30 October 2026

Lifelong personalized tutoring

A personalized tutoring system built for long-term learning. Release v1.7.0 (Oct 28) adds modular installation, a refreshed learning workspace, and narrated video lessons.

• Block notebooks with Dito and narrated video lessons arrived in v1.7.0.
• Learning-path changes are reviewed first, with grouped assessments and bounded tutoring memory (v1.6.15).
• Embedding can resume from checkpoints, and the interface comes in several languages such as French, German, and Ukrainian.

A complete software development method for coding agents

A skills framework with starter instructions that makes a coding agent ask questions first, write a spec, then a plan, before writing code.

• The agent digs into your goal, then shows the spec in short chunks that are easy to read and approve.
• The implementation plan stresses red/green TDD, YAGNI, and DRY.
• Once you say go, a subagent-driven-development process works through each task and checks the result. It supports Claude Code, Codex, Cursor, Gemini CLI, and more.

Design vocabulary for coding agents, with a built-in reviewer

Commands like polish and bolder steer the agent, while critique and audit review the result. The aim is to avoid generic AI-looking interfaces.

• critique scores a design against usability heuristics and ranks fixes. audit checks accessibility, performance, theming, and responsive behavior, with findings from P0 to P3.
• The slop detector has 62 deterministic rules, with no LLM and no API key.
• It works in Claude Code, Cursor, Codex CLI, Copilot, Gemini CLI, and many other tools.

JetBrains agentic IDE built around the pi coding agent

A lightweight IDE that wraps the pi agent in a full interface. The agent learns your project through living specs and reusable skills.

• Each workspace is a git worktree with its own branch, so agents work in parallel without touching your checkout.
• Finished or dormant workspaces move to a per-project shelf and return when you work in them.
• From the topbar you can push, open a PR, see CI checks and the review decision, then merge or hand failing checks to the agent.

Open source Grok Bot alternative: a team of AI bots in a chat app

A chat app with a roster of AI bots. Each bot is a real agent (Claude, Codex, or Grok) with its own personality, model, cloud computer, and connected apps.

• Bots run on the claude, codex, and grok CLIs already installed on your machine, using models you already have.
• You talk to bots like contacts, watch them work, and approve what matters.
• Installers exist for macOS, Windows, and Ubuntu. The README states the project is independent and has no crypto token.

Local-first desktop workspace for persistent AI teammates

A local-first desktop app that gives every agent its own workspace and conversation. It supports Codex and Claude Code, plus Grok CLI, OpenCode, Gemini, Cursor CLI, and Cline CLI through ACP.

• It has crash-safe FIFO message queues, file transfers, an embedded browser, and agent-to-agent messaging.
• It is a development preview: agents run with danger-full-access and approvalPolicy: never, so run only agents and tasks you trust.
• It can use OpenAI-compatible endpoints and detects local model servers such as Ollama and LM Studio.

A collection of design skills for AI coding tools

A collection of self-contained skill folders for Claude Code, Cursor, Codex, Windsurf, and other tools that read a markdown rules file. For now it holds one skill.

• The landing-page-design skill builds a landing page from scratch: intake questions, page structure, conversion copy, and a visual system.
• Design values such as fonts, colors, spacing, and motion sit in one section of the file, so they are easy to change after a fork.
• Install by copying the skill folder into .claude/skills, or copy SKILL.md into another tool's rules file.

Persistent memory across sessions for coding agents

Captures what the agent does during sessions, compresses it with AI, and injects relevant context into future sessions.

• It works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode, and more.
• After a magic-link email sign-in, you pick a memory provider: the claude-mem observer, your own OpenRouter or Gemini key, or your Anthropic plan.
• For Grok Bot, the plugin reads chat log files because there are no host hooks.

Open source AI coworkers that each get a computer of their own

A template to clone: each AI coworker gets its own browser, files, and tools, and every action is decided before it happens and recorded after. Bring any AG-UI agent.

• It is a template, not a product: there is no hosted version, and you replace the example tenant package in examples/ with your own coworkers, channels, and skills.
• You can watch the agent on its own screen, take over when it reaches something it should not do alone, then hand it back.
• It is in beta. It runs on your own machine, and .env.example sets OPENBOT_SINGLE_USER=true.

A collection of prompts that affect AI model behavior

A repo of 32 prompts for 26+ models: system prompts, personas, and configuration files. The README describes it as for research and education.

• It contains system prompts, persona definitions, workspace policies, AGENTS.md files, and Ollama Modelfiles.
• It covers open-weight models such as GLM, DeepSeek, Kimi, and Qwen, and closed models such as Claude, Gemini, and Grok.
• The README warns against using it to harm others or to break laws or provider terms.

Turn a Claude Code session into a full game dev studio

Gives a Claude Code session studio structure: 49 agents in a hierarchy and 74 skills as slash commands for each work phase.

• Directors guard the vision, leads own each area, and specialists do the hands-on work, each with defined responsibilities and escalation paths.
• The agents cover design, programming, art, audio, narrative, QA, and production.
• Slash commands such as /start, /design-system, /create-epics, /create-stories, /dev-story, and /story-done follow the work phases.