Claude Code and Codex agent target
The project is explicitly positioned as a harness for Claude Code and Codex agents, giving those agent environments a trading-focused repository and workflow context.
OpenTrade
Freemium
OpenTrade is an open-source trading harness for Claude Code and Codex agents. It provides a repository-based app, agent templates, and scheduling-oriented run management for automated trading workflows.
OpenTradeとは？
OpenTrade is an open-source trading harness for Claude Code and Codex agents. Hosted in a public GitHub repository, it organizes trading-oriented agent work through an application, agent templates, and supporting hooks and resources. The available project material points to scheduled runs, wake outcomes, monitoring history, runtime limits, and operational logging, while leaving broker support, instrument coverage, setup requirements, and live-trading behavior to the repository documentation and code.
OpenTradeでできること
Agent templates and hooks
The repository includes `templates/agents` and `resources/hooks`, providing project-managed locations for agent instructions and supporting run behavior.
Scheduled wake and run tracking
Project history references scheduled tasks, wake outcomes, and Monitor-tab history rows, supporting workflows that need to record and revisit agent runs over time.
Runtime and stop controls
The application includes a maximum run-time path and a shared handling path for user stops and timeout kills; timed-out wakes are represented as failed with a specific timeout reason.
Headless-run sleep guard
For unattended runs, the project can hold an idle-sleep prevention assertion and release it when the run ends. Failures are reported, and lid-close and low-battery sleep are not blocked.
Local operational records
The project history describes local logging for repeated host or broker failures, with deduplication and recovery reporting rather than writing an identical persistent error on every poll.
利用シーン
“Scheduled agent workflows”
Use the harness when an agent task needs to run at a future time and its wake result should remain visible in later monitoring history.
“Unattended or headless runs”
Run an agent workflow without keeping the application in active use, while applying a maximum runtime and reporting timeout or other run failures.
“Iterating on trading agents”
Use the repository’s agent-template and hook structure to organize repeatable trading-agent instructions and supporting execution behavior.
“Investigating run failures”
Review recorded wake status, timeout reasons, and local operational logs when a scheduled task or broker-related poll does not complete normally.
よくある質問
Which agent environments does OpenTrade target?
The project description explicitly names Claude Code and Codex agents. The available source does not establish compatibility with other agent runtimes.
Is OpenTrade open source?
Yes. OpenTrade is presented as an open-source project in a public GitHub repository and includes a repository license file.
Does the source confirm which brokers or instruments are supported?
No complete broker or instrument list is available in the supplied product evidence. The repository should be checked for its current supported-instrument and integration documentation.
Can runs be stopped or timed out?
The project history describes a maximum run time and a shared path for user stops and timeout kills. A wake that reaches the run limit is recorded as failed with a timed-out reason.
Does the sleep guard prevent every kind of system sleep?
No. The documented behavior covers idle sleep during headless runs; lid-close and low-battery sleep can still occur.
クイック情報
- Category
- Developer Tool
- Source
- Public GitHub repository
- Primary users
- Users building trading workflows with Claude Code or Codex agents
- Project type
- Open-source trading harness
- Repository components
- Application, agent templates, resources, and hooks
- Run model
- Scheduled and headless-oriented agent runs with monitoring history
OpenTrade のトラフィック分析
トラフィックデータは参考情報としてご利用ください。
- ドメイン評価
- 35
OpenTradeの代替品
Flocker
flocker.md
Flocker is an AI agent management platform for creating agent profiles, coordinating multi-agent work, and monitoring activity from a shared dashboard. It is designed for people running multiple agents across tools such as Claude Code, Codex, OpenClaw, OpenCode, and MCP-compatible workflows.
Edgee
www.edgee.ai
Edgee is an agent gateway for coding teams that reduces token usage, routes requests across models, and provides usage visibility. Its Compression V2 works as a drop-in CLI layer for Claude Code, Codex, OpenCode, Cursor, and other supported agents.
Steel
steel.dev
Steel is open-source browser infrastructure for AI agents and web automation. It provides cloud browser sessions through an API, with support for common browser automation tools and workflows such as scraping, research, RPA, and agent-driven web tasks.
DSPy
dspy.ai
DSPy is an open-source framework for programming language-model applications instead of manually prompting them. Its public repository provides the project code, documentation, tests, issue tracking, and contribution workflow.
Graphiti
www.getzep.com
Graphiti is an open-source GitHub project for building real-time knowledge graphs for AI agents. It provides a codebase that developers can inspect, extend, and contribute to through the repository’s development workflow.
Flyte
flyte.org
An open-source runtime for durable AI workflows, model training, serving, and agents.