Code Assistants

Open-source autonomous coding agent for VS Code and JetBrains with human-in-the-loop approval on every file edit and command

Score 89.48

Open-source AI-native observability that self-instruments a repo with OpenTelemetry then groups errors into incidents and posts one agent-investigated mergeable fix PR in Slack

Score 82.25

Reviews every pull request against version-controlled product specs and flags product drift in AI-written code - pricing logic, compliance, scope creep

Score 75.75

Open-source, model-agnostic terminal coding agent that routes across 75+ providers with a client-server architecture

Score 70.44

Terminal coding agent from Meta powered by the Muse Spark 1.2 model with persistent background subagents in isolated git worktrees and a replay-exact local event log for crash recovery

Score 66.47

Agentic command-line coding tool that reads a codebase, edits files, and runs commands across terminal, IDE, and CI

Score 61.64

Open-source self-improving coding and research agent that gives the model one persistent Python kernel as its only tool and lets it refine its own prompts skills and memory with rollback

Score 61.29

Agent observability and evaluation - traces every LLM and tool call, scores quality and risk in real time, and can pause or block risky runs

Score 57.67

Agentic development environment that runs coding agents in shared-context terminal panes alongside visual QA agents that test web and mobile apps and monitoring agents that watch production

Score 56.85

Cloud and CLI coding agent that drafts, edits, reviews, and tests code across CLI, IDE, web, and desktop surfaces

Score 54.21

AI pair programmer offering inline code completions and an agent mode across major IDEs, with selectable models

Score 48.69

Open-source autonomous coding agent that runs in a sandbox to write code, run tests, and open pull requests end-to-end

Score 44.97

AI-first code editor built on a VS Code fork with multi-file agent edits, a Composer mode, and inline autocomplete

Score 44.93

Autonomous AI platform that crawls mobile apps, auto-generates test cases, and self-heals them on real iOS, Android, and Flutter devices

Score 42.90

Tiny open-source coding agent from Vercel Labs written in Zig as a 6MB native binary that runs with local or cloud models and embeds in sandboxes and larger agent systems

Score 40.32

Open-source agent orchestrator IDE that runs Claude Code and Codex and OpenCode and any ACP agent on your existing subscriptions and extends its own interface from a single prompt

Score 30.32

Open-source local desktop harness that wraps the CLI coding agents you already run into a persistent office of agents coordinated by a clone of you with a paid Teams tier for hosted sandboxes

Score 30.32

Open-source command-line agent that runs end-to-end browser and mobile app tests written in plain English on a real local Chrome or simulator with Playwright interop and a headless CI mode

Score 27.93

Native desktop app that attaches to running CLI coding agents and renders them as a live 3D scene, with checkpoints, file locking, and cost tracking

Score 27.03

macOS voice narrator for coding agents - watches Claude Code, Codex, and Cursor sessions and speaks summaries, needed approvals, and failures with per-agent voices

Score 26.46

Agent-first IDE with multi-agent orchestration, a built-in browser for verification, and Gemini model integration

Score 25.63

Voice command center for coding agents - dispatch tasks, steer mid-run, approve commands, and review diffs by phone call while agents run on your Mac

Score 18.49

Spec-driven agentic IDE that turns prompts into structured requirements, designs, and tasks before generating code

Score 17.66

Autonomous AI software engineer that plans, writes, tests, and opens pull requests, with an integrated desktop IDE

Score 15.62

Free native Mac and Windows app that adds two-way voice to coding agents - on-device speech recognition and synthesis with barge-in interruption

Score 14.68

Prompt-driven app builder that generates full-stack React applications with an automatically provisioned Supabase backend

Score 13.00

Plugin that gives Claude Code and Codex missions - decomposes work into a dependency graph run by a coordinated agent team with an independent reviewer and live mission board

Score 11.16