Python agents

Generate adversarial test inputs with a coding agent; your pytest assertions judge them

YouTube transcripts for LLM: load any YouTube transcript as a fragment or tool with the GetYouTubeTranscript API

Drop-in guardrails wrapper for LLM API calls: PII redaction, prompt-injection detection, output schema validation, and usage/latency logging.

General-purpose, multi-objective hyperparameter optimizer for LLM training, with a pluggable model adapter interface.

Tensor math, machine learning, deep learning and model inference helpers: reusable functions and best practices.

Token-level language identification with any LLM provider

Lint LLM backtest prompts and logs for look-ahead (temporal contamination) leakage. Offline, zero dependencies.

LLM clients with a shared model catalog

LLM plugin for the OpenAI Decisions API

A lightweight CLI tool for delegating LLM tasks to expert models across multiple providers.

LLM models from publicai.co

Turn Pydantic models into LLM tools for OpenAI, Anthropic and Gemini.

Turn LLM red-team results (garak, promptfoo) into audit evidence, OSCAL and a model release gate

Prove it before you route: test cheaper LLMs on your own traffic and switch only where quality holds.

Send each prompt to the cheapest model that can handle it, and fall back when one fails

LLM plugin for Runcrate: 200+ open models behind one OpenAI-compatible API

Production-grade, modular dataset sanitization and curation for LLM training.

Cross-platform AI agent skill for llm-security-scanner - OWASP Top 10 for LLMs scanning from Claude, Gemini, Codex and MCP

LLM VRAM calculator and cheapest AWS GPU recommender for vLLM, DeepSeek-R1, Llama 70B, and Qwen. JSON output for AI agents.

Catch silent LLM failures: truncation, reasoning-budget exhaustion, empty output, content filters.

Deterministic output guards for a streamed LLM reply: untrusted links, made-up emails, phone numbers and prices, quoted system prompts and code, stopped before they reach the screen.

Find AI model IDs in your code that are deprecated or about to be shut down (OpenAI, Anthropic, Gemini, Azure, Groq, Cohere, xAI).

Shared foundations for LLM-oriented Python command-line tools.

Proofread Japanese text with the Shodo AI proofreading API

Run a prompt through the claude CLI or the Anthropic Messages API (direct or via a LiteLLM gateway), with one switch and one result shape.

LLM plugin for models available through TrustedRouter

Start and run an LLM-maintained wiki with any agent: presets, workflow guides, search, typed relations and guided navigation.

Open-source framework for cheaper, faster, energy-efficient LLMs. Know before you compress.

Turn any LLM into a decision model: a probability for each predefined option of a question about a text, through a Jev-compatible API

Lightweight Python caching, resilience, semantic caching, rate limiting and observability for LLM/API applications.

Live text channels inside existing warmed Claude, Codex, pi and omo conversations.

Automatically installed offline rescue payload companion for llmcom.

Pick project files in a terminal tree and copy them to the clipboard for an AI chatbot: token counts, sane defaults, no dependencies.

Benchmark Report: a standard data format for LLM inference benchmark runs (schemas, validators, and converters)

A debugger for stepping through LLM layers and interpreting internal representations.

Python (pytest) client for llmentary: deterministic trace-based integration tests for LLM apps

llmf (llmfree): free, keyless LLM chat with automatic multi-provider fallback — stdlib only.

A lightweight, provider-agnostic security and policy enforcement layer for LLM, RAG, and AI-agent applications.

LLM FLock is a CLI tool to interact with local and remote LLM, define multiple scenario variations (model × prompt × options), run them over a dataset and capture structured results for comparison or to build knowledge through meta analysis.

Reliability for LLM calls: retries with backoff, circuit breakers, rate limiting and multi-provider failover.

Cross-model session migration for LLM conversations

Which AI model, and what it costs: LiteLLM prices, Arena ratings and price history, inside Claude and other AI apps.

Extract & repair JSON from LLM replies, parse provider streams, token-aware chunking. Pure Python, zero dependencies.

A pipeline for object detection using LLMs.

LLM-native markup language

LLMON

Keep AI assistant chats, memories and instructions in one git repo, with secrets redacted and checked.

Reports, quick functions and live training tracking for any PyTorch language model: architecture, health, speed, perplexity, behavior, code, and a live training dashboard with alerts.

A lightweight framework to reduce LLM hallucinations, enforce guardrails, manage stateful memory, and monitor performance via local dashboard.

Benchmark-driven router for OpenAI-compatible LLM endpoints

Drop-in prompt-injection firewall for LLM apps

Switch AI models without breaking your product: find model calls, turn real usage into tests, compare models, fix prompts and watch for drift.

Extract video subtitles and translate them into one multilingual SRT.

Lightweight multi-provider LLM streaming parser with sentence buffering for Voice AI, tag guarding, and streaming tool-call assembly. Zero runtime dependencies.

One Python client for chat completions, embeddings and the Responses API across OpenAI, Azure OpenAI, Azure AI Foundry, Anthropic and Vertex AI.

E2E encrypted LLM conversation sync service

Typed structured outputs from LLMs

Live OpenRouter LLM prices + Artificial Analysis intelligence & coding benchmarks dashboard

Call any model with any key, fetched safely from wherever you keep secrets, and see exactly where your money goes

LLM wiki in a Markdown format derived from the Open Knowledge Format: librarian, researcher and CROW classifier routing

Client for Loamz Agents: create, deploy and invoke agents. Placeholder: the first release is coming.

Local-first message bus that lets multiple interactive AI coding agents on one machine discover each other, talk in rooms, and run a human-in-the-loop-by-default task flow — pure stdlib, zero runtime dependencies.

Harness-agnostic controller delegating atomic coding tasks to local Ollama & llama.cpp models via MCP, Agent Skills, and direct CLI.

A small, fully local retrieval-augmented Q&A CLI — no internet needed at query time, no data leaves your machine.

An autonomous web agent driven by a local LLM (LM Studio) and Playwright. Runs entirely on your own machine.

Autonomous AI Testing & Synthetic Chaos Agent

Treat Looker AI-agent instructions and answer correctness as code: lint, compile, verify, attribute, report.

Signed decision, policy, approval and tool-action evidence for Claude Agent SDK

Isolated secure launcher for LoopGrid's Claude Code plugin

Signed, tamper-evident decision evidence for CrewAI Agents using LoopGrid

Signed, tamper-evident decision evidence for Haystack Agents using LoopGrid

Native LangChain agent middleware for LoopGrid signed decision evidence, policy gates and guarded tool execution

Native LangGraph decision evidence, application policy, human-review gates and signed verification for LoopGrid

Native decision evidence middleware for Microsoft Agent Framework and LoopGrid

Signed, tamper-evident decision evidence for OpenAI Agents SDK workflows using LoopGrid

An extensible, evaluable agent SDK built on the Anthropic API.

Multi-CLI validator ensemble for AI work — sprint mode (negotiate/execute/implement/audit), single-shot verbs (ask/review/vote/compare/pipe), declarative graph workflows (flow), natural-language roster management (team), and paste-a-curl onboarding for any AI API. Any AI CLI implements; any AI CLI validates.

LoRA fine-tuning, but more aggressively

Official Python client for the agentx backend (agent-manager, agent-kb, agent-executor).

Local, privacy-friendly macOS disk dashboard: treemap, junk detection, safe cleanup.

Madhav's Act and Transform instruction skill, with a CLI to display and export it.

Magento 1 / OpenMage Core API client (XML-RPC and SOAP)

AIGP governance adapter for AWS Bedrock AgentCore Runtime

AIGP governance adapter for PydanticAI

One command runs the best local AI your PC can handle: measured model picks + a tuned llama.cpp (AMD, NVIDIA, Intel, Apple)

Agentic components for non-life (P&C) insurance pricing, built on mammuthus tools.

Keeps a folder of Markdown notes searchable and lets AI tools edit it by your rules

Mascope's File Agent: watches a folder on an instrument machine and uploads new data files to a Mascope server.

MaskFlow guard for AI coding agents (Claude Code, GitHub Copilot, Cursor, Codex, Antigravity, Gemini CLI, OpenCode, Cline, Devin Desktop): masks PII (Aadhaar, PAN, UPI, mobile, ...) in tool output before the model sees it, and restores values only inside allowlisted commands at run time.

Durable coding goals for local models: bounded context, checked results, and resumable work

Production-ready MCP server scaffold with stateless MCP 2026-07-28, A2A Agent Card, and Agent Skill packaging

MCP server for read-only network device operations (MikroTik, Cisco, Juniper, Aruba, Huawei, Ruckus)

MCP server for auditing Magento projects: update compatibility, module quality, dependency security

MCP Server for OpenAI API via AceDataCloud

MCP Server for OpenAI API via AceDataCloud

Windows-optimized RAG MCP Server with Playwright crawling, Link Graph trees, and local hybrid search.

The prescription for self-improving medical AI. Every model update, by an engineer, a pipeline or an AI agent, is checked against its approved change plan (FDA PCCP, EU AI Act Art. 43(4)).

mediapane plugin: show images Claude Code writes or reads, and an MCP tool so any agent can show media.

Pydantic AI tools and integration for Memanto's persistent memory

A lightweight, extensible Python AI agent framework with model-driven tool calling, durable execution, permission-aware tools, and human-in-the-loop workflows.