Python agents

Governance spine for AgentForge — identity (feat-029) and, as they land, registry / policy / audit drivers

Embedded, file-backed GraphStore for AgentForge (Kùzu)

Reserved - host-agnostic agentic-context engine.

A minimal, dependency-light kernel for a general-purpose AI agent — runs anywhere, brings its own brain.

A reliability harness for scheduled/autonomous agent systems: persist state across runs, prove liveness not just freshness, fail loud, and scan for the failure class before it ships.

Vendor-neutral event logging and redaction for AI agent workflows

The runtime-agnostic AI agent specification. Define once. Run anywhere.

A guard-railed harness that keeps wild agents in check - one gate, sandboxed, reversible context.

AgentProc Protocol SDK + CLI — connect any Agent CLI to a messaging platform

Python package for Agent Profiles.

Self-hosted AgentSave dashboard backend

AgentSave InferRoute — PPD append-prefill routing sidecar for vLLM / SGLang (Enterprise)

AI Agent Runtime Security Platform — runtime security and behavioral monitoring for agentic AI systems

Defensive overlay for AI agents — 8-layer architecture with harness-agnostic adapters; ships 6 of 8 layers (1 skill_vetting, 2 sanitize, 3 structured_output, 4 bash_guard+write_guard, 6 audit, 7 config) plus adapters for Claude Code (PreToolUse) and OpenClaw (before_tool_call)

LLM plugin for ai-mini-box — local and remote inference, RAG

Argus — AI threat-modeling agent (multi-agent, skills)

Axemere AI Gateway Anthropic SDK wrapper — drop-in Anthropic/AsyncAnthropic subclasses

Axemere AI Gateway LangChain integration — ChatAiGateway BaseChatModel

Axemere AI Gateway OpenAI SDK wrapper — drop-in OpenAI/AzureOpenAI subclasses

One-line spending control for AI agents. @budget(max_usd=10.00) decorator for LLM APIs and x402 payments.

Understand your Claude Code usage: token/cost/time analysis and a TUI over local JSONL transcripts.

A typed, durable framework for composable serverless agents on Temporal.

Distributed training and serving framework for models up to 405B parameters.

Claude Code adapter for credence-governor — a thin PreToolUse hook that Bayesian-governs tool calls (proceed/block/ask) via the governor daemon. Requires the credence-governor daemon on :8787 — fails open (silent no-op) without it. Start it: credence-governor-daemon.

CrewAI integration for Infino — vector, BM25, hybrid, and SQL retrieval as agent tools over one engine on object storage.

DataMask LLM 推理引擎 — BERT-NER ONNX 推理、Deepseek 客户端、标注种子管理

Expose djangorestframework-services services and selectors as a Pydantic-AI toolset.

Open mission-assurance commons for secure, authorized, and reviewable AI agents.

Claude Code API for fastllm

LLM provider adapters for GAARD

Hybrid RAG search for Godot documentation

Terminal client for existing ChatGPT web sessions, powered by chatgpt-web-adapter.

Agent-side SDK + CLI for the Handshake Prompt Protocol

Agent SDK — pure agent logic, infra-agnostic

Specialized scientific agent for AD-enabled simulators built on the Jutul framework.

Installer for the kijito-claude toolkit: context tracking, session catch-up, and self-clear scripts plus the Kijito skills, deployed into ~/.claude.

A lightweight Agent framework with extensible model providers

OpenAI component(s) as a standalone Langflow Extension Bundle.

A lightweight metrics exporter for LiteLLM — scrapes Prometheus metrics and serves JSON for dashboards like Homepage and Home Assistant.

LLM plugin for Gonka Broker - OpenAI-compatible access to open-source models on the decentralized Gonka network

Framework-agnostic record/replay test harness for LLM agents

Runtime semantic security layer for LLM-powered applications. The WAF for the AI era.

Local-first, multi-provider tool that captures LLM API spend and exposes it to coding agents via MCP

Fine-tune LLMs (LoRA/QLoRA) and image classifiers locally — browser UI + optional TUI

MAIGP Agent Governance Core — shared lifecycle for all framework adapters (Mediated AI Governance Protocol)

AIGP governance adapter for LangChain/LangGraph

Optics engine for coupled gradient-index light transport simulations.

Privacy-first, SQLite-first observability for local Ollama agents.

A small agent that calls OpenAI-compatible APIs.

统一 AI Agent CLI：整合 Hermes/Codex/OpenClaw 能力

AgentLens: local-first diagnostics & observability for AI agents. Answers 'Why did my agent behave this way?' with one line and a single SQLite file.

A utility library to evaluate LLMs.

Deterministic security middleware for RAG applications to prevent context exfiltration via indirect prompt injection.

LangGraph adapter for Rifft

ZAgent - Lightweight cross-platform AI agent for AICQ, with streaming chat, tool use, workflow automation, and stealth browser capabilities

SigRank local telemetry agent — scans AI-platform logs, computes the 11-metric stack, publishes signed snapshots.

Transparent OpenAI Chat Completions proxy and response normalizer

Agent-first Telegram community management tools.

CLS observability SDK for LangChain/LangGraph — 6-layer span hierarchy, direct upload to Tencent Cloud CLS

Python SDK for creating and managing TryAgent escalations.

Particle Swarm Optimization with a Rust core

Haystack integration for TwelveLabs

Human-in-the-loop review as agent tools for LangChain, CrewAI, LlamaIndex, and AutoGen.

Registro de tráfico (IPs, usuarios, sesiones, endpoints) para apps FastAPI, con visor incluido, listo para Viclix

A local-first coding agent that keeps small local models focused instead of hallucinating.

Memory system for LLM agents -- extraction, retrieval, hybrid search, and MMR reranking

Procedural memory for LLM agents -- skill definitions + invocation history

Long-running-agent foundation -- wake schedules, fires, webhook subscriptions

Expose any Agent Client Protocol (ACP) agent behind an OpenAI-compatible /v1 API

Python orchestration API for Actuate

Aether Core AI v110 — Universal AI Gateway CLI (Python Wrapper)

Local-first process assurance for agentic AI pipelines.

A testing tool for exploring congestion and fault propagation in asynchronous systems

Open framework for simulating compatibility between agents — twin pairing, stress scenarios, no single match score

Deterministic workflows of agents — a Docker-Compose-shaped YAML for composing LLM / code / model / tool nodes into a runnable flow.

Unified MCP server, skill, and hook registration across coding-agent platforms

Context recovery for AI agents — analyze fatigue, compress context, improve token efficiency

A quarantine line for the data your LLM agent ingests: scan untrusted text and outbound actions for prompt injection and exfiltration.

Smart context compression for LLM agents — preserve critical info, reduce tokens 40-80%

Provider-agnostic action defender SDK for AI agents

Real-time memory health monitor and context rot detector for AI agents

Local-first failure diagnosis for AI browser automation, Playwright, crawler, and RPA runs.

Execution-state helpers for app-owned Agent Framework hosting routes.

MCP server exposing AI agent guidance resources, skills, prompts, and project context tools.

A deterministic terminology layer for AI agents.

Repository maintenance checks and diagnostics for AI-assisted Python development.

A Python agent for automated materials science workflows, focusing on surface adsorption analysis

Persistent semantic memory for AI agents with replay, restore, verify, and ignore decisions

Agent OS MVP — agentos_kernel, MCP governance gateway, Neo4j semantic ontology

Claude learns who you are from your own sessions and stays that way

Alpha local personal-data utility for AI agents.

MCP server bridge for Agent Portal.

Deterministic prompt builder on pydantic-stack-core (every prompt = a MetaStack of RenderablePieces) + cognition chains + optional grammar gate — the shared base of the chaincompiler ecosystem.

Release-readiness gates for AI agents: replay a pack of constructed safety scenarios, apply policy-as-code gates, and produce ship/warn/block evidence before an agent, prompt, model, or tool-policy change ships.

Deterministic safety hooks for Claude Code

Static quality linter for AI agent skill definitions (SKILL.md files)

A coordinate-addressed tree of agent skill dirs wired by Read-tool breadcrumbs, with programmatic validation.

Plain-text memory for AI agents — small, filtered, compressed, self-pruning. No database, no embeddings, no dependencies.

A lightweight global Markdown spec and decision vault for AI coding agents.

A CLI tool for scaffolding AI agent projects