Skip to content
View darshjme's full-sized avatar
🌴
On vacation
🌴
On vacation
2 followers · 0 following

Block or report darshjme

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
darshjme/README.md

Darshankumar Joshi

Building production-grade tooling for LLM agents.

GitHub followers GitHub stars

Production Agent Toolkit

A suite of four composable libraries for building reliable LLM agent systems.

┌─────────────────────────────────────────────────────────────┐
│                   production-agent-toolkit                   │
│                                                             │
│  ┌───────────┐    ┌──────────────────┐    ┌─────────────┐  │
│  │ llm-router│───▶│react-guard-patt..│───▶│agent-memory │  │
│  │           │    │                  │    │             │  │
│  │  Route    │    │  Guard during    │    │  Remember   │  │
│  │  first    │    │  execution       │    │  always     │  │
│  └───────────┘    └──────────────────┘    └──────┬──────┘  │
│                                                   │         │
│                                          ┌────────▼──────┐  │
│                                          │  agent-evals  │  │
│                                          │               │  │
│                                          │  Evaluate     │  │
│                                          │  after        │  │
│                                          └───────────────┘  │
└─────────────────────────────────────────────────────────────┘

Route first. Guard during. Remember always. Evaluate after.
Library What it does Tests
agent-evals 3D evaluation framework for LLM agent outputs 45
react-guard-patterns Stop-condition guards for agentic loops 14
llm-router Semantic task routing across LLM providers 10
agent-memory Short-term + episodic memory for agents 13

82 tests total across the toolkit. → production-agent-toolkit

Other Projects

  • brahmand — Universal LLM terminal interface
  • samast — LLM integration framework

Find me

@thedarshanjoshi on X

Popular repositories Loading

  1. mohini mohini Public

    Mohini Agent OS — The world's most complete autonomous agent runtime. 14-crate Rust workspace | 104 skills | 40 channels | 188 models | WASM sandbox | Self-healing | Zero downtime.

    Rust 3

  2. darshjme darshjme Public

    Full-stack dev building AI agents, CLI tools & trading systems in Rust, TypeScript, Python

    1

  3. mycryptocoin mycryptocoin Public

    MyCryptoCoin — The Stripe of Crypto Payments. Multi-chain payment gateway supporting BTC, ETH, USDT & 10+ cryptocurrencies with 0.5% fee.

    TypeScript 1

  4. whatsapp-2fa whatsapp-2fa Public

    WhatsApp 2FA/OTP Gateway — Self-hosted authentication service powered by Baileys. Production-ready REST API for sending & verifying OTPs via WhatsApp.

    JavaScript 1

  5. stable-diffusion-webui stable-diffusion-webui Public

    Forked from AUTOMATIC1111/stable-diffusion-webui

    Stable Diffusion web UI

    Python

  6. knowai-erp knowai-erp Public

    KnowAI ERP v1.0 — AI-powered enterprise management platform. React 19 + Next.js 15 + PostgreSQL. Multi-tenant, real-time inventory & AI insights.

    TypeScript