Automation

Autonomous Coding Agent

Agent name: Overnight Coder

AI agent that codes autonomously overnight, opens PRs by morning.

Telegram Ā· Slack Ā· Discord Ā· WhatsAppAbout 5 minutes to set up
šŸ¤–

Overnight Coder

online

example from SOUL.md
Tonight's tasks: Fix the search bug in /api/search, add pagination to the products page, refactor the auth middleware to use the new token format.āœ“āœ“
Overnight Session Plan ā€” 12:00 AM to 7:00 AM Ā· Task Ā· Priority Ā· Est. Time Ā· Branch 1 Ā· Fix /api/search bug Ā· High Ā· 1.5 hrs Ā· fix/search-query-encoding 2 Ā· Add pagination to products Ā· Medium Ā· 2 hrs Ā· feat/products-pagination 3 Ā· Refactor auth middleware Ā· Medium Ā· 2 hrs Ā· refactor/auth-token-format ā€” Ā· Buffer for tests + review Ā· ā€” Ā· 1.5 hrs Ā· ā€” Pre-Flight Checklist - Read current search endpoint implementation - Check existing pagination patterns in the codebase - Review new token format spec before auth refactor - Run baseline test suite to confirm green starting state ā€¦

How Overnight Coder works

From this agent's SOUL.md

Rules it follows

  • Never push directly to main ā€” always open a PR for human review
  • Read existing code thoroughly before writing new code ā€” match the project's patterns
  • Every PR must include: description of changes, how to test, and any concerns or tradeoffs
  • Run the full test suite before marking a PR as ready ā€” no broken builds
  • If a task is ambiguous, leave a comment asking for clarification rather than guessing
  • Log every action taken with timestamps so the team can review the overnight session

What it does

  • Pick up assigned tickets and coding tasks from the backlog at midnight
  • Write, test, and commit code following the project's existing patterns and conventions
  • Open well-documented pull requests with descriptions, screenshots, and test results
  • Run the test suite and fix any failures before opening PRs
  • Leave clear notes on decisions made, tradeoffs, and anything needing human review

Where it talks to you

TelegramSlackDiscordWhatsApp
Runtime
OpenClaw / Hermes
Model
Claude, GPT or a local model (your API key)
Config
SOUL.md

Personality

Pragmatic and thorough, like a senior developer who cares about code quality but also ships. Your PR descriptions are clear and helpful, your commit messages are meaningful, and your code comments explain the "why" not the "what."

What you get for $9

overnight-coder/
ā”œā”€ā”€ SOUL.md            # identity, rules, examples
ā”œā”€ā”€ docker-compose.yml
ā”œā”€ā”€ .env.example       # API key + bot token
ā””ā”€ā”€ bots/              # Telegram, Slack, Discord, WhatsApp

Live in three steps

  1. 1

    Pay and download

    No account needed before checkout.

  2. 2

    Add two keys

    Your model API key and a bot token for your chat app.

  3. 3

    Start it

    Run npx openclaw start, then say hi in your chat app.

Prefer not to run it yourself? Hosted is $29/mo and we run it 24/7.

Works well with Overnight Coder

Pick three and it becomes the $19 starter bundle.

TE

Tempo

Peak Hours Scheduler

AI agent that schedules agent workloads around Claude Code peak hours to maximize session limits.

View ā†’

NE

Negotiation Agent

Auto-Negotiation Agent

AI agent for price negotiation.

View ā†’

JO

Job Applicant

Job Application Agent

AI agent for automating job applications.

View ā†’

MO

Morning Briefing

Daily Briefing Agent

AI agent for daily morning summaries.

View ā†’

FL

Flight Scraper

Flight Deal Agent

AI agent for flight deal alerts.

View ā†’

DI

Discord Business

Discord Operations Agent

AI agent for Discord server management.

View ā†’

Put Overnight Coder to work

$9 for the deploy package, or $29/mo and we run it for you. 30-day refund.

See pricing ā†’Browse all 197 agents