awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
kirodotdev avatar

kirodotdev/KiroCrew

0
View on GitHub↗

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
kiro.dev/crew
↗

KiroCrew

KiroCrew is a persistent workspace for AI-assisted development built around an autonomous coding agent. Tasks are planned, carried out, checked against the spec, and retried on failure, with state saved to checkpoints so long-running work can resume after a restart. One shared runtime serves the desktop app, web dashboard, command line, and chat channels such as Slack and Discord, so the same ongoing work is visible from each of them.

The agent improves with use: user corrections and task failures are turned into lasting lessons, applied globally or limited to a single repository, and repeated patterns in past work can become reusable skills that can be inspected, refined, or removed. Large problems can be split across isolated subagents that run in parallel and report back to the main conversation. Work can also start without a person present, through timezone-aware schedules, authenticated webhooks, messaging events, and heartbeat checks on ongoing tasks.

Unattended runs are constrained by layered controls: tool approvals, operating system sandboxing, sensitive-path checks, credential redaction, deny rules, audit events, and governance profiles. Capabilities extend through tool servers, markdown skills, lifecycle hooks, and installable focused apps, without changes to the core runtime. A dashboard shows tool calls, subagent progress, context usage, approvals, schedules, memory, and logs.

The workspace can run on a local machine, in a container, or on a remote host, installed through an installer, desktop package, container image, or source build with a selectable release channel. It sends a minimal anonymous daily heartbeat, which can be switched off through a setting, command, environment variable, or policy.

Features

  • Autonomous Task Orchestration - Runs a runtime that plans, executes, verifies, and retries long-running work, delegates to parallel subagents, and resumes from checkpoints unattended.
  • Agent Self-Improvement Frameworks - Improves the coding agent over time by learning from corrections and failures and turning repeated patterns into reusable skills.
  • Agent Governance - Limits what unattended agents can do through approvals, sandboxing, credential redaction, deny rules, audit events, and governance profiles.
  • Multi-Surface Bot Access - Serves the desktop app, web dashboard, command line, and chat channels like Slack and Discord from one runtime sharing the same work.
4,830 stars·986 forks·Python·Apache-2.0·1 view
  • Lesson Memories - Turns user corrections and task failures into stored lessons, global or per repository, that are injected into later prompts so the agent improves.
  • Planned Long-Running Agent Work - Takes a task spec, plans the steps, runs them, checks results against the spec, retries failures, and resumes from checkpoints.
  • Shared Runtime Across Clients - Connects the desktop app, web dashboard, command line, and chat channels like Slack and Discord to one runtime so work continues across them.
  • Agent Action Guardrails - Combines tool approvals, OS sandboxing, sensitive-path checks, credential redaction, deny rules, audit events, and governance profiles to constrain unattended agent runs.
  • Unattended Task Executions - Plans task spec steps, executes and verifies them, retries failures, and resumes from saved checkpoints so lengthy work finishes without supervision.
  • Parallel Subagent Orchestrators - Lets the agent hand work to isolated subagents that run in parallel and report back so findings combine into one answer.
  • Agent Extensibility Frameworks - Extends the agent through tool servers, markdown skills, lifecycle hooks, and installable focused apps without changing the core runtime.
  • Session State Persistence - Saves task and conversation state to durable storage at regular points so long-running work resumes after a restart or failure.
  • Persistent Cloud Workspaces - Keeps context, memory, and running tasks in a long-lived workspace that survives restarts and is reachable from multiple interfaces.
  • Durable User Correction Memory - Turns user corrections and task failures into stored lessons, scoped globally or to one repository, so later runs do not repeat earlier mistakes.
  • Execution-Driven Skill Synthesizers - Detects repeated patterns in past work and turns them into reusable skills that can be inspected, refined, or removed.
  • Timezone-Aware Schedulers - Runs agent work or scripts on timezone-aware schedules and reacts to webhooks and messaging events, delivering results to chosen destinations.
  • Event-Driven Triggers - Starts unattended agent work from incoming messages, authenticated webhooks, and heartbeat signals instead of waiting for someone to issue commands.
  • Agent Run Scheduling - Runs agent work on timezone-aware schedules and in response to webhooks or messaging events, delivering results to chosen destinations.
  • Scheduled and Webhook Bot Triggers - Starts unattended agent tasks from messaging events and authenticated webhooks, with heartbeat checks that watch ongoing tasks until something needs attention.
  • Sandbox Permissions - Passes every agent action through stacked approval rules, operating system sandboxing, path guards, and credential redaction before it reaches the host.
  • Agentic Planning Loops - Breaks a task into steps, runs each one, checks the outcome against the spec, and retries failures until the work passes.
  • Agent Skill Extenders - Adds tool servers, markdown skills, and hooks that broaden what the agent can do without modifying the core runtime.
  • UI and Hook Extensions - Attaches tool servers, markdown skills, and lifecycle hooks to the runtime through defined interfaces so capabilities grow without changing the core.
  • Shared Core Architectures - Runs one central process that holds all state and serves the desktop app, web dashboard, command line, and chat channels.
  • AI Agent Execution Monitors - Shows tool calls, subagent progress, context usage, approvals, schedules, memory, and logs in a dashboard so running agent work stays visible.

    • Star history

    Star history chart for kirodotdev/kirocrewStar history chart for kirodotdev/kirocrew

    How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

    Frequently asked questions

    What does kirodotdev/kirocrew do?

    KiroCrew is a persistent workspace for AI-assisted development built around an autonomous coding agent. Tasks are planned, carried out, checked against the spec, and retried on failure, with state saved to checkpoints so long-running work can resume after a restart. One shared runtime serves the desktop app, web dashboard, command line, and chat channels such as Slack and Discord, so the same ongoing work is visible from each of them.

    What are the main features of kirodotdev/kirocrew?

    The main features of kirodotdev/kirocrew are: Autonomous Task Orchestration, Agent Self-Improvement Frameworks, Agent Governance, Multi-Surface Bot Access, Lesson Memories, Planned Long-Running Agent Work, Shared Runtime Across Clients, Agent Action Guardrails.

    Which projects share features with kirodotdev/kirocrew?

    Projects with overlapping indexed features include: copilotkit/openbot — OpenBot is a self-hosted workspace where teams run AI bots as coworkers. Each bot gets its own isolated computer with… sandbox-quantum/switch — Switch is a self-hosted platform where people and AI agents share chat rooms. Agents built on different coding-agent… loopx-project/loopx — LoopX is a local-first control plane for long-running coding agents. It keeps goals, todos, claims, gates, quota and… cft0808/edict — Edict is a multi-agent orchestration system and framework designed to coordinate specialized large language model… cursor/cookbook — This repository is a collection of example projects and guides for building with Cursor's coding agents. The samples… google-antigravity/antigravity-sdk-python — This is a Python SDK for building asynchronous AI agents. A single async session connects to an agent backend, sends…

    Projects sharing features with KiroCrew

    These projects share indexed features with KiroCrew. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
    • copilotkit/openbotCopilotKit avatar

      CopilotKit/OpenBot

      10,393View on GitHub↗

      OpenBot is a self-hosted workspace where teams run AI bots as coworkers. Each bot gets its own isolated computer with a browser, workspace files and tools. Agents from any framework can be connected over the AG-UI protocol, and built-in bots can run alongside them. People converse with bots one-to-one or in groups, watch the bot's screen and the commands it runs, and can take over the browser when an agent hits a login wall or a two-factor prompt. Governance is central to the project. Every browser, file and tool action passes through a policy gateway that checks permissions and writes an aud

      TypeScriptag-uiagent-governanceai-agents
      View on GitHub↗10,393
    • sandbox-quantum/switchsandbox-quantum avatar

      sandbox-quantum/switch

      4,232View on GitHub↗

      Switch is a self-hosted platform where people and AI agents share chat rooms. Agents built on different coding-agent harnesses, models or providers can join a room as teammates, and the system runs and coordinates their sessions on a laptop, over SSH, or on remote machines. Rooms can be bridged to channels on Slack, Microsoft Teams, Discord, Telegram and Mattermost, so teammates address agents in the tools they already use. Agent sessions are launched through native SDK or protocol adapters, with providers such as Claude Code, Codex, Cursor and OpenCode driven through one shared session inter

      Pythonagent-orchestrationaiai-agents
      View on GitHub↗4,232
    • loopx-project/loopxloopx-project avatar

      loopx-project/loopx

      6,560View on GitHub↗

      LoopX is a local-first control plane for long-running coding agents. It keeps goals, todos, claims, gates, quota and evidence in durable state outside any chat session, so work carries on across restarts, changes of owner and swaps of agent runtime. Agents are woken on a recurring heartbeat, and each turn is gated by quota and authority before one bounded segment of work runs. A completion claim is accepted only after an independent validator checks the real outcome. What sets it apart is how it handles trust and ownership. An agent's own "done" is treated as unverified evidence, and results

      Pythonagent-control-planeagent-harnessagentic-software
      View on GitHub↗6,560
    • cursor/cookbookcursor avatar

      cursor/cookbook

      4,158View on GitHub↗

      This repository is a collection of example projects and guides for building with Cursor's coding agents. The samples cover running agents from TypeScript code against local workspaces or cloud runtimes, streaming their output live, and launching them from a terminal. Other examples include a cloud agent kanban dashboard, a task graph orchestrator, an app-prototyping chat, and a realtime voice intake demo. Several samples go beyond a basic quickstart. The kanban dashboard lists cloud agents as cards grouped by status, repository, or branch, previews their artifacts, and launches new agents fro

      TypeScript
      View on GitHub↗4,158
    Compare all 30 related projects→