Dispatch, watch, verify

Dispatch it.
Prove it.

Dispatch, watch, and verify AI agent tasks from your phone. TwoThumbs turns a raw ask into an evidence-backed brief, challenges weak premises, routes bounded execution, then verifies the artifact and downstream result that actually ran.

Works withClaude Code · Codex · Cursor · CI / headless

Any Google account opens the demo console. No credit card, no TOTP.

https://twothumbs.co/mcp

Remote MCP over HTTPS · no local install

Contact form / 0042● ●
  • 01Prompt engineeredEVIDENCE
  • 02Premise challengedCRITIC
  • 03Artifact executedBOUNDED
  • 04Outcome verifiedPROVEN
Ship it

The loop

See the whole run.
Not a black box.

TwoThumbs is a console for running and verifying AI agent work. Dispatch a task, watch the agents work in real time, approve steps when they need a human, and get a clear pass or fail verdict on the result, all from your phone. Every run is a loop you can see: what was asked, what the agents did, what they produced, and whether it actually worked. No black box, no waiting blind.

01

Real-time task stream

See each step as it happens.

02

Approval cards

The console asks before it does anything that needs your call.

03

Verdicts

A straight answer on whether the work passed, with the evidence.

04

Your workspace is yours

Isolated, private, yours to control.

Remote MCP / quick start

Connect in one minute.

TwoThumbs is a remote Streamable HTTP MCP endpoint. No local package or installer is required.

Choose Streamable HTTP in an MCP-compatible client and use this HTTPS endpoint:

https://twothumbs.co/mcp

The endpoint is reachable worldwide over HTTPS where your client supports remote Streamable HTTP MCP and permits outbound access. Automated or paid use requires a valid key; network policy and client support still apply.

Have an access code? Add the MCP, then choose “Redeem a code” in its authorization screen or call redeem_code. Store the one-time returned key securely.

All setup and CI options →

Better prompts in.
Verified outcomes out.

01

The prompt engineer

Checks the premise, retrieves relevant evidence, exposes contradictions, names the right lane, and writes completion criteria an executor cannot hand-wave.

Open the 60-second quickstart →
02

The verifier

Exercises the real surface and follows the work downstream. Mechanical success, visual comprehension, and the actual outcome are separate gates.

Evidence, not self-scoring

“Submitted” is not
the same as done.

Click

Every visible affordance, across desktop and mobile.

Submit

Every form, with clearly marked synthetic data.

Prove

The row landed, the email arrived, the webhook fired.

TwoThumbs came from real bugs that sat behind polished interfaces for months—bugs AI code review never saw because the code looked plausible. The critic uses outcomes as evidence.

Keep verification
on speed dial.