Practical overview
When it fits
Use it when repository work can be organized as queued tasks and reviewed through Git, especially when a Slack-based intake or continuous background processing is useful.
Who it is for
- Developers managing repository tasks in the background
- Teams that use Slack as an operational interface for coding work
- Builders who want queued tasks separated into isolated workspaces
Documented uses
- Submit coding tasks through Slack or the CLI
- Run multiple tasks in separated workspaces
- Track queued and active work
- Save task results and metadata
- Create Git commits and optionally automate pull requests
Requirements & limits
The repository is public and licensed under MIT. Slack is documented as optional for CLI use; the primary service requires local installation, Claude Code CLI authentication and the listed development tools.
- Slack integration requires creating an app, enabling Socket Mode and configuring tokens and scopes.
- The Claude Code CLI and a completed Claude login are required for the documented workflow.
- Git is required for automatic commits, while the gh CLI is optional for pull-request automation.
- The isolated-workspace and Git automation path is verified by the repository description.
- Replacing Slack with another messaging interface requires modification because only Slack integration is documented.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up the CLI coding workflow
Install Sleepless Agent, authenticate Claude Code and start a task queue without Slack.
Before you start
- Python 3.11+
- Node.js for installing the Claude Code CLI
- Git for automatic commits
- A Claude Code login
Steps
Install the package with pip, or clone the repository and install it in an editable environment.
Install the Claude Code CLI with npm, run claude login, and verify the CLI version.
Start the daemon with sle daemon.
Queue a task with sle think and inspect it with sle check.
Expected outcome
A running CLI-based daemon with a queued coding task and status visibility.
Guide limitations
- This recipe documents task queuing and status inspection; it does not establish that a particular coding task will complete successfully.
- Pull-request automation additionally requires the optional gh CLI.
Connect Slack as the task interface
Configure the documented Slack app path for submitting tasks and checking agent activity.
Before you start
- A Slack workspace
- Slack app setup access
- A Sleepless Agent installation and authenticated Claude Code CLI
- Slack bot and app tokens
Steps
Create a Slack app from scratch and enable Socket Mode.
Create the documented slash commands and configure the listed bot scopes.
Install the app to the workspace and obtain the bot and app tokens.
Copy .env.example to .env, set SLACK_BOT_TOKEN and SLACK_APP_TOKEN, then run sle daemon.
Expected outcome
A Slack-connected Sleepless Agent daemon that can receive the documented commands.
Guide limitations
- Interactive chat mode additionally uses event subscriptions and the listed history and reaction scopes.
- The evidence documents configuration and startup, not a compatibility guarantee for a specific Slack workspace.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 4 October 2026
- Original website title
- context-machine-lab/sleepless-agent · Official homepage title · retained in its source language