llm

Swift package for building provider-agnostic AI agents on Apple platforms, with tools for runtime permissions, hooks, context compaction, session state, validation loops, and traceable execution.

TypeScript middleware that trims or compresses LLM chat history to fit a token budget, with support for OpenAI, Anthropic, and AI SDK integrations.

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.

A research-style demo showing LLM agents evolving through a genetic algorithm to play Pokémon in parallel sandboxed VMs. The page explains the setup, scoring loop, generations, results, and links to the repo and run artifacts.

A ~1000-line C workflow runtime where an LLM can modify a DAG during execution using a restricted four-verb grammar, with every change written to an append-only audit log for replay and inspection.

Open-source Rust coding agent with LLM-native code understanding, shell safety, and support for multiple LLM providers with automatic failover.

Lightport is a lightweight AI gateway that makes LLM providers OpenAI-compatible. The repo shows supported endpoints, a long provider list, quickstart commands, request examples, and active maintenance signals.

An AI-powered Obsidian plugin that turns notes into a self-organizing wiki with auto-generated entity pages, bidirectional links, traceable answers, and local-first support.

A governance plugin for OpenClaw that registers instances with a Recursant registry and intercepts tool calls, LLM prompts, and chat messages for auth, PII redaction, rate limiting, and auditing.

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

A hands-on workshop repository that walks developers through building a small GPT-style language model from scratch, including tokenization, transformer blocks, training, and text generation.

NakshGuard is a self-hosted reverse proxy for OpenAI and Anthropic chat APIs that detects and blocks runaway AI agent loops before they waste tokens or API spend. It includes per-agent session tracking, configurable thresholds, shadow mode, and hot config reload.

Local-first governance layer for AI systems that monitors prompts and responses to block secrets, PII, and runaway LLM spending before requests leave the machine.

Python library for managing LLM conversation history with cache-friendly message ordering, dynamic context, and compaction.

A macOS desktop shell where apps are generated by Claude into interactive HTML mini-apps, with local system features like windows, dock, file storage, notifications, and persona-based themes.

Lathe generates multi-part, hands-on technical tutorials from prompts and provides a local UI plus CLI workflow for working through them with LLM-assisted skills.

A Python filter layer for LLM apps that evaluates responses against a Lean-verified Scherf logic backend and can pass, flag, correct, or block output.

A PyTorch reference implementation of DeepSeek Sparse Attention from the LLMs-from-scratch project, including a lightweight indexer, top-K token selection, and a GPT-style model with KV-cache support.

MagesticAI is an open-source, browser-based platform for managing AI coding tasks with multi-agent orchestration, task boards, terminal access, and code review. The README shows installation steps, Docker deployment, supported platforms, and a concrete local setup path.

A walkthrough for building and deploying an AI agent on Truuze with the AaaS dashboard, including setup, testing, connector configuration, and the platform’s escrow-based transaction flow.

UltraCompress is a Python-based compression tool for large language models. The repo describes lossless 5-bit transformer compression, verification via SHA-256, a CLI on PyPI, and published model packs on Hugging Face.

An open-sourced pipeline that uses Gemini, GoPro footage, Garmin .fit telemetry, Strava segments, and manual labels to automatically build cycling highlight reels.

Reference implementation for using Linux Pressure Stall Information to trim an LLM KV cache under memory pressure. The repo includes requirements, basic usage commands, a simulator, a llama.cpp runner, and benchmark scripts with example results.

Lucid is a browser-based tool for inspecting what a language model is representing before it answers, using Anthropic’s Jacobian lens. It supports small open models, lets users type prompts, view concepts across layers, and export shareable session slices.