ThinkerWave

io.github.thinkerwave-aiPublished by GitHub account thinkerwave-aiv0.1.2Updated Oct 27, 2026

Autonomous research agent for hard problems that labels every claim verified, assumption or guess.

The publisher's own server: Wave AI

VerifiedSTDIODesktop onlyData & AnalyticsAI & ML

Overview

AI-generated overview

ThinkerWave hands hard, open-ended research questions to a cloud autonomous research agent and returns answers with per-claim evidence labels.

What you can do
  • Submit a research problem together with the facts you are confident about, and optionally constraints and a number of rounds.
  • Receive the best answer found, a confidence level, and the evidence behind each claim, labelled verified, assumption or guess.
  • Read back charts and tables produced during the run.
  • Iterate over rounds, with the agent checking its own work and retrying while keeping the best result.
When to use it
Useful when a question has no single right answer and needs multi-round research and quantitative reasoning, such as pricing and churn scenarios, credit risk stress tests, market entry or due diligence. Less useful for quick factual lookups or tasks that need only local files.
Requirements
A local Python environment; the README installs the package with pip (or the MCP extra) and the manifest runs it via uvx over stdio. A ThinkerWave account and an API key supplied in the THINKERWAVE_API_KEY environment variable. Network access, since the research engine runs in ThinkerWave's cloud.
Before you install
The API key is a secret and is passed as the THINKERWAVE_API_KEY environment variable; keep it out of shared configs. The problem statement and the facts you supply are sent to ThinkerWave's cloud service for processing. Runs may be subject to account limits and free-run quotas, so check the documented limits before relying on it heavily.

Installation

In SourceWeft

  1. Open ThinkerWave in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

ThinkerWave

[PyPI] [Python] [License] [MCP server] [CI]

An autonomous research agent for problems without a single right answer — credit risk stress tests, pricing and churn, market entry, due diligence, drug program selection.

You give it the problem and the facts you're sure of. It researches, runs the numbers, checks its own work and tries again, round after round, keeping the best answer. You get the answer, how confident it is, and the evidence behind each claim — verified, assumption or guess — plus any charts and tables it produced.

[A real two-round ThinkerWave run in the terminal]

bash
pip install thinkerwaveexport THINKERWAVE_API_KEY=tw_live_…      # create one at thinkerwave.ai/app → Settings → API keystw run "Should we raise prices 15%, 8%, or hold? 1,200 customers at \$85/month, 3.1% monthly churn." \  --given "last 8% increase spiked churn to 4.6% for three months" --rounds 6 --watch

From Claude Code or Cursor: pip install "thinkerwave[mcp]", then claude mcp add thinkerwave -- thinkerwave-mcp — your coding assistant can now hand ThinkerWave a hard question and read back the evidence.

From Python: see examples/.

What's in this repo: the Python SDK, the tw command line, an MCP server, the API reference and the answer/evidence format. ThinkerWave itself runs in ThinkerWave's cloud; this repository is the way in, not the engine.

Free runs are included with every account. Limits: docs/limits.md. Found a problem it got wrong? Open an issue with the run id.

Source: README.md at commit f77458c

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

2
  1. v0.1.2LatestOct 20, 2026
  2. v0.1.1Oct 19, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

Other Wave integrations

Wave AI

co.waveOfficial publisher

AI note taker for phone calls, meetings and in-person talks. Search, quote, act on recordings.

Streamable HTTPWeb executableAuth
Credential requiredOct 25, 2026

EventWave

Published by GitHub account jgaethle10

EventWave is a remote MCP endpoint for event discovery and promotion workflows, with publishing and commerce kept under human control.

Streamable HTTPWeb executable
Productivity & WorkflowNot indexed yetOct 24, 2026

VirWave Breathe

Published by GitHub account virwave

Lets an assistant start calm, animated breathing sessions paced to the user's breath, shown in chat or opened in a browser.

Streamable HTTPWeb executable
Location & Lifestyle3 toolsOct 24, 2026

SpiritWave Labs Divination Engine

Published by GitHub account vizkr

A remote divination server that lets an assistant draw and interpret Hafez, Tarot, I Ching, Runes and Geomancy readings, alone or as a combined five-oracle view

Streamable HTTPWeb executable
6 toolsOct 24, 2026

WaveGuard

Published by GitHub account gpartin

Physics-simulation anomaly detection API that lets an assistant scan structured data, time series and crypto metrics for outliers and manipulation.

Streamable HTTPWeb executable
Finance19 toolsOct 27, 2026