intent-hq/intent

Monorepo for the Intent platform — tracks intentd, cloudlands-fe, and ios as submodules; docs, CI, and release orchestration

Intent – Overview

Intent is a desktop‑first productivity platform designed for developers who work with AI‑powered coding agents. It bundles a Rust‑based backend daemon (intentd) that stores all state (workspaces, notes, tasks, git repos, terminals, and the agents themselves) in a local SQLite database and exposes a JSON‑RPC 2.0 API over a Unix‑domain socket (for the desktop client) or secure WebSocket (for remote clients such as an iOS companion app).

Core Features

  • Unified workspace – manage code, notes, tasks and terminal sessions from one UI.
  • Coding agents – plug‑in AI code assistants (Claude Code, Codex, etc.) that run as external processes but are orchestrated by the daemon.
  • Git integration – basic git operations are handled by the daemon and surfaced in the UI.
  • Local‑first & privacy‑first – all data stays on the machine; no telemetry is sent. Network traffic is limited to optional auto‑updates and user‑initiated provider sign‑ins.
  • Cross‑platform clients – Electron + SvelteKit desktop app (macOS, Linux, Windows) and a SwiftUI iOS companion (private for now).

Architecture

Desktop app (Electron + SvelteKit)   <-- JSON‑RPC -->   intentd (Rust daemon)
        |                                            |
        └─ iOS app (SwiftUI) –‑ JSON‑RPC over WSS/TLS ──┘
  • The daemon holds the authoritative state and runs the agent runtime.
  • Clients are thin; they simply render UI and forward user actions to the daemon.
  • Persistence is SQLite; communication uses JSON‑RPC 2.0.

Installation & Deployment

  • Standard user: download the packaged app from [intentapp.dev]; the app ships its own bundled intentd.
  • Advanced / remote: install the intentd daemon separately (via a one‑line script, Homebrew, or a .deb package). The daemon includes a self‑updating “sitter” that fetches the latest binary from the public releases repo.
  • Development: clone the repo with submodules and use the provided make targets (make dev, make dev-prod, make dev-daemon). Build requirements are Rust, Cargo, and optional tools for coverage.

Extensibility & Ecosystem

  • Agent providers – the daemon can connect to multiple AI coding‑agent services (Claude, Codex, etc.) via their CLI/OAuth flows.
  • Integrations – optional connections to GitHub, Linear, Sentry, and user‑defined MCP servers.
  • Remote usage – the daemon can expose a secure WebSocket endpoint, allowing other devices (e.g., the iOS app) to pair and use the same workspace.

Community & Governance

  • The repository is currently a read‑only snapshot; external contributions are paused while the project is being prepared for a public release.
  • Issues are accepted via the GitHub issue forms; security reports should be sent privately.
  • Licensed under Apache 2.0.

Bottom line: Intent is a genuine, open‑source tool for orchestrating AI coding agents and related developer workflow artefacts, with a clear separation between a lightweight UI and a Rust backend that handles state, agent execution, and secure remote access.

Related

  • Project
    ZeKaiNie/universal-examprep-skill

    An AI exam-prep tutor that installs as a skill into coding agents (Claude Code, Codex, Cursor, etc.), teaching you from your own lecture notes and homework with page citations, cropped figures, quizzes, mistake tracking, and cheat-sheet generation.

  • Project
    Stability-AI/stable-audio-metrics

    A collection of evaluation metrics for music and audio generative models, providing tools to calculate Fréchet Distance, KL divergence, and CLAP scores for long-form stereo audio.

  • Project
    xiaYuTian11/maskit

    Maskit is a local privacy‑gateway that automatically masks sensitive data in prompts sent to large‑model APIs and restores the original values in the responses, offering desktop, Docker, and browser‑extension deployments.

  • Project
    Prorise-cool/Claude-Code-Multi-Agent

    Claude Code Multi‑Agent is an open‑source framework that gives Anthropic’s Claude Code project awareness and expert‑agent capabilities. It uses a Python Hook system and a locally‑run Ollama model to automatically detect the codebase type, analyse user intent, load 300+ domain‑specific “Skills” (backend, frontend, testing, security, etc.), and enforce automatic documentation (DEVELOPMENT.md, KNOWLEDGE.md, CHANGELOG.md). After cloning, installing Ollama, a model, and uv, you drop your project into the repo and Claude Code instantly knows its context, suggests appropriate tools, and can be driven with slash commands like `/backend‑specialist` or workflow commands such as `/kiro/spec`. Key points: - Zero‑config start; all configuration lives in `.claude/settings.json` and `prompts.json`. - Hooks run on session start, user prompt, tool use, etc., handling project detection, intent analysis, and doc updates. - Skills are defined by `SKILL.md` files and are auto‑discovered. - Works entirely locally (no external API), integrates with Git, and is MIT‑licensed. Ideal for developers or small teams who want a reproducible, on‑premise AI coding assistant.