Python

A guardrails and knowledge-layer tool for coding agents. It briefs agents before tasks, verifies rules with checks, and flags drift when governed files change. Works with Claude Code and other MCP agents.

Project-local MCP server that lets existing agents discover each other, share status, and exchange durable direct messages with explicit acknowledgement.

An MCP server that converts public URLs into clean Markdown for AI agents and MCP-compatible clients. The repo includes setup instructions, a usage example, a direct API option, and notes on supported clients like Claude Desktop and Cursor.

An offline hybrid retrieval engine for RAG that runs on-device, including phones. The repo shows BM25 plus dense retrieval, cross-encoder reranking, sparse expansion support, and bindings for Rust, Python, Swift, and Kotlin.

An in-memory Tree-sitter AST pruning and context firewall MCP server for coding agents like Cursor and Claude. The repo includes installation steps, diagnostics, benchmarks, and a clear usage path for local setup.

A Python package that adds Conduct Guard as a LiteLLM guardrail to enforce runtime policy on LLM calls. The page explains install, proxy wiring, verdict behavior, session tracking, fail modes, and configuration options.

Open-source AI coding factory that turns tickets into reviewed pull requests on your machine. The page shows install options, local-first setup, CLI commands, a desktop app, and product details like adversarial review, tamper checks, and test validation.

A Python SDK that provides LiteLLM-style model routing and translation with a smaller surface area. It supports completions, streaming, embeddings, tool calling, async variants, provider mapping, and local OpenAI-compatible backends.

LeanGuard is a Lean 4-based runtime guardrail system for agent tool use, with native temporal policies, Python integration, and optional MCP support.

Hedgemony is a Python tool that checks AI-written code for fabricated packages, invented APIs, impossible calls, and contradictions between code and stated examples, using interpreter and registry checks rather than another model.

Weir is a CI gate for AI agents that reads OpenTelemetry traces and fails builds when sensitive data reaches forbidden sinks. It includes a gauge for trace coverage, taint tracking, HTML reports, and rule-based scans for agent workflows.

Crilio is a Python CLI and CI gate for testing AI prompts and catching regressions before they ship. It versions prompt tests in crilio.yaml, runs model calls, uses a judge model to score responses, and can block pull requests in GitHub Actions when checks fail.

LayoutLens is an AI-assisted UI testing tool that combines natural-language visual checks with deterministic WCAG audits using axe-core and Playwright. It supports keyless accessibility scans, hybrid LLM-grounded checks, a Python API, and a CLI for running tests in CI.

StateM is a Python CLI for managing long-running AI agents with explicit states, transition checks, runtime history, and resume/compaction support. The repo shows installation, quick-start commands, and a concrete coding-agent runbook example.

A Python computer vision tool that analyzes gameplay videos, detects key events, clips highlights, and generates session reports and montages.

FAAAH is a dependency-free, OpenAI-compatible local proxy that routes requests through text files and AI coding agents. It is aimed at developers who want to reuse an existing agent subscription as a backend for tools that expect the OpenAI API.

Open-source runtime guardrails for AI agents that intercept unsafe tool calls before they execute, aiming to block prompt-injection-driven exfiltration and other policy violations.

Open-source security research tool for auditing Python AI agents and codebases with AST analysis and Docker-based runtime verification.

Open-source runtime action grounding framework for AI agents. It verifies proposed agent actions against live systems of record before execution and can return allow, hold-for-human, or block decisions.

Open-source input scanner for AI agents that strips hidden instructions and scans text, files, images, PDFs, QR codes, audio, and video before they reach an agent.

An open-source CLI for inspecting live LLM inference processes on NVIDIA GPUs. It breaks down VRAM usage by component, shows runtime and model details, and projects memory savings from quantization strategies.

A self-hosted Python runtime for building agentic AI apps with server-driven UI, audit logging, RBAC, sandboxing, and pluggable model orchestration.

A local Rust-plus-Python middleware layer for LLM apps that claims to reduce token usage and add in-process security checks before prompts are sent to a provider.

Wisp is an open-source desktop AI overlay that lets you trigger context-aware prompts with a hotkey, capture selection/clipboard/browser/screen context, and stream replies back into your workflow. It includes voice input/output, privacy controls, model-provider flexibility, and MCP support.