Build your software factory.
A directory of agents, orchestration, execution, sandboxes, verification, and observability for building software factories. Compare model support, deployment, pricing, and source code.
No listings match the current filters. Clear a filter or change the search.
Google open-source SDK for agents on the Gemini stack.
Session and tool-call observability for agent runs.
Open-source Python framework for multi-agent systems.
Open-source CLI coding agent with git-aware edits and commits.
AWS coding assistant with IDE, CLI, and agent features.
Sourcegraph coding agent for editor, CLI, and repository context.
Enterprise coding agent for large private codebases.
Agent sandboxes and compute with persistent sessions.
StackBlitz app-building agent running in WebContainers.
Evaluation platform for datasets, experiments, scoring, and logs.
Hosted browser sessions and APIs for agent computer use.
Anthropic coding agent for terminal, IDE, desktop, and browser workflows.
Open-source VS Code coding agent with visible tool approvals.
Cloudflare SDK for stateful agents on Durable Objects.
Cloudflare isolation layer for untrusted and agent-generated code.
GitHub semantic code analysis with query-based CI checks.
Self-hosted remote workspaces defined with Terraform.
AI pull-request review for GitHub and GitLab.
Current library documentation delivered to coding agents.
Open-source IDE agent for VS Code and JetBrains.
Python multi-agent framework with roles, tasks, and a hosted platform.
Charm's open-source terminal coding agent.
Agent-focused IDE with cloud agents running in isolated VMs.
Programmable containerized CI/CD engine for agent workflows.
Persistent development sandboxes for AI agents.
Pytest-oriented evaluation library for LLM systems.
Hosted static analysis with Autofix pull requests.
Nix-powered reproducible development shells for teams and agents.
Cognition cloud coding agent with Devin Desktop and CLI surfaces.
Stanford framework for programming language-model behavior.
Firecracker microVM sandbox API for AI-generated code.
AI pull-request reviewer that can open follow-up fixes.
Commercial agent platform for software delivery workflows.
OpenAPI-to-SDK and documentation generation platform.
AWS open-source microVM monitor used by sandbox products.
API-controlled Fly.io microVMs for ephemeral agent computers.
Enterprise evaluation and observability for AI systems.
Google open-source terminal coding agent for Gemini models.
Google Cloud IDE coding assistant for Gemini models.
Open testing and red-teaming platform for AI applications.
GitHub-hosted development VMs defined by devcontainers.
GitHub coding assistant with an issue-to-pull-request agent.
Automated development environments with a self-hosting path.
Block open-source local-first agent with repeatable recipes.
Stacked pull requests with CLI workflow and AI review.
AI pull-request review with repository-wide context.
Open task platform for durable, observable background jobs.
deepset open Python framework for pipelines, RAG, and agents.
Open LLM gateway with logs, caching, and spend controls.
Event-driven durable functions for product and agent workloads.
UK AISI open framework for evaluating language-model agents.
Google asynchronous coding agent that opens pull requests.
Open IDE and CLI agent with a multi-model gateway.
Self-hostable open-source LLM and agent observability.
Stateful graph runtime for agents from the LangChain project.
LangChain commercial tracing, evaluation, and prompt platform.
LLM observability and evaluation platform with open components.
Stateful agent framework with long-term memory.
Issue tracker with agent-oriented APIs and integrations.
Data and retrieval framework with workflow and agent abstractions.
Pydantic observability platform for application and agent traces.
Hosted full-stack app builder driven by natural-language prompts.
TypeScript agent framework with workflows and evaluation hooks.
GitHub merge queue and pull-request automation rules.
Microsoft open-source SDK for multi-agent applications.
Documentation platform with AI search and maintenance features.
Serverless containers and GPUs for agent workloads.
Open workflow automation platform with agent and LLM nodes.
Cloud workloads and agent sandboxes with BYOC deployment.
OpenAI SDK for Python and TypeScript agent loops.
OpenAI coding agent for CLI, IDE, and cloud tasks.
Open terminal coding agent with configurable model providers.
Open-source software-agent runtime with terminal, editor, and browser.
Cloud inference router exposing models through one API.
Arize open-source observability and evaluation for LLM applications.
Open terminal coding agent for planning large diffs.
AI gateway combining provider routing, guardrails, and logs.
Python workflow orchestrator for scheduled agent jobs.
Open evaluation and red-team framework for LLM applications.
Typed Python agent framework from the Pydantic team.
Code generation and review agents with test-aware workflows.
Open evaluation metrics for RAG and LLM applications.
Hosted development environment with an agent for building and deploying apps.
Open-source Cline fork with modes and automation.
Teachably fast static analysis with optional AI-assisted rules.
Hugging Face small code-agent library with a code-first interface.
Refactoring and review assistant with IDE and CI surfaces.
GitHub toolkit for repository-native, spec-driven agent work.
Capture AI IDE sessions as reviewable intent artifacts.
Typed SDK generation from OpenAPI specifications.
Open-source browser sandbox API for AI agents.
Princeton open-source agent for SWE-bench-style issue patches.
Self-hosted coding assistant for internal model endpoints.
Enterprise coding assistant with private model deployment options.
Durable workflow engine for sequencing long-running jobs.
Specification-first platform for agent-assisted software development.
Shared cloud workspaces for agent runs, previews, and pull requests.
TypeScript background jobs for long-running application tasks.
Unified linters, formatters, and CI reliability checks.
TypeScript SDK for streaming model interfaces and tool-calling apps.
Vercel-managed sandbox for running untrusted code.
Terminal, multi-model agents, and early-access factory workflows.
StackBlitz in-browser Node runtime for agent execution.