Jev.aitools.fyi

Command Palette

Search for a command to run...

Devin Desktop + Jev

Last updated

Windsurf became Devin Desktop. MCP reflex beside Devin Local, Claude, and Codex agents.

Devin Desktop and the Windsurf rename

Cognition rebranded Windsurf to Devin Desktop: same IDE foundation, new Agent Command Center (Spaces, boards, multi-session management). devin.ai/desktop states existing plans, extensions, and settings migrate over the air.

JetBrains Windsurf builds remain available separately per their FAQ.

ACP and multiple agents

Devin Desktop advertises Agent Client Protocol (ACP) support so you can run Devin Local, Claude, Codex, and custom agents in one surface. Jev does not replace those models. It sits in MCP tools they can all call: one reflex install, many agent personalities.

Verified path: system1-mcp

We did not find a first-party Devin-native Jev plugin in public READMEs. The verified integration is MCP: system1-mcp install lists Windsurf in its autodetected IDE set (the pre-rename name Devin Desktop inherited).

Use uvx system1-mcp install or manual mcpServers blocks per your host's MCP docs.

export TYPESAFE_API_KEY="tsk_..."
uvx system1-mcp install
uvx system1-mcp doctor

Manual MCP entry

Devin Desktop continues Windsurf-style MCP configuration for extensions and MCP servers shown on the marketing site (Slack, Linear, Stripe, and others). Add a local stdio server alongside them:

{
  "mcpServers": {
    "system1": {
      "command": "uvx",
      "args": ["system1-mcp"],
      "env": {
        "TYPESAFE_API_KEY": "${env:TYPESAFE_API_KEY}"
      }
    }
  }
}

Community MCP listings

Devin Desktop's ecosystem page highlights third-party MCP servers. itsmostafa/typesafe-mcp and blakestone-x/jev-mcp are directory listings you can register the same way as other MCP integrations. Read each README for tool names (evaluate vs classify vs fast_guard).

Spaces and shared context

Spaces share git worktrees and context across agents. Put AGENTS.md reflex rules in the repo so every agent session in a Space sees the same fast_guard instructions, not just one local chat thread.

Environment keys

Create a key at https://console.typesafe.ai/ (or route through Vercel AI Gateway / OpenRouter where the integration documents it). Export TYPESAFE_API_KEY in the environment for Devin Desktop MCP subprocesses.

Some MCP clients accept AI_GATEWAY_API_KEY when using gateway-hosted Jev models; follow that integration's README. Never commit keys into repos or paste them into chat logs.

export TYPESAFE_API_KEY="tsk_..."
# Optional pin:
export TYPESAFE_MODEL="jev-latest"

Pitfalls

Installer strings may still say Windsurf while the product says Devin Desktop; rerun doctor after major upgrades. Do not assume cloud Devin sessions share your laptop MCP config.

MCP tools are advisory: Spaces with parallel agents need consistent policy text or one agent may approve what another blocked.

Related directory listings

  • Agent toolingGo62
    MCP
    Go CLI and single-binary MCP server exposing TypeSafe judgments to Claude Desktop, Claude Code, and Codex.

    by @0xLogicrw

    agent toolingmcpagents
  • Agent toolingPython7
    MCP
    MCP server for TypeSafe Jev: typed classify, score, check, match and screen for any agent, with confidence on every answer.

    by @0xLogicrw

    agent toolingmcpagents
  • Agent toolingTypeScript4
    MCP
    Brainwires jevwire packages a Jev decision layer for agents: Streamable MCP server, embeddable DecisionModel library, and a Claude Code plugin marketpla
    agent toolingmcpagents
  • Official205
    OfficialDemoFeatured
    Official Claude Code / Codex skill pack for Jev primitives, patterns, and evaluation shape.
    officialagentseval

Related agent guides

FAQ

Is there an official Devin Jev extension?
We did not find one in public READMEs indexed here. Use MCP servers documented on PyPI or the catalog.
Windsurf MCP configs after rename?
Devin Desktop FAQ says settings carry over OTA. Validate MCP entries with system1-mcp doctor after updating.
Devin Local vs Desktop Jev?
Cloud Devin products use different connector models. This guide focuses on Desktop MCP on your machine.
Can I use LangChain middleware instead?
Only inside Python services you control. Desktop agents need MCP or HTTP tools, not in-process LangChain middleware.
Which tool preset should I pick?
system1-mcp fast_* tools for IDE reflexes. typesafe-mcp evaluate when you design custom parallel questions.

Primary sources

By Rishit