#llm

FastroAI wraps PydanticAI with production essentials: cost tracking in microcents, multi-step pipelines, and tools that handle failures gracefully.

LLM agent evaluation framework with 7 Harness Gates (A–G) and 58 metrics. Supports LangChain, CrewAI, AutoGen, DSPy, PydanticAI. Native LLM-as-Judge, OTEL tracing, FastAPI dashboard.

Block engine for AI-composable web apps — snap-together 'Lego' blocks + the AI catalog contract (manifest → validate → render static HTML).

DataForSEO API modelcontextprotocol server

An opinionated full-stack agent harness with native memories, long running tasks, and multi-agent relay

Evi — local-first personal AI assistant (CLI + web + desktop) over local LLM backends

Fast, pure-Go document-to-Markdown conversion for PDFs, Office files, HTML, CSV, ZIPs, and text—available as a CLI and modular library.

Your backlog is your agent's memory

Provider-agnostic multi-LLM abstraction for the Greentic platform

No repository description published.

Your prompts need tests too. Run prompts against real datasets, score outputs with LLM judges, version everything, and compare runs to see what got better.

Agentic engineering supersystem — one CLI to run the full /mavis-ship chain. Bundles 38 curated integrations + 1890 skills + the v2.7.1 arsenal (orchestrate.py + catalog.py + autopilot + mavis-ship-uni + big-boss persona). MIT.

Shrinks your AI agent's tool output so it burns fewer tokens, without changing what the agent reads. Lossless by default, no decode step.

distributed agent framework

A RWKV management and startup tool, full automation, only 8MB. And provides an interface compatible with the OpenAI API. RWKV is a large language model that is fully open source and available for commercial use.

Providing unified trait abstractions for AI models

A pure-Go, single-file AI memory and knowledge graph library and plugin.

Smart Ai Voice Assistant written in Python

Research-grade investment decision engine for AI agents: isolated multi-agent committee, auditable verdicts, backtests with lookahead protection, published negative results

Provider-agnostic TypeScript agent SDK — pluggable providers, built-in tools, skills, memory, and MCP

An AI-driven dynamic protocol fuzzer for the Model Context Protocol (MCP). Prove runtime exploitability by discovering state violations, transport crashes, and application-layer logic flaws (SSRF, LFI) before your AI agents do.

No repository description published.

Unofficial Go SDK for E2B cloud sandboxes — create, manage, and run commands in secure microVMs.

Run any local LLM engine, auto-tuned to your GPU — polished web UI + OpenAI/Anthropic-compatible API. Point Claude Code at your own machine in one command. No Electron, no Python, offline-first.

31 MCP tools that give Claude, Cursor and any MCP client the live web — scrape, crawl, search, real Google rank, change tracking, document parsing, plus an autonomous agent that researches from a plain-English prompt with no URLs. Clean Markdown and schema-validated JSON, not raw HTML. Local-Ollama extraction by default. MIT, 1,000 free credits.

Route inference calls to different providers based on a fallback condition, or a routing rule you define.

yet another tElegram bot api library for node.js (deno and bun TOO!!!!!!!!!!!!!!!)

Multi-agent collaboration platform for persistent, proactive AI agents across rooms, workspaces, skills, and external services.

Reusable Rust engine (crate) for self-hosted advisory AI PR reviewers — reviews GitHub, GitLab & Bitbucket PRs. Line-anchored inline comments, committable suggestions, tree-sitter structural context, OSV.dev CVE scans. Ships as one binary (npm, PyPI, crates.io, curl) with JSON output, so the bot around it can be any language that spawns a process.

Local MCP server that helps coding agents stop re-reading the same code every turn. Repository intelligence, cache-write optimization, and long-term session memory. 100% local. No API keys.