Python agents

LangChain integration for QuanticData — document loaders, a web retriever and an agent toolkit for scraping, SERP search, crawling and 74 ready-made collectors through residential proxies.

LangChain middleware that cleans tool outputs: fixes broken encoding, HTML and invisible characters before your agent ever reads them.

LangChain tools for the Web Metadata, OpenGraph & Contact Extractor API — give your agent URL-to-structured-data extraction, AI-ready Markdown, contact discovery, and SEO auditing.

Full, tamper-evident traceability for LangGraph agents — a port of DeepSeek Harness (dsh)'s traceability design: hash-chained, content-addressed, replayable, rewindable.

SOVP to AgenTrust TRACE bridge

LlamaIndex reader for Gluedly scrape snapshots

LlamaIndex tools for OpenDMA - let agents interact with ECM systems

Sprites tools for LlamaIndex agents

LlamaIndex integration for LastSearch — research infrastructure with Grounded Intelligence for AI agents

LlamaIndex reader and tools for the Web Metadata, OpenGraph & Contact Extractor API — ingest any URL as a clean-Markdown Document for RAG, plus agent tools for contacts, tech stack, and SEO auditing.

Local LLM serving for a single RTX 3090, with a browser control panel

Official Mistral API package for llm-api-adapter

Auto-instrumenting observability SDK for LLM inference.

Local, zero-dependency cost tracking and anomaly detection for LLM/agent calls

Confidently navigate LLM deployments from concept to production

A CLI and SDK for routing LLM calls and gating according to price.

Work out which AI models your computer can run, fine-tune, or train.

Parse, normalize, and compare llm:// references (llmns, a name system for language models)

Autopilot for local AI: evidence-driven planning, controlled setup, and measured verification.

Open-source security scanner for AI-powered and agentic applications.

Deterministic project extractor and structured code-access tool for local projects.

LLM-compiled personal wiki toolkit: BM25 + wikilink-graph recall, lint, and chat channels over plain Markdown notes.

Fail-safe client discipline for local LLMs on memory-constrained machines: preflight memory checks, starvation-aware retries, thinking-mode-off, token budgets.

ClickHouse adapter for Meridian V1 evidence and time-series operations

Meridian V1 in-process runtime, binding resolver, and adapter SPI

Meridian V1 evidence Catalog, audit, lineage, provenance, and OTel bridge contracts

Apache Kafka engine adapter for the Meridian Streaming Catalog

Provider-neutral Meridian Object Catalog contracts, streaming utilities, and conformance

PostgreSQL and PostGIS adapter for Meridian V1

Meridian V1 projection, cache, migration, and portability primitives

Engine-neutral query AST, validation, planning, and cursor contracts for Meridian Storage

S3 object adapter for the Meridian V1 Object Catalog

Meridian V1 logical schema, resource, data, and metadata semantics

Provider-neutral Meridian V1 Streaming Catalog contracts and conformance suite

Search Toolkit plugin for Mistral Workflows - durable document ingestion pipelines

Python wrapper to simplify fetching information about models on Amazon Bedrock - particularly pricing

NAVA: Autonomous Personal Agent OS with Governed Execution & Rich Cowork Studio

The sleep cycle for AI memory: every memory carries its evidence; a scheduled pass keeps the evidence honest.

Allows to use server environment with storage backend

OpenAI infrastructure FinOps SDK — scan, analyze, and export your OpenAI org resources

Palapa — local autonomous AI agent built on Ollama

Zero-infrastructure in-memory RAG prototyping framework for Python.

apcore integration for pydantic-ai: governed modules as agent tools, and agent tools as governed modules

DeepKeep AI Firewall custom capability for Pydantic AI

Pytest fixtures for projects built on agentic-fabric

TLLM runs 70B large language models on a single 4GB GPU without quantization, distillation or pruning. 405B Llama 3.1 on 8GB, DeepSeek-V3 671B on ~12GB, Kimi K3 2.8T on under 4GB, Qwen3.8-27B on 3.3GB.

Placeholder release reserving the r3agent name.

A visual debugger for RAG pipelines — see exactly where your pipeline breaks

A local, explainable RAG index for codebases and coding agents

⚡ Evaluate & benchmark RAG strategies, LLMs and embedding models across every popular provider — 18 strategies (incl. graph + multimodal), 100+ models, built-in dashboard.

Evals for RAG pipelines — typed, async, honest.

RAMP: Resource-Aware Model Proxy - an elastic local LLM daemon that auto-calibrates model size to available RAM, VRAM and disk

Kernel-side runtime shim for Zero recursion

A Python module and MCP server for coding agents to understand and write Sassembly/Sanskrit code.

Official SiteGPT SDK for Python — a zero-dependency client for the SiteGPT API v2 (chatbots, knowledge, conversations, leads, messages, onboarding).

结构化独立规划引擎：符号系统主导 + LLM 语义补充的生产级规划 Agent（LLM 全局重构 + 向量记忆 + 多 Agent 协作）

Central provider, model and application routing policy for Subactor LLM consumers.

The published contract a partner builds a Swisper domain agent against

天枢 Agent — 中国本土独立自主通用 AI Agent 框架

Scheduled agent loops that audit a repository and the harness that judges it

Official OpenAI embeddings provider for VERA

A demo hello-world package

Placeholder bootstrap release reserving the vscode-agent-host name; implementation is forthcoming.

Nextcloud plugin for Waldur Site Agent

Watchlight governance hooks for Claude Agent SDK agents: authorize every tool call and sub-agent spawn before it runs, fail-closed. Model turns are not authorized. Independent plugin, not affiliated with Anthropic.

Governance primitives for LangGraph agents with Watchlight: authorize gated actions, validate plans, spawn attenuated sub-agents.

Watchlight governance for Pydantic AI agents: authorize the tools you wrap with guarded_tool() and open every Agent.run as a governed run (nested runs as attenuated sub-agents), fail-closed. Undecorated tools and model turns are not authorized.

Haystack RAG components for Twitter search and public user timelines through TwexAPI. Not affiliated with X Corp.

Typed Python SDK for the xAgent ID customer API.

Statistical evaluation and reporting for LLM-based software engineering systems.

Fresh Claude Code session manager: settings-only template, durable OAuth, quota fallback wrapper, and a web console.

Agentic Control Plane governance for Pydantic AI agents. One ACPHooks() capability; ACP decides allow/deny/redact on every tool call.

Traceable text watermarking for AI-agent long-form deliverables — anti-leak attribution

Install and run Aether Agent, the open-source terminal coding agent, from pip or pipx.

Python support library for the Agent Approval Interchange Specification

Reference emitter for the agent-audit OpenTelemetry semantic convention: proposed -> decided -> executed.

Local control plane for terminal AI agents: run them, message them, coordinate them through GitHub Issues, reach them from Telegram.

LLM agent execution tracer with failure attribution

Static, pre-deployment analysis of what an agentic system can reach in an AWS account from untrusted model input.

Quality gates for agent-built code: an anti-simulation tripwire, an AWS-account tripwire, a local-path tripwire, a denylist tripwire, and an eval harness with confidence intervals and significance testing.

Archive local AI chat histories to S3, freeing up disk space while keeping conversations searchable

자율 에이전트 및 데이터 파이프라인을 위한 경량 설정·로깅·도구 프레임워크 (Lightweight configuration, logging, and tooling framework for autonomous agents and data pipelines)

Deterministic lint and regression gate for AGENTS.md, Claude, Cursor, Copilot, and Gemini instructions.

Reproducible testing, benchmarking, and regression testing for Agent Skills.

Definitions for agent security tools: public rule sets and risk corpora, scored against outside benchmarks, watched for changes, and preserved after the source disappears.

Official Python SDK for the Agent Enrollment Protocol

Intelligent, safety-first, cross-platform file organizer for Windows, macOS, Linux, AI Agent Skills, and FastMCP.

Encrypted, capability-scoped, versioned Markdown file-tree memory for MCP agents

MemorySync for Microsoft Agent Framework: a ContextProvider with budgeted instruction-layer recall, per-run identity, once-per-turn duplicate-proof capture, and fail-open everything.

Outbound change control for AI agents — gate shell, file-write and HTTP side effects before they happen

Artifact-driven development harness for coding agents

Ahp public package coordinate scaffold

Ahp public package coordinate scaffold

Governed project memory for AI coding agents across sessions and tools

模型协议残留处理：识别漏进正文的协议头、孤儿 JSON 尾、思考标记。纯标准库。

A local-first, auditable job runner for AI-agent automation

Ultra-low latency (<3ms) deterministic security, prompt injection, Universal Modular Escrow & Truth Adapters on Polygon, Base, Arbitrum, and Solana Mainnet.

An agent that rewrites its own system prompt from execution feedback

Policy-enforced self-edit gateway for coding agents