Agent frameworks & infrastructure

Prompt engineering & prompt libraries

913 agents · most starred first

Prompt engineering & prompt libraries, page 9 of 10

llm-prompt-json

Plugin to add a --json flag to the llm prompt command

llm-prompt-manager

A flexible system for managing and processing prompt templates

llm-prompt-optimizer

A comprehensive framework for systematic A/B testing, optimization, performance analytics, security, and monitoring of LLM prompts across multiple providers with enterprise-ready API

llm-prompt-refiner

A lightweight Python library for optimizing and cleaning LLM inputs

llm-prompt-semantic-diff

A CLI tool for managing and comparing LLM prompts using semantic diffing

llm-promptdiff

Git-style diff and version control for LLM prompts

llm-prompting-gen
llm-reasoner

Transform any LLM into a methodical thinker that excels at systematic reasoning like OpenAI o1 and DeepSeek R1

llm-steer

Steer LLM outputs towards a certain topic/subject and enhance response capabilities using activation engineering by adding steer vectors

llm-structure

An llm plugin to add structured JSON output support to LLMs

llm-structured-output

Constrain LLM generation to structured output, such as function calling and tool use

llm-suite

Universal tools for prompt engineering and LLM interactions.

llm-task-agents

A collection of preset efficient prompts packaged into LLM agents.

llm-templates

Instruction/chat prompts creation library for text generation LLMs. It supports local and Hugging Face models.

llm-templates-fabric

Load LLM templates from Fabric

llm-templates-fsspec

Load LLM templates from local paths and cloud storage (S3, GCS, Azure) using any supported fsspec URL scheme.

llm-templates-github

Laod LLM templates from GitHub repositories

llm-templates-latitude

LLM template loader for Latitude - Load prompts from Latitude as LLM templates

llm-templates-sourcehut

Load LLM templates from SourceHut repositories

llm-text-compressor

Remove characters from text while keeping it understandable for LLMs, saving on inference costs.

llm-tmux-fragments

LLM fragment & template loaders for tmux scrollback and helpful prompt templates

llm-txt-generator

从 YAML 配置生成标准化的 AI 协作规则文档 (llm.txt)

llmcast

Type-safe LLM output parsing with Pydantic models and Jinja2 prompt templates

llmclean

Utilities for cleaning and normalizing raw LLM output

llmdiff-cli

git diff for LLM prompts

llmformat

Format LLM language by using LALR(1) grammar. Supports JSON, XML, etc.

llmhq-promptops

Git blame for prompts: incident archaeology, git-native prompt versioning, and a no-SaaS production runtime for LLM workflows

llmlingua

To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

llmlingua-promptflow

To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

llmpivot

Versioned prompts and Agent Skills with progressive context loading for production LLM apps

llmpromptkit

A comprehensive LLM Prompt Management System

llmpromptvault

Version, compare, and manage your LLM prompts. No API keys required. Bring your own model.

llmstudio

Prompt Perfection at Your Fingertips

llmtricks

A package for LLM tricks

mcts-llm

MCTS + LLM + Prompt Engineering => Enhanced LLM Reponse Quality.

motif-llm

A prompt algebra for LLM orchestration. The code reads as its own specification.

npllm

No Prompt For LLM

olingo-llm-parser

Elegant Jinja2-based chat template parser for AI applications

optimallm

A lightweight Python library to optimize functions using OpenAI's GPT model

parserllm

Force LLMs to use a specific context-free grammar in completions

passive-agent-kahiroka

A passive agent for prompt engineering experiments with instruction-based context building

pguard-llm

Prompt version control and A/B testing for LLM applications

pragma-prompt

A powerful toolkit for building, managing, and composing hierarchical prompts as code.

prellm

preLLM — One function for small LLM preprocessing before large LLM execution. Like litellm.completion() but with decomposition.

pretty-please-llm

🙏🏼 Politely injects PLEASE into your LLM prompts. Because manners matter (kind of).

prompt-llm

A zero-dependency CLI tool to prompt LLMs by reego.software

prompt-storage

Library for managing and caching Google Docs prompts with templating support.

promptify-llm

A simple, zero-dependency library to format prompts for various open-source LLMs.

promptly-llm

A lightweight library for LLM prompt management, observability/tracing, and optimization

pydantic-ai-optimizers

A library for optimizing PydanticAI agents prompts through iterative improvement and evaluation, built on top of PydanticAI + Pydantic Evals.

regex-llm

Controlled language model generation using regex

rellm

Get exact structure out of any language models with regular expressions.

repr-llm

Creating lightweight representations of objects for Large Language Model consumption

schellma

Clean, simplified schemas from Pydantic models for LLM prompts - less verbose than JSON Schema

searchagent

A library of prompts for GenAI

sellm

LLM-powered ProServe manifest generator

sigil-agent

Python SDK for the Sigil persona registry - inject personas into LLM system prompts

smarty-gpt

A wrapper of LLMs that biases its behaviour using prompts and contexts in a transparent manner to the end-users

solvers-hub-llm-json

A Python library to extract and correct JSON from LLM outputs

struct-gpt

structured llm outputs

structllm

Universal Python library for Structured Outputs with any LLM provider

structout-llm

Guaranteed structured output from any LLM — with automatic retries and validation

swarmauri_prompt_j2prompttemplate

Jinja2 Prompt Template for Swarmauri.

swarms-spm

A lightweight, standalone library for managing hierarchical system prompts with support for global, primary, secondary, and numbered prompts

system-prompt-storage

The official Python library for the system-prompt-storage API

task-refiner-llm

Refine unstructured briefs into implementation-ready JSON via LLM7 and llmatch.

taskllm

A library for optimizing LLM tasks, including prompt engineering and bandit-based training.

the-agency-agents

AI agent personalities for every tool. https://github.com/itallstartedwithaidea/agency-agents

tokencrush-langchain

LangChain Runnable client for TokenCrush prompt optimization API (/v1/crush)

toon-llm

Token-Oriented Object Notation – Optimised serialization JSON for LLMs

zenbase_llml

A lightweight markup language for structured text generation

Pretty Prompt

Improve your AI prompts instantly for better results.

@polyskill/tallesborges.agentic-system-prompts

Claude Code Guidelines - Agentic System Prompts Repository

@playsthisgame/melon

A dependency manager for agentic markdown — skills, agents, and prompts for your AI tools.

langchain-prompt-ide

A web-based IDE for LangChain prompts

rule-coherence-graph

Rule Coherence Graph — detect conflicts in AI agent rule corpora.

@os-eco/canopy-cli

Git-native prompt management for AI agent workflows

jrpyllmaip

Jumping Rivers: Prompt Craft

annihilate-llm

Fully automatic censorship removal for language models

dspy-auto-signature

Generate DSPy Signatures from any prompt using a DSPy meta-program

nlp2skillm

Natural language to skillm DSL

llmtrim

Static, deterministic LLM prompt/payload compression that cuts input tokens 30-90% with zero extra model calls.

fix-llm-xml

Practical, ready-to-use tools for LLM generated XML

llmbelt

A tiny, dependency-free tool belt for LLMs: token counting, cost estimation, retries, prompt templates, and text chunking.

agent-prompt-engineering

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.

llm-id-compressor

Bidirectional UUID<->numeric ID mapping to shrink LLM prompt token counts for large item lists.

llm-prompt-engine

Safe LLM prompt templating with validation guards

n8n-nodes-toon-encode

Encode JSON → TOON (40-60% token savings for LLMs!)

@anacatavc/amiga-ia

Universal declarative skills for agents

cohereny

cohereny — runtime prompt adaptation for multi-LLM workflows (coming soon)

litellm-prompt-optimizer

Provider-agnostic prompt optimization pipeline powered by LiteLLM.

ragkit-llmparse

Robustly extract, repair, and coerce structured data (JSON) from messy LLM text output.

agentopt-ai

Self-improving loop for your existing AI agent: write GOALS.md, agentopt measures the gap and closes it - prompts, params, tools, workflow. No model training. Under active development.

@code-ministry/aip

AI profiles shared across agentic harnesses

agent-self-edit

An agent that rewrites its own system prompt from execution feedback

gosce-claude-schema-output-agent

Turn a prompt + a field schema into validated, typed JSON (Instructor over Gemini 2.5 Flash on Vertex AI).

gosce-multi-provider-schema-output-agent

Turn a prompt + a field schema into validated, typed JSON (Instructor over Gemini 2.5 Flash on Vertex AI).

atlante

Atlante command-line interface for structured, versionable AI agent prompts

agent-charters

Structured corpus of AGENTS.md — the written rules humans hand to AI agents | 人写给 AI 智能体的书面规约的结构化语料库

llm-ids

Encode data as token-efficient, LLM-friendly strings

More in Agent frameworks & infrastructure