env": {
        "TENTLY_API_KEY": "<TENTLY_API_KEY>"
      }
    }
  }
}

Configuration

  • TENTLY_API_KEYTently workspace API key, from the Connect page at app.tently.devRequired
  • TENTLY_API_URLTently API origin (default https://api.tently.dev)

Tools this server provides (2)

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

search_decisions
Returns team decisions that apply to the code about to change, marked blocking or advisory.
query_impact_zone
Computes the blast radius of a symbol locally on the checkout.

What it does

Decision lookup
Finds the team decisions that apply to files about to change, marked blocking or advisory.
Local impact analysis
Traces callers and dependencies of a symbol on your own checkout.
One-command setup
Registers the server and installs a skill for Claude Code or Cursor.
PR review gate
The paired GitHub App posts check runs and inline comments with evidence.
Decisions mined from history
Proposes decisions from merged PRs, review threads and docs for the team to approve.

Good for

  • Have an agent check team decisions before editing a module
  • Find which callers a signature change would break
  • Keep generated code consistent with past design decisions
  • Review a pull request against decisions and downstream consumers

Pricing

Freemium, from $99 a month

Solo free (30 PRs/month); Crew $99/mo; Fleet $399/mo; early access

Prices change. Check tently.dev before you buy.

Tently FAQ

What do I need?

A Tently workspace and an API key from its Connect page; the service is in early access.

Does my source code leave my machine?

Impact checks run locally. Decision search sends the query text, file paths and repository owner and name.

Which agents does init support?

Claude Code and Cursor, and the site also mentions Codex.

What does it cost?

Solo is free with 30 PRs a month; Crew is $99 a month and Fleet $399 a month, with seats and the MCP gate unlimited.

Is Tently free?

Tently has a free plan, with paid plans for more usage or features. Paid plans start at $99 a month.

How do I add Tently to Claude Code?

Run this command in your terminal: claude mcp add tently-mcp --env TENTLY_API_KEY='<TENTLY_API_KEY>' -- npx -y @tently/mcp

What are the best alternatives to Tently?

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

Alternatives to Tently

  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