Blave Agent

Turn trading ideas into tested strategies and live deterministic execution with Claude Code or Codex.

Stars
★ 358
Last updated
1d ago
License
Apache-2.0
Primary language
Python

At a glance

How it runs
Desktop appHosted service
Works with
Universal · cross-platformCodex · Claude Code
Cost
Free tier plus a paid hosted plan
Setup effort
Medium · a few setup steps
You'll need
macOS 13 or laterNode.js 22.12 or later with npmpython3 on PATHClaude Code or Codex installed and signed in, or a Blave accountShell / CLINetwork accessLocal filesystem
Typical use
A macOS-based solo quant who wants Claude Code or Codex to turn a natural-language single-asset idea into code and a backtest.
Not a fit if
  • Teams needing the desktop app on Linux or Windows
  • Traders requiring desktop orders to continue after the app quits
  • Users needing desktop auto-trading for portfolio strategies
Website
blave.org ↗ set on GitHub
Maintenance
Commits in the last 30 dayslast commit 2026-11-20latest release desktop-v0.1.57 (2026-11-19)+1.7 ★/day over 7 days
Source review
78/100 · Good

What does this agent do, and when should you use it?

Blave Agent is a macOS quant workspace whose Electron desktop shell lives in shell/. It connects to a user's Claude Code or Codex installation, or to usage-billed Blave AI, and lets the agent research an idea, write strategy code, run backtests, and prepare it for deployment. The shared lib/ layer covers data access, backtesting, Monte Carlo permutation testing, parameter scans, rolling walk-forward analysis, reports, charts, and exchange account and order helpers. Live execution keeps the LLM out of the order loop: scheduled deterministic code produces targets and manager/reconciler.py moves the account toward them, while state/HALT blocks new exposure. Users can run it locally through the macOS app or pay for a dedicated cloud machine accessed through the web workspace, Telegram, or SSH. The local software is free, but users may still need their own AI subscription, data access, and trading accounts, and its statistical checks do not guarantee future returns.

After a user describes an idea, the workspace classifies it as Type A, B, or C. Type A covers one fixed symbol and interval with a long, short, or flat position; Type C covers scheduled rebalancing of a multi-symbol weight vector; both require backtests. Screeners, grids, arbitrage, alerts, and one-off execution fall under Type B and do not require a backtest. The agent writes strategy files under strategies/ and runs the backtest chain through components such as lib/runner.py; Type A runs an MCPT through lib/validation.py by default, lib/param_scan.py searches for robust parameter plateaus, lib/walk_forward.py measures out-of-sample behavior, and lib/quality_check.py flags a zero fee. Versioned Type A and C strategies are marked “Live · file changed” when live code no longer matches the tested version, although the flag does not stop execution. For live runs, scheduled deterministic strategy code feeds target positions to manager/reconciler.py; state/HALT prevents new exposure while allowing closes and stops. Every order attempt is written to state/audit.jsonl, and the system queries the exchange before reporting that an order was placed. The desktop supports tested order libraries for Binance, OKX, BingX, Gate.io, and Bybit, plus keyless paper trading through lib/account_paper.py and lib/order_paper.py.

  1. A macOS-based solo quant who wants Claude Code or Codex to turn a natural-language single-asset idea into code and a backtest.
  2. A research team evaluating whether apparent performance survives MCPT, parameter-plateau analysis, and rolling out-of-sample tests.
  3. A crypto trader who wants to paper trade before connecting a Binance, BingX, OKX, Gate.io, or Bybit account.
  4. A Taiwan-equity researcher using free daily bars sourced directly from TWSE and TPEx in the desktop app.
  5. A user who needs strategies to continue while their computer is off and is willing to use Blave's paid dedicated cloud machine.
  6. A developer testing scheduled multi-asset rebalancing, while accepting that Type C portfolios cannot yet be auto-traded from the desktop app.

How do you install or deploy this agent?

Running from source requires macOS 13 or later, Node.js 22.12 or later with npm, and python3 on PATH. Install and sign in to Claude Code or Codex, or have a Blave account available. Run:

git clone https://github.com/Blave-TW/blave-agent.git
cd blave-agent/shell
npm install
npm start

On first launch, select your own Claude Code/Codex installation or Blave AI. Source installs use the system python3 to create a virtual environment; the packaged app bundles Python 3.12. A universal Apple Silicon and Intel macOS build is also available from GitHub Releases. To force the English interface, run:

BLAVE_LANG=en npm start

How do you use this agent?

Choose the agent provider during first launch. Using your own Claude Code or Codex requires no Blave account; the desktop app launches the CLI and leaves its credentials under that product's control. Blave AI instead requires a Blave login and is billed by usage. Enter a concrete strategy request such as:

Backtest BTCUSDT on the 4h chart: long when the 20-period SMA crosses above the 60-period SMA, flat when it crosses back below. Use a 0.05% fee per side.

For a portfolio workflow, try:

Build a portfolio of BTC, ETH and SOL with equal weights, rebalanced weekly, and backtest it.

For live trading, connect a supported venue through Auto trading › Connect an exchange in the desktop app; do not paste a key into chat. Grant read and trade permissions only, never withdrawal permission. Funding and resuming trading must be performed by the user through the Auto trading page. The built-in paper venue requires no keys. Desktop orders are sent only while Blave is running, and reopening the application leaves trading paused until the user presses Start trading.

What are this agent's strengths and limitations?

Pros
  • Type A backtests run an MCPT and record a p-value by default, with parameter-plateau scanning and rolling walk-forward analysis available alongside it.
  • Live schedules execute deterministic code without an LLM in the order loop, and state/HALT blocks new exposure at the order-library level.
  • The workspace detects when versioned live strategy code differs from the version that was backtested.
  • It includes tested order libraries for five crypto exchanges, a keyless paper venue, and templates for adding other brokers.
  • The local path supports either Claude Code or Codex, while Apache-2.0 permits commercial use, modification, and redistribution.
Limitations
  • The desktop application requires macOS 13 or later; the Capital Futures integration is separately limited to a Windows workspace.
  • Desktop orders stop when the application quits, and trading must be manually restarted after reopening it.
  • Type C portfolio strategies cannot currently be auto-traded from the desktop app.
  • Free local data is limited; Blave indicators, Taiwan intraday and flow data, futures, and the macro calendar require a Blave account with data access.
  • Always-on cloud machines, Blave market data, and the Blave LLM proxy are paid services outside this repository.

How does this agent compare with similar options?

With a user's own Claude Code or Codex, the desktop app only launches the CLI, requires no Blave account, and adds no Blave AI charge; the user remains responsible for their existing AI subscription or service costs. Blave AI is the built-in alternative and requires a Blave account with usage-based billing. Local desktop operation is free and keeps workspace credentials on the user's machine, but orders run only while the app is open. The paid cloud-machine option runs continuously and can be reached through the web workspace, Telegram, or SSH.

Key facts side by side with the most closely related agents.

Agent Source review Form / cost Stars Updated Language Full support on
Blave Agent This agent 78 · Good Desktop appFreemium ★ 358 1d ago Python Codex · Claude Code
Lumibot Trading Framework 49 · Major gaps Library / SDKFree + model costs ★ 2.2k today Python Codex · Claude Code
Vibe-Trading 68 · Some gaps CLIFree + model costs ★ 39k 1d ago Python OpenAI API · Claude API
Alpaca Multi-Agent Trading Workbench 47 · Major gaps CLIFree + model costs ★ 286 25d ago Python OpenAI API

How does FollowAgents rate this agent?

FollowAgents source review · FARS-2.1
Good
78/ 100 5-point scale 3.9 / 5
This review covers the source as of Nov 07, 2026. The project has shipped 7 stable release(s) since, which have not been re-reviewed. The score and verdict describe the reviewed version only. Recent releases ↓
Trust 21/29
Reliability 11/14
Adaptability 16/18
Convention 14/18
Effectiveness 10/13
Verifiability 6/8
Why each dimension lost points
Trust21 / 29 · 3.6/5

The browser uses an isolated session, denies permissions and downloads, and applies page-request restrictions for private addresses, some submissions, and file uploads. AGENTS.md reserves funding allocation and resuming trades to the user, requires exchange-side order checks, and names an audit log. These explicit boundaries support high scores for confirmation and external effects. Scores are reduced because the sample does not show the full credential-storage, network-policy, or trading-log implementations, or a recovery procedure for strategy and account state.

Reliability11 / 14 · 3.9/5

AGENTS.md says failed calls must be treated as failed steps and reported as such, and requires verification before reporting orders or backtest results. The Windows release workflow describes gates, version checks, artifact verification, and signing boundaries. Scores are reduced because the snippets do not establish consistency across the complete product, and the packet does not show dependency availability planning or comprehensive failure-handling documentation.

Adaptability16 / 18 · 4.4/5

The product targets traders using Claude Code or Codex to design, backtest, and deploy strategies, and describes desktop and cloud use, asset classes, platform branches, and several workflows. AGENTS.md gives specific triggers and boundaries for data, deployment, external agents, and desktop operation. Scores are reduced because the README's quick start centers on one BTC example, and some rules rely on reference files absent from this packet.

Convention14 / 18 · 3.9/5

The Spanish README provides a quick start, platform requirements, developer entry points, release information, and security notes; the Apache-2.0 license file is present. It discloses the unsigned Windows installer, points to Releases for updates, and the workflow explains Windows release and signing arrangements. Scores are reduced because this translation identifies an earlier English source commit, maintainer material is deferred to the English README, and the sample does not show a full changelog or clearly assigned maintenance owner.

Effectiveness10 / 13 · 3.8/5

The product brings strategy management, backtesting, parameter sweeps, versioning, and trading into one desktop workspace. The README includes a usable sample prompt and risk reminders, showing clear value for its intended audience. Scores are reduced because the packet does not show representative output formats or results, and does not quantify the costs behind the claimed benefits.

Verifiability6 / 8 · 3.8/5

Some README claims link to documentation or Releases, AGENTS.md requires reports to reflect actual tool output, and the Windows workflow checks versions and artifact hashes. The README also distinguishes historical backtest results from future trading risk. Scores are reduced because this is a limited sample and many claims point to references not included here, so corroboration across the full product cannot be assessed.

Risks and how to mitigate them
  • The README says the Windows installer is currently unsigned and may be blocked by SmartScreen; obtain installers only through the project's stated release channels.
  • The documentation warns that backtests describe past performance and live trading can lose all funds. This static review did not run the software or verify trading results.
Evidence confidence: Low Reviewed Nov 07, 2026 Reviewed revision 7d09adecd060 New commits since this review; the score may not cover them
See the full review method →
Maintain this project? Add the FARS badge to your README

The badge shows the current review score and verdict and links back to this page. It refreshes about once an hour after a re-review.

FARS

[![FARS review](https://followagents.com/badge/blave-agent.svg)](https://followagents.com/en/agents/blave-agent)
<a href="https://followagents.com/en/agents/blave-agent"><img src="https://followagents.com/badge/blave-agent.svg" alt="FARS review"></a>

Recent releases

Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.

  1. desktop-v0.1.57 Blave 0.1.57
    Release notes excerpt
    * 下單程式卡住時會自動重啟;一小時內卡住三次就暫停下單,並通知你 * 送單時卡住會先暫停,告訴你哪幾筆可能已成交 * 電腦睡醒後照最新訊號下單;減倉、平倉立刻做,新倉與加碼最多等 3 分鐘 * Telegram 通知不再拖慢下單 * 統一期貨:改帳號後一定用新帳號登入 * 統一期貨:沒有部位的帳號也能完成「確認登入」 * 統一期貨:密碼錯誤時也能直接〔改帳號〕,〔用新密碼確認〕不再失敗 * 安裝套件不再受你電腦上的 Python 設定影響 * 加強保護你的交易所金鑰 * 左下第一格改成選單:報告、設定、說明文件、登出都在這裡;有新報告會亮一個點 * 新增策略框多了三個範例,按一下就幫你填好 * 自動下單頁還沒有策略時,先帶你新增策略;〔連接帳戶〕改成旁邊的連結 * 優化的回看天數預設 365 天 * 選單項目和幾顆按鈕放大了可以點的範圍 下載:Mac 用 .dmg(通用版),Windows 用 Blave-Setup-0.1.57.exe。 已安裝的版本會自動提示更新。 --- * If the order program hangs, it restarts on its own. Three hangs in an hour pause trading and notify you …
    View on GitHub ↗
  2. desktop-v0.1.56 Blave 0.1.56
    Release notes excerpt
    * 出錯時會先問你要不要送錯誤內容給我們,按「看會送出的內容」可以先看;也可以在設定 › 隱私改成自動送或不送 * 策略組合的「優化」拿掉〔跑優化〕:選好策略和方法就自動算權重,部位大小直接在頁面上填 * 權重被另一次計算換掉時,給一個〔重算權重〕連結,不會自己一直重算 * 策略庫不再掛「已驗證」標籤 * 手機配對框改寫:先說用手機可以做什麼,步驟從四步減成兩步 * 事件列「K 棒沒到位」寫策略名稱,等待時間只算開盤時間,週末和收盤不算 * 「解除綁定」統一改叫「解除連接」 * agent 回報回測時,費用寫「交易成本」 * 幾顆小按鈕放大了可以點的範圍 下載:Mac 用 .dmg(通用版),Windows 用 Blave-Setup-0.1.56.exe。 已安裝的版本會自動提示更新。 --- * When something goes wrong, Blave now asks before sending the error details; See What Will Be Sent shows them first. You can switch to always send or never send in Settings › Privacy …
    View on GitHub ↗
  3. desktop-v0.1.55 Blave 0.1.55
    Release notes excerpt
    * 台指期休市時按「暫停並關閉部位」或解除連接,平倉照樣送出;券商不收就寫清楚「沒有平倉:休市中」,部位留著讓你開盤後自己平,不會在開盤時自動幫你平 * 休市時解除連接被拒,下單不會因此暫停,金鑰也留著,開盤後再平一次就好 * 換交易所時先暫停下單、再換金鑰,換的那幾秒不會用新帳戶下單 * 交易所合約停牌或下架時,不會再被當成「已平倉」,部位沒平完就不解除連接 * 送出的單還沒成交時,不再跳出「買進 0 口」的通知 * OKX 的策略改用 OKX 自己的 K 線回測與下單;跟 Binance 同名但不同種類的合約會先擋下 * Gate.io 合約保證金不足時,改成中文說明 * Windows 第一次開啟不會停在黑畫面;第一次下載引擎斷線時會接著下載 * 回測與參數掃描的「手續費」改叫「交易成本」,裡面含滑價等成本 * 策略頁的拉回改成圖示,拉回後直接跳到回測並顯示進度 * 已暫停時頁首不再出現「關閉部位」,台指期開盤提醒拿掉 下載:Mac 用 .dmg(通用版),Windows 用 Blave-Setup-0.1.55.exe。 已安裝的版本會自動提示更新。 --- …
    View on GitHub ↗

FAQ

Is local use free?
The repository and local application are free. Blave does not charge for AI when users connect their own Claude Code or Codex, although those products and other external services may have their own costs. Blave AI, cloud machines, and some market data are paid offerings.
Where are exchange keys stored, and which permissions are required?
The desktop stores them in ~/Blave/workspace/.env on the Mac. A cloud workspace stores them on the user's dedicated machine, while keys bound through the web page are encrypted by Blave. Keys should have read and trade permissions only, never withdrawal; Gate.io does not expose the withdrawal flag, so users must verify it themselves.
Can the agent fund an account or resume trading?
No. The user must handle funding and resume trading through the desktop Auto trading page or the cloud web workspace. The agent may trip the kill switch itself, but it will not resume trading for the user.
Do the validation tools guarantee profitable live results?
No. MCPT, parameter scans, and walk-forward analysis reduce the chance of mistaking luck or overfitting for a robust strategy, but they cannot predict or guarantee future returns. Trading can lose all committed capital.
Can it connect to an exchange or broker not listed?
The agent can create helpers from lib/account_TEMPLATE.py and lib/order_TEMPLATE.py, but Blave has not tested those custom integrations, so the adopter must validate them.
Is Blave Agent open source?
Yes. The source code is on GitHub (Blave-TW/blave-agent) under the Apache-2.0 licence.
Is Blave Agent still maintained?
Yes. The latest commit on GitHub is from 2026-11-20, within the last 30 days. The latest release is desktop-v0.1.57 (2026-11-19).
Who makes Blave Agent?
Blave Agent is published on GitHub by the account Blave-TW. FollowAgents has not verified which person or company is behind that account.
What is Blave Agent's official website?
The project's GitHub repository lists https://blave.org as its website. That link is set by the repository owner; the source code itself is on GitHub (Blave-TW/blave-agent).
Was this page helpful?
View on GitHub ↗ Install ↓

Compare agents like this one

The same FARS review applied across the shortlist this agent qualifies for.

Related concepts