What is TecAdRiseBot?
TecAdRiseBot is an Electron, React and TypeScript desktop app for Windows 10 and 11 that hosts several AI agents at once. Each agent gets its own model, a sticky instruction file it calls a Soul, a workspace folder, MCP servers and scheduled routines (loop, interval, daily, weekly, once or cron). Agents can message each other, write to a shared wiki-style memory, and read and change their own configuration. It ships with Playwright and AgentMail MCP servers set up on first launch.
Agents run on the Cursor TypeScript SDK, which keeps the agent loop on your PC but sends inference to Cursor's hosted models and bills your Cursor plan, or on any OpenAI-compatible key you bring (Hugging Face, OpenRouter, OpenAI, Claude, NVIDIA NIM). A local HTTP API on 127.0.0.1:8787 lets other programs list agents, post messages and patch their settings.
The maker, TecAdRise of Wysoka, Poland, sells AI automation services and pitches the app as a free alternative to Grok Bot. The repository was created on 19 September 2026 under MIT; v0.1.5 shipped on 6 November 2026 as an installer hosted on tecadrise.ai rather than GitHub. It is not affiliated with xAI or Cursor.
Is TecAdRiseBot worth it?
TecAdRiseBot's pitch is Grok Bot without the subscription: a desk of agents on your own Windows PC, each with its own model, instructions, tools and timetable. Grok Bot comes with Cursor's Premium plan at $30 a month; this app is free, and you pay only for the model calls.
The feature list is longer than most hobby projects. Routines fire on intervals, daily, weekly or on a cron expression; step workflows have one agent do the work and a second review it; a shared memory folder in the Obsidian style lets agents keep facts across sessions. Agents can also patch their own configuration, which the README calls self-healing and a security reviewer would call self-modifying.
The Cursor SDK is the default engine. Its docs say the agent loop runs locally but inference always uses Cursor's hosted models, billed like IDE runs to the plan behind your API key; Plus at $8 a month covers only Cursor and Grok models, and anything else is charged at API list price. Local SDK agents also run tool calls without asking for approval unless you add hooks or a sandbox. The bring-your-own-key mode added in v0.1.5 avoids Cursor entirely.
The security posture is mixed. The renderer uses context isolation without Node integration, and keys sit in Electron's safeStorage. But the control API on port 8787 has no authentication, so any program on the PC can message agents or rewrite their instructions, and the lockfile pins Electron 36.9.5, a line that left support on 27 October 2025 while the current release is 45.3.0.
The project has 0 stars, 0 forks and a single committer, with the installer served from the maker's own server. Against OpenClaw or Hermes Agent, which have large communities and run headless on servers, it offers a friendlier Windows window and much less scrutiny.
Where it earns its place
A freelance marketer on a Windows laptop
Have one agent check competitors' pages every morning and another draft a summary
Set a daily routine on a research agent with Playwright MCP, let it write findings to shared memory, and have a writer agent read them. On a BYOK OpenRouter key the cost is whatever the chosen model charges; routines skip if the laptop is shut.
A developer already paying for Cursor
Run background agents on a schedule without another subscription
Paste a Cursor user API key and the SDK bills runs to that plan like IDE usage; add hooks or a sandbox first, because local SDK agents run tool calls without asking.
An IT lead at a 20-person office
Judge whether staff may install it
The unauthenticated port 8787, self-modifying agents and an Electron line out of support since October 2025 make it unsuitable for managed machines until those change.
Before you commit
- Setup
- Download the 64-bit installer from tecadrise.ai (about 105 MB, SHA-256 published), paste a Cursor API key or a BYOK provider key in Settings, pick a model and create an agent. From source: Node.js 22.13+ and npm 10+, then npm install and npm run dev.
- Integrations
- Cursor TypeScript SDK or any OpenAI-compatible endpoint; MCP servers globally or per agent, with Playwright and AgentMail bundled; OAuth connectors for GitHub, Canva and ElevenLabs; a local REST API on 127.0.0.1:8787.
- Data and privacy
- Chats, memory and workspaces stay in the app's data folder; prompts and context go to Cursor or your chosen provider. Keys are kept in Electron safeStorage. The control API binds to localhost only but has no authentication. No telemetry is documented. MIT.
- Cost at scale
- The app is free; cost is model usage, either your Cursor plan (Plus $8, Premium $30 a month, other models at API list price) or your provider's per-token price. The in-app token table is a local log, not an invoice.
- Limits
- Windows 10 and 11 only; routines fire only while the app runs; single maintainer, no outside issues or contributors yet; Electron 36.9.5 is outside Electron's three supported major versions.
What’s good
- Free MIT app with per-agent model, instructions, MCP and workspace
- Cron, interval and weekly routines with pause and run-now
- Bring-your-own-key mode for OpenRouter, OpenAI, Claude and others
- Shared wiki memory and agent-to-agent messages
- Published SHA-256 for every installer
What’s not
- Unauthenticated local API on port 8787 can rewrite agents
- Pinned to Electron 36, out of support since October 2025
- One committer, 0 stars, seven weeks old
- Windows only; routines run only while the app is open
- Cursor SDK mode bills your Cursor plan and runs tools without approval by default
Best for
Windows power users comfortable reading TypeScript who want a free local host for several scheduled agents and accept an early single-maintainer project.
TecAdRiseBot alternatives
TecAdRiseBot: common questions
- Is TecAdRiseBot worth it?
- A capable one-person agent desktop with real scheduling and memory, but it is seven weeks old, has no outside users yet and runs on an Electron release a year out of support. We rate it 2.9 out of 5.
- How much does TecAdRiseBot cost?
- Free and open source (MIT). Model costs are yours: Cursor SDK runs bill your Cursor plan (Plus $8/month covers Cursor and Grok models; other models at API list price), or your own provider key is billed by that provider.
- What are the downsides of TecAdRiseBot?
- Unauthenticated local API on port 8787 can rewrite agents; pinned to Electron 36, out of support since October 2025; one committer, 0 stars, seven weeks old; windows only; routines run only while the app is open; cursor SDK mode bills your Cursor plan and runs tools without approval by default.
- Who is TecAdRiseBot best for?
- Windows power users comfortable reading TypeScript who want a free local host for several scheduled agents and accept an early single-maintainer project.
- How much does TecAdRiseBot cost for a team or in production?
- The app is free; cost is model usage, either your Cursor plan (Plus $8, Premium $30 a month, other models at API list price) or your provider's per-token price. The in-app token table is a local log, not an invoice.
- Is TecAdRiseBot safe for company data?
- Chats, memory and workspaces stay in the app's data folder; prompts and context go to Cursor or your chosen provider. Keys are kept in Electron safeStorage. The control API binds to localhost only but has no authentication. No telemetry is documented. MIT.
- What are the best TecAdRiseBot alternatives?
- The closest alternatives we have reviewed are OpenClaw and Hermes Agent.
Answers checked Nov 11, 2026.