Practical overview
When it fits
Use Aiden when the work spans multiple computer surfaces or connected services and benefits from an agent that can plan, act, recover and check its output.
Who it is for
- Developers building autonomous computer-use workflows
- Operators automating multi-step work across connected tools
- Teams that need browser, terminal, file and API actions coordinated by an AI agent
Documented uses
- Plan and execute multi-step operational tasks
- Coordinate browser, terminal, file, application and API actions
- Automate workflows that may need failure recovery
- Retain context across work and verify the result
Requirements & limits
The repository is public and identifies Aiden as open source under the AGPL-3.0 licence. Documented environments include Windows, Linux, WSL and macOS API mode.
- The project is TypeScript-based and uses Node.js 20 or 22, npm, ESM and esbuild.
- Browser automation is documented through Playwright and headless Chromium.
- SQLite with FTS5 and better-sqlite3 provides the documented storage and search layer.
- Docker sandboxing and MCP 1.0 are part of the documented technical stack.
- Provider choice includes Groq, Anthropic Claude, OpenAI, Google Gemini and local Ollama.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Configure Aiden with a provider and model
Complete first-run setup and select a provider and model for the interactive terminal.
Before you start
- Aiden installed and available as the
aidencommand. - A supported provider with authentication and model access; availability is checked during setup.
Steps
Run
aidento start the first-run setup wizard.Review the disclaimer and let the wizard initialize system checks, skills, tools, memory, and runtime.
Choose a provider, then let Aiden fetch its available model catalog where supported and probe authentication, model access, and tool-use support.
After the success screen, use
/modelinside the REPL to open the provider/model picker and save a model choice toconfig.yamlif you want to change it.
Expected outcome
The setup wizard validates the provider connection and hands off to the interactive terminal; a model selected with `/model` is saved to `config.yaml`.
Guide limitations
- Live model fetching is available only where supported by the selected provider.
- Output quality depends heavily on the selected model's capability.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 4 October 2026
- Original website title
- taracodlabs/aiden · Official homepage title · retained in its source language

