In this tutorial, we implement a SkillNet use case as a practical framework for discovering, installing, inspecting, evaluating, and organizing reusable AI skills. We start by setting up a robust SkillNet client with SDK and REST fallback support, then compare keyword search with semantic search to understand how skills can be found for different task requirements. From there, we install curated skills from GitHub, inspect their metadata, apply a quality gate across key evaluation dimensions, and visualize relationships between skills as a graph. Finally, we build a skill-augmented agent planner that breaks a complex goal into subtasks, discovers relevant skills, filters them, and assembles an execution pipeline.

Copy CodeCopiedUse a different Browser

import sys, subprocess def _pip(*pkgs): subprocess.run([sys.executable, "-m", "pip", "install", "-q", *pkgs], check=False) print("Installing dependencies (skillnet-ai, networkx, matplotlib, requests)...") _pip("skillnet-ai", "networkx", "matplotlib", "requests") import os, re, json, textwrap, pathlib, traceback import requests API_KEY = os.environ.get("API_KEY", "") BASE_URL = os.environ.get("BASE_URL", "https://api.openai.com/v1") MODEL = os.environ.get("SKILLNET_MODEL", "gpt-4o") GITHUB_TOKEN = os.environ.get("GITHUB_TOKEN", "") GITHUB_MIRROR = os.environ.get("GITHUB_MIRROR", "") if not API_KEY: try: from google.colab import userdata API_KEY = userdata.get("API_KEY") or "" except Exception: pass REST_BASE = "http://api-skillnet.openkg.cn/v1" WORKDIR = pathlib.Path("./skillnet_demo"); WORKDIR.mkdir(exist_ok=True) SKILLS_DIR = WORKDIR / "my_skills"; SKILLS_DIR.mkdir(exist_ok=True) def banner(title): line = "=" * 78 print(f"\n{line}\n {title}\n{line}")

We install the required dependencies and prepare the basic environment for the SkillNet tutorial. We configure API keys, model settings, GitHub options, and working directories to ensure the rest of the workflow runs smoothly. We also define a reusable banner function to keep the tutorial output organized and readable.

Copy CodeCopiedUse a different Browser

banner("1) Initialize SkillNet client (SDK with REST fallback)") USE_SDK = False client = None try: from skillnet_ai import SkillNetClient client = SkillNetClient( api_key=API_KEY or None, base_url=BASE_URL, github_token=GITHUB_TOKEN or None, ) USE_SDK = True print("SDK loaded: skillnet_ai.SkillNetClient") except Exception as e: print(f"SDK unavailable ({e!r}); using REST fallback for search/download.") def _norm(item): if isinstance(item, dict): g = item.get else: g = lambda k, d=None: getattr(item, k, d) return { "skill_name": g("skill_name") or g("name") or "?", "skill_description": g("skill_description") or g("description") or "", "author": g("author") or "", "stars": g("stars") or 0, "skill_url": g("skill_url") or g("url") or "", "category": g("category") or "", } def search(q, mode="keyword", limit=5, min_stars=0, sort_by="stars", threshold=0.8): if USE_SDK: try: kw = dict(q=q, limit=limit, mode=mode) if mode == "keyword": kw.update(min_stars=min_stars, sort_by=sort_by) else: kw.update(threshold=threshold) res = client.search(**kw) return [_norm(x) for x in (res or [])] except Exception as e: print(f" [SDK search failed -> REST] {e!r}") params = {"q": q, "mode": mode, "limit": limit} if mode == "keyword": params.update(min_stars=min_stars, sort_by=sort_by) else: params.update(threshold=threshold) try: r = requests.get(f"{REST_BASE}/search", params=params, timeout=30) r.raise_for_status() return [_norm(x) for x in r.json().get("data", [])] except Exception as e: print(f" [REST search failed] {e!r}") return [] def show_results(results, title=""): if title: print(f"\n-- {title} --") if not results: print(" (no results / endpoint unreachable)") return for i, s in enumerate(results, 1): desc = textwrap.shorten(s["skill_description"], 70, placeholder="...") print(f" {i}. {s['skill_name']: {path}") print(f"\nInstalled {len(installed)} skill(s).")