Awesome AI Agents › IDE Integrations
AutoHarness is a plugin for Claude Code that maintains the skill layer of the agent automatically instead of leaving it to be hand-written. It watches ordinary working sessions, distills each episode into a skill derived from the work that was already happening, and writes the result into the project .claude/skills directory in the background. There is no separate data-collection phase and no replay loop. A reflector component compares each new episode against the skills already present and folds same-scenario skills together rather than appending near-duplicates, so the layer consolidates by category as it grows. Skills are validated by adherence in later turns, that is by whether they actually get used, rather than by a held-out benchmark score, which means no oracle sits on the active path and no tokens are spent on a dedicated evaluation run. Skills that stop being used are pruned. The plugin only ever touches skills it generated itself through this plugin; hand-written or separately installed skills are left alone. Every create, update and prune decision is written with its scenario to a per-skill ledger, which the authors describe as the raw material for building a benchmark from real usage later. The stated motivation is the gap between model and harness: the README cites a reported jump from 42 to 78 percent on CORE-Bench for the same model under a different harness, and argues that the harness is still rebuilt by hand each model generation, so at least the skill layer should maintain itself. Installation goes through the Claude Code plugin marketplace with two slash commands followed by a reload, and updating requires refreshing the marketplace catalog first. It requires only python3 on PATH, has no third-party dependencies, runs on Linux and macOS, and is MIT licensed.
https://github.com/tigerless-labs/autoharness
Continue is an open-source platform that enables developers to create, share, and use custom AI code assistants within popular IDEs, enhancing coding productivity and collaboration through features like Chat, Autocomplete, Edit, and Agent.
Coder is a platform for creating secure, self-hosted cloud development environments that streamline developer onboarding and optimize resource usage through automation and integrations.
Sweep is an AI-powered coding assistant plugin designed to enhance productivity and coding efficiency within JetBrains IDEs.
LMQL is a Python-superset programming language designed for constraint-guided and efficient interaction with large language models, enabling advanced control flow, decoding, and integration with multiple LLM providers.
Aide is a Visual Studio Code extension that leverages AI to provide one-click code conversion, commenting, optimization, and batch processing features to enhance developer productivity.
Plugin marketplace and shared configuration for Claude Code, Codex CLI, Cursor and Gemini CLI, bundling skills, slash commands, hooks, subagents and MCP servers.
VS Code extension that runs twelve AI coding agents in one chat interface, with a brainstorm mode where two agents collaborate, selectable personas, permission controls and at-mention routing.
MCP LLMS-TXT Documentation Server is an open-source MCP server that exposes llms.txt files to IDEs and applications, providing controlled, auditable access to LLM documentation for enhanced development workflows.