Featured

Deploy OpenClaw in 60 seconds — 20% off logoDeploy OpenClaw in 60 seconds — 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free
6,000+ web scrapers for your AI agent, start free logo6,000+ web scrapers for your AI agent, start free

Apify gives your agent live web data: 6,000+ prebuilt scrapers and actors, MCP-ready. Sign up free with $5 in usage credits.

Try Apify free
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data — no proxies, no parsers, no maintenance.

Start building free
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it — secured from day one.

Get it set up for you
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free
Reach 48,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here
product-management logo

product-management

ngochuy13/intern-skills-1

Otheropenclawby ngochuy13

Summary

OpenClaw plugin exposing 8 skills.

Install to Claude Code

openclaw plugin add ngochuy13/intern-skills-1

Run in Claude Code. Add the marketplace first with /plugin marketplace add ngochuy13/intern-skills-1 if you haven't already.

README.md

Autonomous Intern — Onboarding Skill Spec

Overview

A self-contained OpenClaw skill that runs on every Intern device. It handles first-time user onboarding (role detection → skill installation) and ongoing skill auto-updates via a cronjob.

---

Architecture

workspace/skills/autonomous-intern-onboarding/
├── SKILL.md          ← Skill definition (trigger rules, flow, cronjob setup)
└── onboarding.json   ← Local state (installed role, version, timestamp)

Remote Skill Repository (GitHub)

https://raw.githubusercontent.com/autonomous-ecm/intern-skills/main/
├── version.txt                             ← Global version string (e.g. "1.0.3")
├── manifest.json                           ← Role → skills mapping
├── skills_zip/
│   ├── generic.zip                         ← Skills installed for ALL users
│   ├── developer.zip                       ← Developer-specific skills
│   ├── marketing.zip                       ← Marketing-specific skills
│   ├── designer.zip                        ← Designer-specific skills
│   ├── sales.zip                           ← Sales-specific skills
│   └── ops.zip                             ← Operations-specific skills
│   └── ...

Each .zip contains N sub-skill folders:

developer.zip
├── github-integration/SKILL.md
├── code-review/SKILL.md
├── terminal-helper/SKILL.md
└── ...

manifest.json Example

{
  "version": "1.0.3",
  "roles": {
    "developer": {
      "label": "Software Developer / Engineer",
      "keywords": ["developer", "engineer", "programmer", "coder", "devops", "sre", "backend", "frontend", "fullstack"],
      "skills_zip": "developer.zip",
      "description": "Code review, GitHub, terminal, debugging tools"
    },
    "marketing": {
      "label": "Marketing / Growth",
      "keywords": ["marketing", "growth", "seo", "content", "social media", "ads", "copywriter"],
      "skills_zip": "marketing.zip",
      "description": "Analytics, email campaigns, SEO, content tools"
    },
    "designer": {
      "label": "Designer / Creative",
      "keywords": ["designer", "ui", "ux", "graphic", "creative", "figma", "photoshop"],
      "skills_zip": "designer.zip",
      "description": "Image tools, design review, asset management"
    },
    "sales": {
      "label": "Sales / Business Development",
      "keywords": ["sales", "business", "account", "crm", "lead"],
      "skills_zip": "sales.zip",
      "description": "CRM integration, email outreach, lead research"
    },
    "ops": {
      "label": "Operations / Admin",
      "keywords": ["operations", "admin", "hr", "finance", "accounting", "office"],
      "skills_zip": "ops.zip",
      "description": "Scheduling, document management, workflow automation"
    }
  },
  "generic_zip": "generic.zip"
}

---

Onboarding Flow

Trigger

  • Every session start → skill checks onboarding.json
  • If onboarding.json doesn't exist or installed_role is empty → trigger onboarding

Step 1: Greet & Ask Role

👋 Welcome to Autonomous Intern! I'm your AI assistant.

To set things up for you, what's your role or job?
Examples: Developer, Marketing, Designer, Sales, Operations

Just type your role or describe what you do!

Step 2: Match Role

  • Parse user response against manifest.json keywords
  • If confident match → proceed to Step 3
  • If ambiguous → present options:
  I found a few possible matches:
  1. Software Developer — Code review, GitHub, terminal tools
  2. DevOps / SRE — Infrastructure, monitoring, deployment tools
  
  Which one fits best? (or describe more)
  • If no match → ask user to clarify, suggest available roles

Step 3: Install Skills

1. Download skills_zip/generic.zip → extract to workspace/skills/ 2. Download skills_zip/{role}.zip → extract to workspace/skills/ 3. Save state to onboarding.json:

   {
     "installed_role": "developer",
     "installed_version": "1.0.3",
     "installed_at": "2026-03-18T06:30:00Z",
     "generic_installed": true,
     "skills": ["github-integration", "code-review", "terminal-helper"]
   }

4. Confirm to user:

   ✅ All set! I've installed these skills for you:
   
   📦 Generic: web-search, calendar, file-manager
   🛠️ Developer: github-integration, code-review, terminal-helper
   
   You can change your role anytime by saying "change my role".

Step 4: Role Change (User-Initiated)

  • User says "change my role" / "switch role" / "I'm now in marketing"
  • Skill detects intent → asks to confirm
  • On confirm:

1. Remove old role-specific skills (keep generic) 2. Download new role zip → extract 3. Update onboarding.json 4. Confirm to user

---

Auto-Update Cronjob

Setup (by SKILL.md)

SKILL.md instructs OpenClaw to create a cronjob on first run:

  • Schedule: Every 6 hours (or configurable)
  • Action: Check remote version.txt
  • Logic:

1. Fetch version.txt from CDN 2. Compare with onboarding.json.installed_version 3. If newer:

  • Download skills_zip/generic.zip → overwrite existing generic skills
  • Download skills_zip/{installed_role}.zip → overwrite existing role skills
  • Update onboarding.json.installed_version
  • Notify user: "🔄 Skills updated to version X.Y.Z"

4. If same → do nothing

Version Strategy

  • version.txt contains a single semver string
  • Any version bump triggers full re-download of applicable zips
  • Skills are overwritten (not merged) to ensure clean state

---

State File: onboarding.json

{
  "installed_role": "developer",
  "installed_version": "1.0.3",
  "installed_at": "2026-03-18T06:30:00Z",
  "updated_at": "2026-03-20T12:00:00Z",
  "generic_installed": true,
  "skills": [
    "web-search",
    "calendar", 
    "file-manager",
    "github-integration",
    "code-review",
    "terminal-helper"
  ],
  "update_cronjob_id": "abc-123-def"
}

---

Edge Cases

| Case | Behavior | |------|----------| | No internet during onboarding | Retry on next session, inform user | | Zip download fails | Retry 3x, then notify user + skip | | User gives gibberish role | Ask again, max 3 attempts, then install generic only | | Multiple users on same device | Last onboarded role wins (single-user device) | | Skill conflicts (same name) | Role-specific overwrites generic if same folder name | | User wants no skills | "skip" keyword → install generic only, mark onboarded |

---

Implementation Notes

  • GitHub raw URL is configurable in SKILL.md (default: https://raw.githubusercontent.com/autonomous-ecm/intern-skills/main/)
  • Zip files are at skills_zip/*.zip in the repo (fetched via {base_url}/skills_zip/{role}.zip)
  • Zip extraction target: the skills directory where this skill is installed (resolved via {baseDir}/../)
  • Skill removal on role change: delete folders listed in onboarding.json.skills that came from old role zip
  • Manifest cache: cache manifest.json locally, refresh on version change
  • No firmware dependency: skills update independently from intern-server OTA

---

TODO

  • [x] Define CDN URL / repo location — using GitHub raw URLs from this repo
  • [x] Create initial role → skills mapping (which skills go in which zip)
  • [x] Write the actual SKILL.md content for the onboarding skill
  • [x] Build generic skills pack
  • [x] Build role-specific skills packs
  • [ ] Test onboarding flow end-to-end on staging device

Related plugins

Browse all →