Personal assistants
assistants and chat-based task agents
NousResearch/hermes-agent NousResearch/hermes-agent is the clear leader — the most-starred, full-featured, self-hostable personal agent with a built-in closed learning loop; pick it if you want an out-of-the-box, persistent, self-improving assistant you can run yourself. Expect many alternatives because teams need different tradeoffs around privacy, size, UIs, vertical workflows, and extensibility that Hermes does not optimize for.
Which one matches your setup?
Answer any of the questions — the shortlist updates as you go. Recommendations come from the capability passports below, nothing else.
Comparison matrix
Axes are extracted from each project's docs by our review pipeline; the maturity score is computed from stars, growth and commit activity — not an opinion. Click a column to sort.
| Runs in | Models | Context | Cost to run | |||||
|---|---|---|---|---|---|---|---|---|
⭐ 241.2k +3638/7d | CLIWeb app | BYOKOpenAIAnthropic | Related files | Approval & allowlist | ●●●●● | Self-hostable | ●●●●● | Your API key or Nous Portal subscription |
⭐ 53.9k +299/7d | CLIWeb app | BYOKOpenAIAnthropicGeminiOllama / local | Diff/File only | None mentioned | ●●●●● | Self-hostable | ●●●●● | Your API key for cloud models; can use local models. |
⭐ 47.7k +220/7d | CLIWeb app | BYOKAnthropicOllama / local | Diff + related files | Basic filters/config | ●●●●● | Fully local possible | ●●●●● | Your API key; optional paid hosting |
⭐ 40.7k +3258/7d | CLI | AnthropicFixed provider | Repo + documents | Reviewer + dedupe | ●●●●● | Third-party cloud | ●●●●● | Requires Claude Code (Anthropic) CLI / third‑party access. |
⭐ 38.7k +918/7d | CLIWeb app | BYOKOpenAIAnthropicGeminiOllama / local | Multi-doc RAG | Basic filters | ●●●●● | Fully local | ●●●●● | Free to self-host; uses your model API key or local LLMs |
⭐ 34.9k +283/7d | CLIWeb appIDE | BYOKOllama / localFixed provider | Diff + related files | Approval gates & guards | ●●●●● | Runs fully local | ●●●●● | Local runtime free; can use your cloud API keys if configured. |
⭐ 5.9k +41/7d | CLI | Fixed provider | Conversation only | No noise controls | ●●●●● | Self-hostable | ●●●●● | Free to download, local runtime. |
⭐ 1.1k +115/7d | CLIWeb app | BYOKOpenAIAnthropicOllama / local | Context pack + recall | Threshold gating | ●●●●● | Fully local possible | ●●●●● | Your API key (per-provider) or run local models |
⭐ 1.9k +35/7d | CLI | OpenAI | Whole-repo analysis | Config & normalization | ●●●●● | Local tool + cloud LLMs | ●●●●● | Your API key (OpenAI recommended) |
⭐ 1.7k +73/7d | CLIWeb app | BYOKOpenAIAnthropicGeminiFixed provider | Related files | Gate + evals | ●●●●● | Self-hostable | ●●●●● | Your API key; provider billing applies |
⭐ 1.2k +9/7d | CLI | Ollama / localFixed provider | Related files & skills | Basic filters & prompts | ●●●●● | Fully local | ●●●●● | Free/local models; optional remote provider costs |
⭐ 1.5k | Web app | BYOKOpenAIAnthropicOllama / local | Session-only context | Dedup & filters | ●●●●● | Self-hostable | ●●●●● | Your API key; optional NagaModel gateway |
Capability profiles
Six axes, 0–5 each. The shape tells you the strategy: a wide hexagon is a generalist, a spike is a specialist. Showing the 8 most established — the rest are in the full catalog.
A self-hostable personal agent with a built-in closed learning loop that creates, persists, and self-improves skills from its own experience.
A native, local-first agent with desktop app, CLI, and an extensible plugin ecosystem (Model Context Protocol) plus broad provider support.
An ultra-lightweight, self-hosted personal agent runtime that pairs a small, readable core with built-in WebUI, persistent memory, automation, and flexible model routing for long-running workflows.
Local-first, end-to-end job-application framework that scrapes portals, tailors LaTeX CVs/cover letters, runs a drafter–reviewer agent pipeline, and archives applications for repeatable, owned job hunting.
A self-hostable, multi-agent tutoring platform with built-in RAG, three-layer persistent memory and a pluginable Skills ecosystem for lifelong personalized learning.
Local-first Agent OS with built-in local models, three-layer self-evolving memory (ReMe), and kernel-level sandboxing for secure, extensible personal assistants.
A personality-first, lifelike QQ chatbot that emphasizes human-style conversation and long-term user modeling while offering an extensible plugin/event system.
Three-layer memory (learner profile, native wiki, cross-session recall) plus a proactive scheduler and a workspace Practice Lab — an education-optimized, evidence-driven agent design layered on the Pi SDK.
All repositories (25)
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with…
Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, M…
The job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings…
DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.
Your Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple cha…
MaiSaka, an LLM-based intelligent agent, is a digital lifeform devoted to understanding you and interacting in…
Waku Waku! Waku Agent is a local-first AI agent harness you actually own, including loop, memory, eval, all in…
A simple yet powerful agent framework for personal assistants, designed to enable intelligent interaction, mul…
PhoneClaw turns phones into local AI agent runtimes with on-device models, native mobile Skills, LiveLand, and…
An open-source personal learning agent with three-layer memory (learner profile / wiki knowledge base / cross-…
Agent-led job application workflow with a local progress-tracking dashboard — works with any AI coding agent t…
Smart Promise Tracker AI 2026 – Commit Memory Engine
Panel-based personal assistant with a plugin architecture for productivity workflows. AI agents share a worksp…
Lightweight local swappable semantic search & ask CLI agent for Gmail
A travel planning Agent powered by LangGraph, Amap POI, route clustering and LLM-based itinerary generation.
Open source CLI agent that generates complete lesson bundles in your teaching voice. 48+ tools, zero-touch 9-f…
Local-first AI desktop assistant that scrapes job boards, matches roles to your resume, and auto-generates res…
Assessment-driven, local-first learning agent built on an observable Agent Runtime and eval harness—grounded i…
🪼 A personal AI assistant that lives in Telegram — every topic is its own isolated Claude Code agent with lay…
Ask your video library anything — an AI agent that watches, reasons in a tool-use loop, and answers with playa…
Agente autónomo para LALIGA Fantasy: lee el juego, decide y actúa (pujas, ventas, clausulazos, alineación). Py…
Hosted alternatives
If running your own reviewer is more ops than you want, these managed services cover the same job.
A hosted personal-assistant builder: wire up email, calendar and messaging agents from templates and let Lindy run them, instead of keeping a bot process alive yourself.
Try Lindy →A managed autonomous assistant for research, scheduling and recurring tasks — the same everyday-agent idea, run in the cloud with no self-hosting.
Try Cognosys →