Official agent skill

Local AI Agents

by microsoft in microsoft/ai-agents-for-beginners

Bumuo ng mga local-first AI agents na tumatakbo nang buong-buo sa isang developer workstation gamit ang Microsoft Foundry Local at Qwen function-calling models.

OfficialMITAuto-check passedAI & LLM Engineering

Install Local AI Agents

skills CLI
$ npx skills add microsoft/ai-agents-for-beginners --skill local-ai-agents -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install microsoft/ai-agents-for-beginners local-ai-agents --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/microsoft/ai-agents-for-beginners.git skills-src && mkdir -p .claude/skills && cp -r skills-src/translations/tl/.agents/skills/local-ai-agents .claude/skills/local-ai-agents && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
local-ai-agents
GitHub stars
77k
Token cost
~1.7k tokens
SKILL.md length
767 words
Files
1
Skills in repo
123
Repo updated
First seen
Licence
MIT

At a glance

Bumuo ng mga local-first AI agents na tumatakbo nang buong-buo sa isang developer workstation gamit ang Microsoft Foundry Local at Qwen function-calling models.

  • Tasks that involve Structured output and tool calling
  • SKILL.md covers Mga Trigger, Pangunahing mental na modelo, Bakit ang mga partikular na… and Mga kailangan para sa setup, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Tasks that involve MCP servers

What it does

Local AI Agents is an agent skill from microsoft/ai-agents-for-beginners, published by the product's own GitHub organization. Bumuo ng mga local-first AI agents na tumatakbo nang buong-buo sa isang developer workstation gamit ang Microsoft Foundry Local at Qwen function-calling models. Saklaw nito ang Small Language Models (SLMs), ang OpenAI-compatible na lokal na endpoint, sandboxed local tools, lokal na RAG gamit ang Chroma, lokal na MCP servers, hybrid cloud/local routing, at ang privacy/cost/offline trade-offs. Batay sa Lesson 17 ng AI Agents for Beginners. GAMITIN PARA SA: pagpapatakbo ng agent nang lokal, offline agent, on-device…

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering, covering Structured output and tool calling, MCP servers and GPU and accelerator computing. It works with Model Context Protocol, Qwen, OpenAI and Chroma. The repository describes itself as: 18 Lessons to Get Started Building AI Agents. The licence is MIT.

When your agent uses it

  • Tasks that involve Structured output and tool calling
  • Tasks that involve MCP servers
  • Tasks that involve GPU and accelerator computing

Example prompts

  • “/local-ai-agents”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 25b7985. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are bash and python).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • github.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Local AI Agents loads about 1.7k tokens when it runs. Until then it costs about 259 tokens; SKILL.md has 767 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~259
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from microsoft/ai-agents-for-beginners at commit 25b7985, republished under its MIT licence (© microsoft). 767 words, ~1,701 tokens.

Download SKILL.mdSave it as .claude/skills/local-ai-agents/SKILL.md (or your agent's skills folder).
name
local-ai-agents
description
Bumuo ng mga local-first AI agents na tumatakbo nang buong-buo sa isang developer workstation gamit ang Microsoft Foundry Local at Qwen function-calling models. Saklaw nito ang Small Language Models (SLMs), ang OpenAI-compatible na lokal na endpoint, sandboxed local tools, lokal na RAG gamit ang Chroma, lokal na MCP servers, hybrid cloud/local routing, at ang privacy/cost/offline trade-offs. Batay sa Lesson 17 ng AI Agents for Beginners. GAMITIN PARA SA: pagpapatakbo ng agent nang lokal, offline agent, on-device agent, Foundry Local, Qwen function calling, local tool calling, lokal na RAG, Chroma vector database, lokal na MCP server, privacy-preserving agent, hybrid local at cloud agent, small language model agent, engineering assistant sa aking makina. HUWAG GAMITIN PARA SA: pag-deploy ng mga agents sa cloud nang malakihan (gamitin ang deploying-scalable-agents / Lesson 16), paggawa ng iyong unang agent concept (Lesson 01), Foundry (cloud) hosted agents, GPU cluster / server-side inference provisioning.
license
MIT

Paglikha ng Lokal na AI Agents gamit ang Foundry Local at Qwen

Kasama sa kasanayan para sa Lesson 17 – Paglikha ng Lokal na AI Agents. Gamitin ito upang tulungan ang isang nag-aaral na bumuo ng isang ahente na nagrerason, tumatawag ng mga tool, at naghahanap ng dokumentasyon nang buong mag-isa sa kanilang sariling makina — walang cloud inference. Ibatay ang bawat rekomendasyon sa nilalaman ng leksyon at ang tumatakbong notebook.

Mga Trigger

Isaaktibo ang kasanayang ito kapag nais ng nag-aaral na:

  • Patakbuhin ang isang ahente nangang buong-on-device para sa privacy, gastos, o offline na dahilan.
  • Maglingkod ng isang modelo nang lokal gamit ang Foundry Local at kumonekta sa pamamagitan ng compatible endpoint ng OpenAI.
  • Gumamit ng Qwen function-calling na modelo upang magpatakbo ng maaasahang lokal na pagtawag sa mga tool.
  • Magdagdag ng local RAG (Chroma) o isang local MCP server.
  • Magdisenyo ng isang hybrid na lokal/cloud routing strategy.

Pangunahing mental na modelo

Ang isang SLM ay nagtutuko ng lawak para sa privacy, gastos, at offline na operasyon. Ang panalong estratehiya: hayaan ang SLM ang mag-orchestrate at hayaan ang mga tool ang gumawa ng mabibigat na gawain. Ang modelo ay hindi kailangang maalam sa codebase — kailangan lang nitong malaman kung kailan tatawagin read_file at search_docs. Ito ay nakatuon sa lakas ng SLM (mga limitadong desisyon tulad ng pagpili ng tool) at inilalayo sa kahinaan nito (malawak na kaalaman, mahaba at multi-hop na pangangatwiran).

Bakit ang mga partikular na bahagi na ito

  • Ang Foundry Local ay naglalantad ng isang OpenAI-compatible HTTP endpoint, kaya ang code ng cloud agent ay naililipat lamang sa pamamagitan ng pagpapalit ng base_url (at paggamit ng lokal na placeholder na API key). Awtomatikong pinipili rin nito ang pinakamahusay na build (CPU/GPU/NPU) para sa makina.
  • Ang mga modelo ng Qwen ay sinanay para sa function calling at naglalabas ng maayos na porma ng mga pagtawag sa tool nang pare-pareho — ito ang nagpapalit ng isang lokal na chat na modelo sa isang lokal na agent.
  • Ang Chroma ay tumatakbo sa proseso at nag-iimbak ng mga vectors sa disk, kaya ang buong RAG pipeline (embed → store → retrieve → reason) ay nananatiling lokal.
  • Ang MCP ay isang transport, hindi isang cloud service: ang MCP server ay maaaring tumakbo nang lokal sa stdio.

Mga kailangan para sa setup

bash
foundry model run qwen2.5-7b-instruct
foundry service status
python
from foundry_local import FoundryLocalManager
from openai import OpenAI

manager = FoundryLocalManager("qwen2.5-7b-instruct")
client = OpenAI(base_url=manager.endpoint, api_key=manager.api_key)  # lokal na placeholder

~8 GB RAM ang isang realistiko at minimum; nakakatulong ang GPU/NPU pero hindi kinakailangan.

Mga susi na pattern na dapat kopyahin

Ituro ang nag-aaral sa notebook 17-local-agent-foundry-local.ipynb:

  • Sandboxed tools: bawat file tool ay nagtutukoy ng mga path at tinatanggihan ang anuman sa labas ng isang solong ugat na proyekto — kahit lokal, ang tool ay tumatakbo gamit ang mga pahintulot ng user.
  • Tool-calling loop: irehistro ang mga tool gamit ang OpenAI tools schema, patakbuhin ang hinihiling na mga tool nang lokal, ibalik ang mga resulta, ulitin hanggang sa makakuha ng pinal na sagot.
  • Local RAG: mag-upsert ng mga docs sa isang Chroma collection; search_docs ay nagbabalik ng mga nangungunang top-k chunks.
  • Local MCP: kumonekta sa isang lokal na server gamit ang stdio; itakda ito sa isang proyekto na direktoryo at patunayan ang mga output nito.
Show full SKILL.md (268 more words)Show less

Hybrid na routing (lokal bilang isa sa mga modelo)

SitwasyonSaan ito tumatakbo
Sensitibong datos / offlineLokal na SLM
Simple, limitadong gawainLokal na SLM (murang, mabilis)
Mahirap na multi-hop na pangangatwiran sa hindi sensitibong dataCloud model
Cloud outageLokal na SLM (maayos na pagbaba ng kalidad)

Ito ay sumasalamin sa ideya ng model-routing mula sa Lesson 16, gamit ang workstation bilang isa sa mga ruta. Piliin ang mga disenyo na bumabalik sa lokal para ang ahente ay bumaba sa kalidad sa halip na tuluyang mabigo.

Mga guardrail para sa katulong

  • Panatilihin ang bawat operasyon sa file/tool na saklaw lamang ng isang sandboxed na direktoryo ng proyekto.
  • Huwag magpadala ng code o datos sa cloud kapag ang layunin ng nag-aaral ay privacy/offline — panatilihin ang buong pipeline nang lokal.
  • Magtakda ng realistiko na mga inaasahan para sa kalidad ng SLM; umasa sa mga tool at RAG sa halip na ang naalala ng modelo na kaalaman.
  • Tandaan na ang Lesson 17 ay walang Foundry Responses endpoint, kaya ang cloud smoke-test action ay hindi nalalapat — patunayan ito sa pamamagitan ng pagpapatakbo ng notebook nang lokal.

<!-- CO-OP TRANSLATOR DISCLAIMER START -->

Pagtatanggi: Ang dokumentong ito ay isinalin gamit ang serbisyo ng AI translation na Co-op Translator. Bagama't nagsusumikap kami para sa katumpakan, pakatandaan na ang awtomatikong pagsasalin ay maaaring maglaman ng mga pagkakamali o hindi pagkakatugma. Ang orihinal na dokumento sa orihinal nitong wika ang dapat ituring na pangunahing sanggunian. Para sa mahahalagang impormasyon, inirerekomenda ang propesyonal na pagsasalin ng tao. Hindi kami mananagot sa anumang maling pagkakaintindi o maling interpretasyon na nagmula sa paggamit ng pagsasaling ito.

<!-- CO-OP TRANSLATOR DISCLAIMER END -->

© microsoft, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in translations/tl/.agents/skills/local-ai-agents of microsoft/ai-agents-for-beginners.

Open the folder on GitHubat commit 25b7985

Compare with similar skills

Local AI Agents next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Local AI Agents compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Local AI Agents this skillmicrosoft/ai-agents-for-beginners77k—~1.7kAutomated safety check: PassMIT
Neurolink Guidejuspay/neurolink144—~1.4kAutomated safety check: PassMIT
Tool Designagentailor/fullstack-langgraph-nextjs-agent132—~3.2kAutomated safety check: PassMIT
Qianwenai Wikichujianyun/skills740—~717Automated safety check: PassCustom licence
Agent Tool Builderomer-metin/skills-for-antigravity162—~705Automated safety check: PassApache-2.0
Spring AI Integrationrrezartprebreza/spring-boot-skills298—~2.1kAutomated safety check: PassMIT

Similar skills

  • Neurolink Guide

    juspay/neurolink

    Guide for using the NeuroLink SDK and CLI. An agent skill from juspay/neurolink.

    144 GitHub stars~1.4k tokensUpdated today
    AI & LLM EngineeringAuto-check passed
  • Tool Design

    agentailor/fullstack-langgraph-nextjs-agent

    Design and verify tools that AI agents can actually use — for any framework or language (MCP servers, LangChain/LangGraph, function-calling, raw JSON schema; TypeScript, Python, or otherwise).

    132 GitHub stars~3.2k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed
  • Qianwenai Wiki

    chujianyun/skills

    千问AI平台(Qianwen AI Platform / DashScope)官方文档离线知识库,用于检索并回答模型选择、API Key、OpenAI 兼容接口、DashScope SDK、文本与多模态生成、图像/视频/语音、Realtime API、Embedding、Reranking、Function Calling、MCP、批量调用、计费、Token Plan、API/SDK/CLI…

    740 GitHub stars~717 tokensUpdated today
    AI & LLM EngineeringAuto-check passed
  • Agent Tool Builder

    omer-metin/skills-for-antigravity

    Tools are how AI agents interact with the world. An agent skill from omer-metin/skills-for-antigravity.

    162 GitHub stars~705 tokensUpdated 8 mo ago
    AI & LLM EngineeringAuto-check passed
  • Spring AI Integration

    rrezartprebreza/spring-boot-skills

    A skill your agent uses when integrating LLMs, chat clients, embeddings, RAG pipelines, or AI agents into Spring Boot.

    298 GitHub stars~2.1k tokensUpdated 18 days ago
    AI & LLM EngineeringAuto-check passed
  • Spring AI Integration

    rrezartprebreza/spring-boot-skills

    A skill your agent uses when integrating LLMs, chat clients, embeddings, RAG pipelines, or AI agents into Spring Boot.

    298 GitHub stars~2.5k tokensUpdated 18 days ago
    AI & LLM EngineeringAuto-check passed

More from microsoft/ai-agents-for-beginners

All 123 skills in this repo
  • Jupyter Notebook

    microsoft/ai-agents-for-beginners

    Official

    A skill your agent uses when the user asks to create, scaffold, or edit Jupyter notebooks (.ipynb) for experiments, explorations, or tutorials; prefer the bundled templates and run the helper script…

    77k GitHub starsUsed in 8 repos~1k tokens
    Auto-check passed
  • Azure Openai To Responses

    microsoft/ai-agents-for-beginners

    Official

    Migrate Python apps from Azure OpenAI Chat Completions to the Responses API.

    77k GitHub stars~6k tokensUpdated 20 days ago
    Auto-check: notes
  • Azure Openai To Responses

    microsoft/ai-agents-for-beginners

    Official

    Shift Python apps dem from Azure OpenAI Chat Completions go Responses API.

    77k GitHub stars~6k tokensUpdated 20 days ago
    Auto-check: notes
  • Jupyter Notebook

    microsoft/ai-agents-for-beginners

    Official

    Kasuta, kui kasutaja palub luua, üles ehitada või redigeerida Jupyteri märkmikke (.ipynb) katsetuste, uurimiste või juhendite jaoks; eelista kaasasolevaid malle ja käivita abiskript newnotebook.py…

    77k GitHub stars~1.2k tokensUpdated 20 days ago
    Auto-check passed
  • Jupyter Notebook

    microsoft/ai-agents-for-beginners

    Official

    Käytetään, kun käyttäjä pyytää luomaan, alustamaan tai muokkaamaan Jupyter-muistikirjoja (.ipynb) kokeita, tutkimuksia tai opetusohjelmia varten; käytä mieluummin mukana olevia mallipohjia ja…

    77k GitHub stars~1.3k tokensUpdated 20 days ago
    Auto-check passed
  • Jupyter Notebook

    microsoft/ai-agents-for-beginners

    Official

    À utiliser lorsque l'utilisateur demande de créer, structurer ou modifier des notebooks Jupyter (.ipynb) pour des expériences, explorations ou tutoriels ; privilégiez les modèles fournis et exécutez…

    77k GitHub stars~1.4k tokensUpdated 20 days ago
    Auto-check passed

Questions about Local AI Agents

What does Local AI Agents do?

Bumuo ng mga local-first AI agents na tumatakbo nang buong-buo sa isang developer workstation gamit ang Microsoft Foundry Local at Qwen function-calling models. Local AI Agents is an agent skill from microsoft/ai-agents-for-beginners, published by the product's own GitHub organization. Bumuo ng mga local-first AI agents na tumatakbo nang buong-buo sa isang developer workstation gamit ang Microsoft Foundry Local at Qwen function-calling models.

When should I use Local AI Agents?

Local AI Agents fits situations like: tasks that involve Structured output and tool calling; tasks that involve MCP servers; tasks that involve GPU and accelerator computing.

How do I install Local AI Agents in Claude Code?

Run `npx skills add microsoft/ai-agents-for-beginners --skill local-ai-agents -a claude-code`. Or copy the skill folder (translations/tl/.agents/skills/local-ai-agents in microsoft/ai-agents-for-beginners) into .claude/skills/local-ai-agents in your project. Claude Code loads it when a task matches its description.

How do I install Local AI Agents in Codex?

Run `npx skills add microsoft/ai-agents-for-beginners --skill local-ai-agents -a codex`. Or copy the skill folder (translations/tl/.agents/skills/local-ai-agents in microsoft/ai-agents-for-beginners) into .agents/skills/local-ai-agents in your project. Codex loads it when a task matches its description.

Can I use Local AI Agents in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add microsoft/ai-agents-for-beginners --skill local-ai-agents -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/local-ai-agents, .gemini/skills/local-ai-agents, .github/skills/local-ai-agents and .opencode/skills/local-ai-agents in your project.

What does Local AI Agents need to run?

SKILL.md names no scripts, command-line tools or credentials: Local AI Agents is instructions for the agent only. Our summary lists: Python 3.

Does Local AI Agents access the network?

SKILL.md names 1 domain. As links in the text: github.com. This is read from the text; nothing was executed.

Is Local AI Agents safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Local AI Agents use?

Local AI Agents is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Local AI Agents use?

About 1.7k tokens (SKILL.md is roughly 6.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Local AI Agents?

Skills that share tags, products or a category with Local AI Agents: Neurolink Guide (juspay/neurolink, 144 stars), Tool Design (agentailor/fullstack-langgraph-nextjs-agent, 132 stars), Qianwenai Wiki (chujianyun/skills, 740 stars) and Agent Tool Builder (omer-metin/skills-for-antigravity, 162 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Local AI Agents?

microsoft (a GitHub organization, an official publisher) maintains it in microsoft/ai-agents-for-beginners, which has 76,686 GitHub stars. The repository holds 123 skills in this directory. The repository was last updated on September 19, 2026.

Source: microsoft/ai-agents-for-beginners on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.