Skip to content

Open source

Open source is how we build. It’s in our DNA.

Dailybot was built on decades of free-and-open software — every library, every server, every CLI we depend on. So we give it back. Everything we design to run outside our product is public on GitHub, MIT-licensed, and shaped by the developers who use it.

How we build in the open

  • MIT top to bottom

    No dual-licence games. Fork it, strip it, rebrand it.

  • Public roadmap

    Every issue, every PR, every changelog is public.

  • No vendor lock-in

    Every tool keeps working if we disappear tomorrow.

  • Shaped in public

    The people who use it help decide where it goes.

Our open-source commitment, in numbers

8

public repositories on DailybotHQ

MIT

licensed, top to bottom

Public

issues, PRs, and roadmap

Zero

vendor lock-in, by design

Where we bet the biggest.

Three flagship repositories — the tools we designed to run outside our product. If Dailybot disappears tomorrow, these three keep working.

Dailybot Agent Skill

MIT

Fourteen capabilities that turn your AI coding agent into a real Dailybot teammate.

Fourteen coordinated capabilities — report, checkin, chat, conversation, kudos, forms, email, ask, teams, health, messages, workflow, channels, env — that map 1:1 to the Dailybot public API. Your agent can send messages, create forms, give kudos, open group DMs, and more. Reads `dailybot login` or `$DAILYBOT_API_KEY`, chooses whichever is available.

  • 14 capabilities mapped to the Dailybot API
  • Chat, forms, check-ins, kudos, conversations, and more
  • Works with every major AI coding agent
  • MIT-licensed
Language: Bash / Markdown

Install

npx skills add DailybotHQ/agent-skill

Deep Work Plan

MIT

The gated-plan methodology any AI coding agent can execute.

DWP turns an ambiguous goal into a numbered, atomic, validated plan under `.dwp/plans/`. Ships with eight sub-skills (create, execute, refine, resume, status, verify, onboard, author) and an optional Dailybot addon that wires the plan lifecycle into your team’s standup.

  • Eight composable sub-skills
  • Works with Claude Code, Cursor, Codex, Gemini, Copilot
  • Optional Dailybot progress-report addon
  • MIT-licensed
Language: Bash / Markdown

Install

npx skills add DailybotHQ/deepworkplan-skill

Dailybot CLI

MIT

The command-line client for the whole Dailybot API.

A single binary — `dailybot` (current release 3.7.3) — with 15+ commands that give any script, CI job, or AI coding agent full access to the Dailybot platform. Report progress, send messages, create forms, give kudos, open group conversations — all from the command line. Dual auth: OTP login or API key.

  • 15+ shipping commands
  • 6 install channels (curl, brew, npm, apt, winget, binary)
  • Non-blocking, JSON-first output
  • MIT-licensed
Language: Python

Install

curl -fsSL https://cli.dailybot.com/install.sh | bash

The wider ecosystem — five more OSS libraries.

Prompt contracts, PR reviews, emoji parsing, text highlighting, and the Deep Work Plan website. Small, focused, MIT-licensed — the same standard as the flagship.

Deep Work Plan website

MIT

The open-source documentation hub for the Deep Work Plan methodology.

The static Astro site behind deepworkplan.com — landing pages, methodology docs, and adoption guides. Read it, fork it, self-host it. Every page in the methodology docs is intentionally shippable as its own PR.

  • Full methodology reference
  • Deploys to Cloudflare Pages
  • Localised structure ready for translation
Language: Astro

promptschema

MIT

The Zod for AI prompts — schema-validated, versioned, multi-provider, cross-language.

Stop hardcoding LLM prompts as raw strings. Define them as contracts: typed input, semantic versioning, readable diffs, and a shared registry that TypeScript and Python both consume. Any model — OpenAI, Anthropic, Gemini, Ollama, or your own adapter.

  • Zod (TS) and Pydantic (Python) schemas
  • Semantic versioning with automatic diff
  • Shared JSON-Schema registry across languages
  • ~12 kB bundle, zero vendor lock-in
Language: TypeScript + Python

Install

npm i @dailybot/promptschema

AI Diff Reviewer

MIT

One review methodology, two surfaces — a GitHub Action for CI and a coding-agent skill. On the Marketplace.

One product, two surfaces: a GitHub Marketplace Action for PR CI and a coding-agent skill for local “review my branch” passes. Same severity model (critical / warning / info), same prompts when you pin the same tag. Inline comments, merge gating, no extra infra — just your provider’s API key.

  • Dual surface: GitHub Action + coding-agent skill
  • Severity gating (critical / warning / info)
  • Providers: Anthropic, Claude Code, Cursor, Codex
  • Listed on the GitHub Marketplace
Language: Python + Markdown

Install

uses: DailybotHQ/ai-diff-reviewer@v2

universal-emoji-parser

MIT

Parse unicode and emoji shortcodes into consistent HTML images — battle-tested since Dailybot 1.0.

A tiny library that turns any mix of unicode emoji and shortcodes (`:rocket:`) into HTML `<img>` tags backed by Twemoji, keeping emoji rendering identical across Slack, Teams, Discord, Google Chat, and the browser.

  • 1,898 curated emoji entries
  • Unicode ↔ shortcode both ways
  • Custom CDN and HTML tag options
  • Twemoji-backed image rendering
Language: TypeScript

Install

npm i universal-emoji-parser

search-text-highlight

MIT

Find and highlight substrings in text with an HTML wrapper — full unicode support.

A focused utility to highlight query matches inside a string, with configurable HTML tag, class name, case sensitivity, and match-all / match-first behaviour. Small, dependency-light, unicode-aware — production-tested across Dailybot search UIs.

  • Highlight all matches or just the first
  • Custom HTML tag and class name
  • Case-sensitive or case-insensitive
  • Unicode / emoji safe
Language: TypeScript

Install

npm i search-text-highlight

Why we build in the open.

Two reasons. First — Dailybot itself is built on decades of free-and-open software; not giving back would be extractive. Second — trust. A tool that runs on a proprietary rail is a rail we can turn off. Everything we design to run outside our product should keep working if we disappear tomorrow. That is the promise, written in a licence.

  • MIT

    You can fork it. You can strip our branding. You can rebrand and ship it.

  • Public roadmap

    Every issue, PR, discussion, and CHANGELOG lives on GitHub. Full transparency, from roadmap to release.

  • You own the data

    Nothing sensitive stays local to a proprietary runtime. Every project is portable by design.

How to contribute

Skim the repo, open an issue, ship a PR. Every repo has good-first-issue labels and a CONTRIBUTING.md.

Step 1

Pick a repo

Skim the README, open an issue if the fit isn’t obvious. Every repo has a good-first-issue label.

Step 2

Follow the standards

Every repo has a CONTRIBUTING.md and passes a validation gate before PRs merge.

Step 3

Ship it

Small PRs review faster. Ask for early feedback in the issue thread.

The road so far

A moving target — an updated public log lives in each repo’s CHANGELOG.

Past

CLI × API-key parity

Both credentials — `dailybot login` and `DAILYBOT_API_KEY` — grant the same access. Parity is a written guarantee, not a footnote.

Now

8-sub-skill Agent Skill pack

Report, checkin, chat, kudos, forms, teams, health, email. Deep Work Plan wired into every major AI CLI.

Next

More surface, more harnesses

More sub-skills, more supported agent frameworks, tighter feedback loops with the community.

Star the org. Never miss a release.

One click. Every ship shows up in your GitHub feed.