Recently Evaluated

Local push-to-talk dictation for macOS that uses Whisper, stays on-device, and pastes text at the cursor.

HookGuard is a CLI security scanner for AI coding agent configuration files. It looks for malicious hooks, invisible Unicode, credential exfiltration patterns, and prompt-injection text in files like CLAUDE.md, AGENTS.md, Cursor rules, and GitHub Copilot instructions.

Open-source, self-hostable sandbox for AI agents. It lets agents work inside isolated Linux containers through MCP tools, with CLI and host-level admin controls for creating, managing, and exposing boxes.

AImpostor is an Electron-based ChatGPT wrapper for macOS that mimics Claude’s visual style and color scheme. The repo includes an install script for placing the app in ~/Applications and a screenshot asset.

Strict Go static analyzer built for LLM-written code, with opinionated checks, annotation-based exceptions, caching, and Claude Code hook integration.

A public visualization that tracks flagship AI models’ Elo history over time using the Arena AI Leaderboard dataset, with notes on caveats and methodology.

A replay page for AgenTank matches, showing battle outcome, match timing, and options to share or export a replay video. It also exposes replay links for humans and agents, suggesting a game built for AI-vs-AI battles and comparison.

An agent skill for starting or extending Django projects, with ready-made scaffolding, stack choices, and safeguards for common production pieces like auth, jobs, Redis, email, and deploy setup.

Open-sourced local multi-agent OS that runs three agents on consumer hardware, lets them form goals from environmental pressure, and supports self-modifying workflows through tool creation and change requests.

AgentKanban is a VS Code task board for agentic coding workflows, capturing Copilot chats and task context so work can be resumed, shared, and tracked across sessions.

An open-source probabilistic spreadsheet that lets cells hold distributions, confidence intervals, and ranges instead of single values, with formulas that propagate uncertainty.

Open-source infrastructure for running and automating sandboxed coding agents, with local Docker setup, integrations, snapshots, sessions, and automations.

An agent-first web browser for automating web tasks, with Playwright integration and a CLI for browser control, page inspection, screenshots, navigation, and file actions.

A loadable SQLite extension that adds column-oriented storage and analytics for fast local OLAP-style queries, with benchmark data and build instructions.

Self-hosted AI observability for tracking LLM requests, costs, latency, prompt traces, reasoning tokens, and PII masking across many model providers.

Hopper is an agentic development environment for mainframes and z/OS. It uses AI agents through Model Context Protocol to navigate TN3270, inspect datasets, write JCL, debug jobs, and work inside mainframe systems from a modern interface.

An open-source 26M parameter function-calling model designed to run on very small devices, with a local quickstart, Python usage, and a playground for testing and finetuning.

Single-binary viewer for HTML reports generated by AI agents, with local browsing, live previews, search, and shareable links.

An experimental programming language and compiler designed for LLM-written code, with a Python-based toolchain that compiles to native macOS binaries. The repo also includes a full NES emulator written in Laze and usage examples for both the compiler and emulator.

A cognitive memory engine for AI agents with temporal decay, contradiction detection, autonomous consolidation, knowledge graph support, and ANN recall via HNSW. The repo shows Rust and Python usage, MCP client setup, and a quick-start path for agents and developers.

Local-first code intelligence for Claude Code that adds structural graph search, semantic memory, cross-project links, MCP integration, and automated hooks for coding workflows.

Community edition of Merlin, a local-first deduplication tool for LLM context and RAG pipelines. The repo includes quick-start install steps, MCP and editor integrations, savings claims, and rollback guidance.

Tessera is a workspace for organizing AI coding sessions across projects, collections, tabs, panes, and Git worktrees. It supports Claude Code, Codex, and OpenCode, with demos for Kanban-style task flow, worktree tracking, tool logs, diffs, and cross-platform runtimes.

A Swedish AI inference API offering GLM 5.1 on owned H200 hardware in Stockholm, with OpenAI-compatible endpoints, SEK pricing, and an emphasis on EU data residency.