Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

awesome-prompts

Curated list of chatgpt prompts from the top-rated GPTs in the GPTs Store. Prompt Engineering, prompt attack & prompt protect. Advanced Prompt Engineering papers.
https://github.com/ai-boost/awesome-prompts

Last synced: 15 days ago
JSON representation

  • Agent Ecosystem

    • A2A — Agent-to-Agent Protocol

    • Agent Skills

    • Frameworks

      • **LangGraph** - grade workflows (Nov 2025 stable release) |
      • **CrewAI** - based multi-agent teams |
      • **Magentic-One** - capability agents (web + file + code + terminal) |
      • **OpenAI Agents SDK** - native orchestration (Mar 2025) |
      • **OpenAI Agents SDK for JS/TS** - agents-js?style=flat-square) |
      • **GitHub Agentic Workflows (gh-aw)** - first agentic workflows for GitHub Actions — Markdown workflow specs, sandboxed execution, structured outputs, approval-aware automation (2026) ![](https://img.shields.io/github/stars/github/gh-aw?style=flat-square) |
      • **Google ADK** - native development (Apr 2025) |
      • **Claude Code**
      • **karpathy/autoresearch** - line self-improving agent — reads its own training code, forms hypotheses, runs experiments overnight (Mar 2026) ![](https://img.shields.io/github/stars/karpathy/autoresearch?style=flat-square) |
      • **Microsoft Agent Framework** - driven actor model, multi-agent orchestration (RC 2026) ![](https://img.shields.io/github/stars/microsoft/agent-framework?style=flat-square) |
      • **openai/codex** - mini powered, runs in terminal (Apr 2025, active 2026) ![](https://img.shields.io/github/stars/openai/codex?style=flat-square) |
      • **DeerFlow 2.0** - horizon "SuperAgent" — filesystem, sandboxed execution, persistent memory, parallel sub-agents, skill system; LangGraph-based; hit #1 GitHub Trending on launch day (Feb 28, 2026) ![](https://img.shields.io/github/stars/bytedance/deer-flow?style=flat-square) |
      • **smolagents** - first agent framework (~1000 LOC core) — MCP integration, multi-agent hierarchies, multimodal I/O, 100+ model providers ![](https://img.shields.io/github/stars/huggingface/smolagents?style=flat-square) |
      • **browser-use** - driven browser automation — agents control a real browser to complete web tasks; 89% on WebVoyager benchmark ![](https://img.shields.io/github/stars/browser-use/browser-use?style=flat-square) |
      • **Mastra** - first AI agent framework — Agent/Workflow/RAG/Evals primitives, 40+ model providers, native MCP server support (YC W25, 2026) ![](https://img.shields.io/github/stars/mastra-ai/mastra?style=flat-square) |
      • **PraisonAI** - ready multi-agent framework — 100+ LLM providers, MCP integration, memory/RAG/guardrails, 24/7 delivery to Telegram/Discord/WhatsApp, fastest agent instantiation (2026) ![](https://img.shields.io/github/stars/MervinPraison/PraisonAI?style=flat-square) |
      • **Portia AI** - source predictable agent framework — 1000+ cloud/MCP tools, built-in auth, auditability and security focus for enterprise workflows (2026) ![](https://img.shields.io/github/stars/portiaAI/portia?style=flat-square) |
      • **Paperclip** - human-company multi-agent orchestration — org charts, budgets, goal management, CEO→Manager→Worker delegation; 48k stars in 3 weeks (Mar 2026) ![](https://img.shields.io/github/stars/paperclipai/paperclip?style=flat-square) |
      • **Gemini CLI** - source terminal AI agent — ReAct loop, MCP support, 1M context window, Gemini 2.5 Pro/3 Flash/3.1 Pro; free tier (60 req/min); Apache 2.0; v2.0 Apr 2026 ![](https://img.shields.io/github/stars/google-gemini/gemini-cli?style=flat-square) |
      • **oh-my-codex** - agent execution, notification routing; 23k+ stars (2026) ![](https://img.shields.io/github/stars/Yeachan-Heo/oh-my-codex?style=flat-square) |
      • **Hermes Agent** - improving agent framework built on Hermes 3 — persistent memory across sessions, learns from interactions, multi-platform messaging; 32k+ stars (2026) ![](https://img.shields.io/github/stars/NousResearch/hermes-agent?style=flat-square) |
      • **claw-code** - development demo in Rust — human sets direction via chat, claws self-coordinate (plan/build/test/review/push); notification routing kept outside agent context; fastest repo to 100K stars (Mar 2026) ![](https://img.shields.io/github/stars/ultraworkers/claw-code?style=flat-square) |
      • **PilotDeck** - isolated agent OS — white-box memory, smart model routing (~70% cost savings), always-on background execution, MCP-native; productivity platform for multi-project agent workflows (May 2026) ![](https://img.shields.io/github/stars/OpenBMB/PilotDeck?style=flat-square) |
      • **AOS CE** - harness loops, MCP bridge; composable user-space layer for harnesses and agent-native software (July 2026) ![](https://img.shields.io/github/stars/unicity-aos/aos-ce?style=flat-square) |
      • **agent-browser** - labs/agent-browser?style=flat-square) |
      • **nanobot** - lightweight self-hosted personal AI agent framework in Python — WebUI, CLI, chat apps, tools, memory, MCP, multi-agent workflows, automation, OpenAI-compatible API (Feb 2026) ![](https://img.shields.io/github/stars/HKUDS/nanobot?style=flat-square) |
      • **Flue** - harness framework — sessions, tools, skills, sandboxes, durability, and subagents; compose the full harness an agent needs to do real work, run locally via CLI or deploy to a hosted runtime (Feb 2026) ![](https://img.shields.io/github/stars/withastro/flue?style=flat-square) |
      • **Agno** - first agent framework — memory, knowledge, tools, multi-agent teams, and structured workflows; rebrand of phidata (2026) ![](https://img.shields.io/github/stars/agno-agi/agno?style=flat-square) |
      • **Qwen-MM-Plugins** - native — vision, audio, and video plugins that wire into existing agent frameworks via MCP/tool interfaces (July 2026) ![](https://img.shields.io/github/stars/QwenLM/Qwen-MM-Plugins?style=flat-square) |
      • **codebase-memory-mcp** - performance code-intelligence MCP server — tree-sitter + Hybrid LSP knowledge graph, 15 MCP tools, indexes Linux kernel in 3 min, 120× fewer tokens than file-by-file exploration (Feb 2026) ![](https://img.shields.io/github/stars/DeusData/codebase-memory-mcp?style=flat-square) |
      • **TencentDB Agent Memory** - level memory hub for AI agents — turns conversations, docs, and code into four reusable memory assets (Chat Memory, Skill, LLM-Wiki, Code-Graph) that are governed, shared, and equipped across agents and frameworks (Apr 2026) ![](https://img.shields.io/github/stars/TencentCloud/TencentDB-Agent-Memory?style=flat-square) |
      • **agentmemory** - scored facts/procedures/sessions, hybrid dense+keyword+graph retrieval, 54 MCP tools, 12 auto hooks, 95.2% R@5, 92% fewer tokens; supports Claude Code, Cursor, Codex, Gemini CLI, Hermes, OpenClaw, pi, OpenCode, and any MCP client (Feb 2026) ![](https://img.shields.io/github/stars/rohitg00/agentmemory?style=flat-square) |
      • **eve** - first framework for durable backend AI agents — instructions, tools, skills, channels, schedules, connections, and subagents as files; path-named capabilities, typed tools, `eve eval` harness (June 2026) ![](https://img.shields.io/github/stars/vercel/eve?style=flat-square) |
      • **kimi-code** - source terminal AI coding agent — single-binary TUI, Kimi K3 + OpenAI-compatible providers, `/goal` judge mode, coder/explore/plan subagents, AI-native MCP config, Skills, lifecycle hooks, video input; MIT (May 2026) ![](https://img.shields.io/github/stars/MoonshotAI/kimi-code?style=flat-square) |
      • **Orca** - backed (Mar 2026) ![](https://img.shields.io/github/stars/stablyai/orca?style=flat-square) |
      • **herdr** - native runtime for coding agents — background server with persistent sessions, agent-aware pane states (working/blocked/idle), detach/reattach across terminals and SSH; Rust, Apache-2.0, 34k+ stars (Mar 2026) ![](https://img.shields.io/github/stars/herdrdev/herdr?style=flat-square) |
      • **OpenSRE** - source AI SRE agent framework — investigate production incidents across 60+ tool integrations, synthetic RCA simulations, real-world e2e tests across Kubernetes/EC2/CloudWatch/Lambda, reversible PII masking, headless CLI and REPL (Jan 2026) ![](https://img.shields.io/github/stars/Tracer-Cloud/opensre?style=flat-square) |
      • **DeepSeek Harness** - first open-source agent harness — everything (tools, skills, UI, memory, models) is a hot-swappable plugin; Cordis-based composability; ships with Web UI and headless CLI (developer preview, Aug 2026) ![](https://img.shields.io/github/stars/deepseek-ai/deepseek-harness?style=flat-square) |
      • **TrueForge** - source agent harness — runtime layer that turns an LLM into a working agent; chat UI, HTTP API + TypeScript SDK, MCP tools, git-backed skills, sandbox-as-tool, approvals, context compaction; local SQLite or hosted Postgres/Redis (MIT, Aug 2026) ![](https://img.shields.io/github/stars/truefoundry/trueforge?style=flat-square) |
      • **OpenBot** - source AI coworkers that each get a computer of their own — browser, files and tools; every action decided before it happens and recorded after; bring any AG-UI agent (MIT, Aug 2026) ![](https://img.shields.io/github/stars/CopilotKit/OpenBot?style=flat-square) |
      • **qm** - agnostic core (Pi, OpenCode, Codex, Claude Code all drive the same loop), admin-gated org security posture, scope-shared skills with pack imports, web apps and background crons (TypeScript, MIT, 14.5k+ stars, Aug 2026) ![](https://img.shields.io/github/stars/yc-software/qm?style=flat-square) |
      • **Omnigent** - source meta-harness — a common orchestration layer over Claude Code, Codex, Cursor, OpenCode, Hermes, Pi, and custom YAML-defined agents; mix and supervise multiple agents in one session, swap harnesses without rewriting, policies/sandboxing/approvals, cloud sandbox backends (Modal, E2B, K8s, Databricks…), sessions synced across terminal/browser/phone/desktop (Python, Apache-2.0, 9.7k+ stars, June 2026) ![](https://img.shields.io/github/stars/omnigent-ai/omnigent?style=flat-square) |
      • **Claude Agent SDK** - agent-sdk-python?style=flat-square) |
      • **commerce-agents** - builder Claude Code plugin that scaffolds and reviews your own deployment (Apache-2.0, 2.8k+ stars, Sept 2026) ![](https://img.shields.io/github/stars/anthropics/commerce-agents?style=flat-square) |
      • **OpenHuman** - first personal AI harness built in Rust — a "brain that remembers everything" (persistent memory), plus agent orchestration and deep-research workflows, with the human kept in the loop (GPL-3.0, 40k+ stars, Feb 2026, actively maintained) ![](https://img.shields.io/github/stars/tinyhumansai/openhuman?style=flat-square) |
      • **phone-harness** - compatible agents (MIT, 2.7k+ stars, Aug 2026, actively maintained) ![](https://img.shields.io/github/stars/ShawnPana/phone-harness?style=flat-square) |
      • **Artemis** - language Android automation that lets AI assistants drive real devices like a human — cross-app workflows, multimodal element targeting (indices + coordinate/visual fallbacks), reactive observe-and-act loop (~3–5s/step) with asynchronous history summaries, proactive exploration with blocked-action recovery; MCP-native diagnostics (Logcat, screenshots) for Claude Code, Codex, and Windsurf; 99%+ task completion on AndroidWorld; Gemini/Claude/GPT-4o/Qwen-VL multimodal (Python, Apache-2.0, 7.5k+ stars, Aug 2026, actively maintained) ![](https://img.shields.io/github/stars/google/artemis?style=flat-square) |
      • **Reef** - Agent Society | Continual-learning infra for self-improving agents — connects agent inference, feedback, learning, and versioned delivery in one serve → observe → grow → commit loop; either train model weights (Slime/SGLang integration) or evolve the harness itself (prompts, rules, skills) with no local training GPUs; versioned artifact history with candidate evaluation and selection policies (Python, Apache-2.0, 2.8k+ stars, Aug 2026, actively maintained) ![](https://img.shields.io/github/stars/Human-Agent-Society/reef?style=flat-square) |
      • **Dormice** - hosted, E2B-compatible sandbox platform for AI agents. Inverts cloud sandbox economics: one daemon + one SQLite ledger on a machine you already pay for, and sandboxes are permanent — they cool down an idle ladder (`active → frozen → stopped → archived`) so idle costs nothing (~5 MiB resident frozen, ~50 ms wake, files intact). `acquireSandbox(key)` is the entire mental model (idempotent create/wake/restore). Docker + gVisor-isolated execution, one-binary deploy (no K8s), multi-node fleet mode, signed file URLs, real PTY streaming — and the official `e2b` SDK works unmodified by changing two URLs; ships an Agent Skill so coding agents can drive it directly (TypeScript, Apache-2.0, 1.2k+ stars, July 2026, early development, actively maintained) ![](https://img.shields.io/github/stars/BitMiracle-AI/Dormice?style=flat-square) |
      • **OpenConnector** - source connector gateway for AI agents — an alternative to Pipedream/Composio. Connect user app accounts once, then expose 1,000+ providers / 10,000+ prebuilt Actions through SDK, CLI, MCP, HTTP, and OpenAPI from one inspectable runtime. Credential handling for API keys, OAuth2, and custom credentials; scoped runtime tokens, action allow/block policies, redacted run logs — provider credentials never enter the agent process. Self-host with Docker/Node.js (SQLite or Postgres) or use the hosted runtime (TypeScript, Apache-2.0, 5.8k+ stars, June 2026, actively maintained) ![](https://img.shields.io/github/stars/oomol-lab/open-connector?style=flat-square) |

    • Harness Engineering

      • Aakash Gupta
      • Harness Engineering — OpenAI - first world" |
      • Improving Deep Agents with Harness Engineering — LangChain
      • The Importance of Agent Harness in 2026 — Philipp Schmid
      • Skill Issue: Harness Engineering for Coding Agents — HumanLayer - agents as context firewalls, practical patterns |
      • Effective Harnesses for Long-Running Agents — Anthropic - running agent design |
      • SethGammon/Citadel - tier routing, parallel worktrees, lifecycle hooks, 6 skills |
      • langchain-ai/deepagents
      • Building a C Compiler with Parallel Claudes — Anthropic - agents to build a C compiler — generator/evaluator harness patterns |
      • strukto-ai/mirage - ai/mirage?style=flat-square) | Unified virtual filesystem for AI agents — mounts S3, GDrive, Slack, Gmail, Redis as one tree; agents use bash across every backend; Python/TypeScript SDKs, cache, snapshots (May 2026) |
      • QoderAI/better-harness - harness?style=flat-square) | Open-source loop/harness improvement skill — turns project and session evidence into prioritized improvements and verifiable next steps for Claude Code, Codex, Cursor, and other coding agents (July 2026) |
      • **harness-engineering** - engineering?style=flat-square) | Ryan Lopopolo's anthology, field guide, and agent context bundle for harness engineering — shaping context and tools so agents can recover intent, operate systems, respect authority, prove outcomes, and leave the next run better equipped (CC-BY-4.0, July 2026) |
      • **loop-engineering** - engineering?style=flat-square) | "Stop prompting. Design the loop." — practical patterns, starters & CLI (`loop-audit`, `loop-init`, `loop-cost`) for systems that discover work, hand it to agents, verify results, and persist state across Claude Code, Codex, Grok, and OpenCode; report-only week one, scores loops on a "Loop Ready" rubric (MIT, 11k+ stars, June 2026, actively maintained) |
      • **ECC** - m/ECC?style=flat-square) | The agent harness performance optimization system — skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond (MIT, 242k+ stars, Jan 2026) |
      • **SoL-Pi** - Pi?style=flat-square) | NVIDIA's efficiency extension for the Pi coding agent — four opt-in harness mechanisms distilled from scaled auto-research loops ([arXiv 2609.20519](https://arxiv.org/abs/2609.20519)): Action Fusion (run follow-up validation in the same tool call), ObservationPack (stable handles with exact paged recall for repeated large outputs), evidence-preserving reducer (long logs → compact receipts where every quotation is verifiable against the archive), and online context compaction gated by economics/window pressure; no patches to Pi, evidence always preserved (TypeScript, MIT, 2.9k+ stars, Sept 2026, actively maintained) |

    • MCP — Model Context Protocol

  • Basic Prompt Engineering

  • Context Engineering

    • Prompt Attack & Defense

      • MIT Technology Review
      • Effective Context Engineering for AI Agents — Anthropic
      • Context Engineering Guide — Prompt Engineering Guide
      • davidkimai/Context-Engineering - Engineering?style=flat-square) — first-principles handbook on context design, orchestration, and optimization
      • Meirtz/Awesome-Context-Engineering
      • **NanoNets/Graft** - square) — Open-source codebase context layer for coding agents — builds a persistent, queryable code graph and pulls matching nodes into each prompt; 46% fewer tool calls, 42% token savings, 60% faster on a 162-run benchmark, 66% SWE-bench Verified (vs 54% cold); supports Claude Code, Cursor, Codex, Gemini via MCP/CLI (MIT, July 2026)
      • muratcankoylan/Agent-Skills-for-Context-Engineering - Skills-for-Context-Engineering?style=flat-square) — comprehensive, MIT-licensed collection of Agent Skills for context engineering, multi-agent architectures, and production agent systems — context fundamentals/degradation/compression, memory systems, tool design, harness engineering, self-improvement loops; cited in academic research as foundational work on static skill architecture (17.9k+ stars, actively maintained, 2026)
      • **redhat-et/ripwire** - et/ripwire?style=flat-square) — "The ripgrep of AI context" (Red Hat): zero-dependency C++23 CLI + MCP server that lets coding agents find what they need without reading the whole repo — tree-sitter signatures at 74.7% fewer bytes than bodies, blast-radius queries, tests-to-run suggestions, and post-change verification with quality deltas; every heuristic guess labelled, every context loss published (Apache-2.0, 2.4k+ stars, actively maintained Sept 2026)

  • Frameworks

    • Automatic Prompt Optimization

      • **TextGrad** - group/textgrad?style=flat-square) | Treats LLM feedback as "textual gradients" and backpropagates them to optimize prompts. Published in Nature. |
      • **GEPA** - ai/gepa?style=flat-square) | Reflective Text Evolution — optimizes prompts, code, and agent configs. Claims +6–20 pts over GRPO on 6 tasks with fewer rollouts. |
      • **Hermes Agent Self-Evolution** - agent-self-evolution?style=flat-square) | Evolutionary self-improvement for Hermes Agent — DSPy + GEPA (Genetic-Pareto Prompt Evolution) automatically evolves skills, tool descriptions, system prompts, and code via reflective search over execution traces (understands *why* things fail, not just that they failed); constraint gates (tests, size limits, benchmarks) before a best-variant PR; no GPU training required, ~$2–10 per optimization run (Nous Research, MIT, 5.4k+ stars, Mar 2026) |

    • Eval & Observability

      • **DeepEval** - ai/deepeval?style=flat-square) | Unit testing for LLMs — G-Eval, hallucination, RAG faithfulness, agentic task metrics. |
      • **Langfuse** - square) | Open-source LLM engineering platform — tracing, evals, prompt management, A/B experiments. |
      • **Phoenix** - ai/phoenix?style=flat-square) | Open-source AI observability & evaluation platform (Arize) — OpenTelemetry-native tracing for agents, LLM-as-judge evals, versioned datasets & experiments for prompt regression testing, prompt management with version control and replay, plus an MCP endpoint so Claude Code/Cursor can query traces directly; framework-agnostic (OpenAI Agents SDK, Claude Agent SDK, LangGraph, DSPy, LlamaIndex, Vercel AI SDK); self-hosted, actively maintained (2026) |
      • **Tracely-ai** - ai?style=flat-square) | Trace-native CI/CD for AI agents — grades every production trace as it lands (LLM-as-judge evaluators as trace-table columns), clusters failures into issues, freezes failing runs into hermetic replayable regression cases ($0 replay, no API keys), blocks the PR via CI gate, alerts via Slack/email/webhook; OTLP ingest (MIT, 1.4k+ stars, June 2026, actively maintained) |

    • Eval & Testing

      • **OpenAI Evals** - square) | Open eval framework and benchmark registry — standardizes LLM performance measurement. |
      • **promptfoo** - square) | Test-driven prompt engineering: regression tests, red teaming, model comparison, CI/CD integration. [Acquired by OpenAI (Mar 2026)](https://openai.com/index/openai-to-acquire-promptfoo/) — remains open source. |
      • **Terminal-Bench** - terminal agent benchmark (Stanford/Laude) — compile code, train models, set up servers in Docker-sandboxed environments; the de facto benchmark for agentic coding (2026). |

    • Legacy (2023 era — kept for reference)

      • dair-ai/Prompt-Engineering-Guide - ai/Prompt-Engineering-Guide?style=flat-square) — the canonical entry point. Covers techniques, adversarial prompting, RAG, agents, papers, and notebooks.

    • Low-Code & Workflow Platforms

      • **Dify** - square) | Production-grade RAG and agent workflow platform — visual pipeline builder, multi-model support, plugin architecture. |
      • **Langflow** - ai/langflow?style=flat-square) | Drag-and-drop agent and chain builder — good for rapid prototyping of complex pipelines. |

    • Prompt Programming

      • **DSPy** - square) | Write LM pipelines declaratively, then *compile* — DSPy auto-optimizes prompts and few-shot demonstrations. The strongest engineering-first approach. |
      • **Guidance** - ai/guidance?style=flat-square) | Interleave generation with constraints, regex/CFG, and control flow. Precision output control that goes beyond what prompts alone can achieve. |

    • Red Team & Security

      • **garak** - square) | LLM vulnerability scanner by NVIDIA — red teaming, prompt injection, jailbreak, and leakage detection. |
      • **OpenAI: Prompt Injection Defense**
      • **The Promptware Kill Chain** - stage malware kill chain; 21/36 documented attacks already traverse 4+ stages. Featured at Black Hat 2026. | [PDF](https://arxiv.org/pdf/2601.09625) |
      • **Microsoft Agent Governance Toolkit** - governance-toolkit?style=flat-square) | 7 packages (Python/Rust/TS/Go/.NET) — policy enforcement (<0.1ms), zero-trust agent identity (Ed25519 + SPIFFE), sandboxed execution; covers all OWASP Agentic Top 10; adapters for LangChain/CrewAI/ADK/OpenAI Agents SDK (Apr 2026) |
      • **agent-drift** - drift?style=flat-square) | Stress-test agents for goal drift and system-prompt violations across 6 value dimensions — multi-turn escalation, LLM-as-judge, interactive HTML reports; inspired by ICLR 2026 workshop paper (Apr 2026) |
      • **T3MP3ST** - plinius/T3MP3ST?style=flat-square) | Autonomous red-teaming meta-harness for AI coding agents — recon → exploit → report against authorized targets, multi-agent offensive-security workflows, offline-model support; by elder-plinius (AGPL-3.0, 5.3k+ stars, July 2026) |
      • **OpenAI Codex Security** - security?style=flat-square) | Official OpenAI CLI and TypeScript SDK for finding, validating, and fixing security vulnerabilities — standard/deep scans, diff and working-tree targets, SARIF/CSV/JSON export, CI-native exit codes, pre-commit hooks (Apache-2.0, 8k+ stars, July 2026) |
      • **SkillSpector** - square) | Security scanner for AI agent skills — detects vulnerabilities, malicious patterns, prompt injection, data exfiltration, and supply-chain risks in Claude Code, Codex, and MCP skills before installation (Apache-2.0, 14.7k+ stars, Mar 2026) |

    • Tool Use & Reliability

      • **forge** - square) | Reliability layer for self-hosted LLM tool-calling — guardrails (rescue parsing, retry nudges, response validation), optional workflow constraints (`required_steps`, `prerequisites`, `terminal_tool`), and built-in eval suite. MIT, 2.2k+ stars, Feb 2026 |
      • **reverify** - square) | Hallucination gate for agents — the model proposes claims, deterministic tools check each against ground truth and return VERIFIED/REFUTED with evidence; only what survives counts as fact. Ships as MCP server + CLI, with `reverify rollover` for lossless context handoff across resets. Caught every hallucination on a 71-file binary reverse-engineering benchmark (0 wrong claims accepted). MIT, 978 stars, Aug 2026 |

  • Official Agent Building & Prompt Engineering Guides

  • Official Guides

  • Papers

Programming Languages
Python 43 TypeScript 26 JavaScript 6 Rust 6 Jupyter Notebook 5 Go 2 C++ 1 Shell 1 JetBrains MPS 1 C# 1
Categories
Papers 150 Agent Ecosystem 77 Video courses 62 Table of Contents 48 Official Guides 34 Tools & Libraries 28 Frameworks 25 Prompting libraries & tools (in alphabetical order) 16 Official Agent Building & Prompt Engineering Guides 15 Papers on advanced prompting to improve reasoning 9 Context Engineering 8 System Prompt Leaks 6 Prompting guides 5 Prompt Attack and Prompt Protect 2 Basic Prompt Engineering 1
Sub Categories
Frameworks 50 Harness Engineering 49 Voice & Realtime Agents 32 Reasoning Techniques 16 Agent Evaluation 14 Agent Safety 14 Context & Memory 13 Automatic Optimization 12 Agent Reliability 11 Self-Improving Agents 8 Red Team & Security 8 Prompt Attack & Defense 8 Low-Code & Workflow Platforms 8 Multi-Agent Coordination 8 Instruction Following 7 Agent Skills 7 RAG & Knowledge 7 Multimodal Prompting 7 Foundations 6 Tool Use 6 Medical & Health AI 6 Embodied AI & World Models 6 Surveys 5 Eval & Observability 4 Automatic Prompt Optimization 3 A2A — Agent-to-Agent Protocol 3 Eval & Testing 3 Prompt Programming 2 MCP — Model Context Protocol 2 Tool Use & Reliability 2 Legacy (2023 era — kept for reference) 1
Keywords
llm 32 ai 26 ai-agents 25 python 18 agents 17 openai 15 claude-code 14 prompt-engineering 14 codex 13 chatgpt 12 agent 12 typescript 12 mcp 11 anthropic 11 developer-tools 10 rag 10 langchain 9 cli 7 claude 7 generative-ai 7 framework 7 gemini 7 large-language-models 6 llmops 6 llm-evaluation 6 prompts 5 awesome-list 5 agentic-ai 5 cursor 5 multi-agent 5 context-engineering 5 llms 5 harness 5 mcp-server 4 genai 4 agent-framework 4 workflows 4 coding-agents 4 machine-learning 4 open-source 4 sdk 4 orchestration 3 awesome 3 automation 3 language-model 3 tree-sitter 3 self-hosted 3 javascript 3 nodejs 3 openclaw 3