Open-source workbench
Free
MIT; external agents/models/hosting billed separately
- Included scope
- Local terminals, task board, sessions and optional relay access
Windows coding-agent terminal workbench · Official-source review 2026-10-09
Termexo manages coding-agent terminals, saved sessions, model profiles and task boards on Windows. Its local MCP service lets a permitted agent find another running terminal, send instructions and read that terminal’s actual output.
Relevant to Windows developers coordinating existing coding CLIs and monitoring approvals across projects.
Windows 10/11 desktop; relay supports other operating systems. Existing CLIs must be installed/signed in and running. Terminal interaction supports Claude Code, Codex, OpenCode, Grok Build and Antigravity; task-board execution excludes Antigravity. Host must stay powered/on and connected for remote use. Relay login and desktop token are separate checks. Shared workspaces, granular team permissions and central audit remain planned. Local-first does not prevent configured model providers receiving code. Credential Manager storage and encrypted relay frames are vendor-described, not audited. Sending a task is not proof of completion. No installation, terminal control or relay deployment performed.
Enable only required MCP scopes, first list a non-sensitive workspace read-only, and inspect task output before approving code changes.
Windows 10/11 desktop; relay supports other operating systems. Existing CLIs must be installed/signed in and running. Terminal interaction supports Claude Code, Codex, OpenCode, Grok Build and Antigravity; task-board execution excludes Antigravity. Host must stay powered/on and connected for remote use. Relay login and desktop token are separate checks. Shared workspaces, granular team permissions and central audit remain planned. Local-first does not prevent configured model providers receiving code. Credential Manager storage and encrypted relay frames are vendor-described, not audited. Sending a task is not proof of completion. No installation, terminal control or relay deployment performed.
Official product evidence reviewed on 2026-10-09; product artwork inspected visually. No hands-on vendor application test was performed. Application performance, security claims and plan enforcement were not independently tested.
What it can do
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Product workflow | Termexo manages coding-agent terminals, saved sessions, model profiles and task boards on Windows. Its local MCP service lets a permitted agent find another running terminal, send instructions and read that terminal’s actual output. | Facts sourced | 2026-10-09 |
| Pricing and restrictions | Windows 10/11 desktop; relay supports other operating systems. Existing CLIs must be installed/signed in and running. Terminal interaction supports Claude Code, Codex, OpenCode, Grok Build and Antigravity; task-board execution excludes Antigravity. Host must stay powered/on and connected for remote use. Relay login and desktop token are separate checks. Shared workspaces, granular team permissions and central audit remain planned. Local-first does not prevent configured model providers receiving code. Credential Manager storage and encrypted relay frames are vendor-described, not audited. Sending a task is not proof of completion. No installation, terminal control or relay deployment performed. | Facts sourced | 2026-10-09 |
Understand the total cost
Open-source workbench
Free
MIT; external agents/models/hosting billed separately
scritty is a terminal emulator that captures attributable turns from supported coding CLIs into a searchable local corpus. Agents can query shared memory through MCP, while people can search through the terminal or CLI.
Explore toolOpenComputer deploys TypeScript agent functions onto managed Linux sessions. It handles model selection, tools, MCP, streaming, schedules and durable sessions that can hibernate and resume.
Explore toolTrigger.dev runs TypeScript tasks and durable AI agents with retries, queues, streaming and observability. Its chat-agent workflow supports paused tool approvals and continued execution, while deployments version task code and Realtime exposes run status.
Explore toolThe practical questions
No. This listing is based on dated official-source evidence and a visual artwork check. Unconfirmed costs and restrictions are identified explicitly.
Termexo manages coding-agent terminals, saved sessions, model profiles and task boards on Windows. Its local MCP service lets a permitted agent find another running terminal, send instructions and read that terminal’s actual output.
MIT open-source workbench with no Termexo account. Agent subscriptions, model requests and self-hosted relay costs are separate.. This record lists ongoing free access; check the plan limits before starting.
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Local MCP exposes terminal, task and permitted-setting tools. Remote access uses a self-hosted relay or trusted LAN/VPN.. API access and subscription access may have different terms; consult the linked sources.
Windows 10/11 desktop; relay supports other operating systems. Existing CLIs must be installed/signed in and running. Terminal interaction supports Claude Code, Codex, OpenCode, Grok Build and Antigravity; task-board execution excludes Antigravity. Host must stay powered/on and connected for remote use. Relay login and desktop token are separate checks. Shared workspaces, granular team permissions and central audit remain planned. Local-first does not prevent configured model providers receiving code. Credential Manager storage and encrypted relay frames are vendor-described, not audited. Sending a task is not proof of completion. No installation, terminal control or relay deployment performed.
No retained pricing changes yet. A current price alone does not establish a historical trend.
Current official product source read individually; product artwork inspected visually.
Read original source ↗Official product page reviewed for disclosed costs; unverified prices and allowances are identified in the listing.
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.