whatsapp-triage — autonomous AI employee workflow whatsapp-triage, Personal-AI-Employee, community, autonomous AI employee workflow, ide skills, whatsapp-triage install and configuration, AI agent workflow optimization, whatsapp-triage for CEO productivity, Claude Code, Cursor, Windsurf

v1.0.0
GitHub

About this Skill

Ideal for Task Management Agents requiring automated workflow triage and CEO briefing capabilities via WhatsApp. whatsapp-triage is a local-first autonomous AI employee that detects intent, assigns priority, and creates action plans based on WhatsApp messages.

Features

Monitors emails, finances, and tasks via MCP agents
Detects intent and assigns priority based on urgent keywords
Creates action plans and routes outgoing actions to /Pending_Approval/
Delivers weekly CEO briefings with real productivity gains
Requires human-in-the-loop (HITL) approval for all WhatsApp sends
Processes WhatsApp messages stored in WHATSAPP_*.md files

# Core Topics

muhammadwaheedairi muhammadwaheedairi
[0]
[1]
Updated: 2/26/2026

Agent Capability Analysis

The whatsapp-triage skill by muhammadwaheedairi is an open-source community AI agent skill for Claude Code and other IDE workflows, helping agents execute tasks with better context, repeatability, and domain-specific guidance. Optimized for autonomous AI employee workflow, whatsapp-triage install and configuration, AI agent workflow optimization.

Ideal Agent Persona

Ideal for Task Management Agents requiring automated workflow triage and CEO briefing capabilities via WhatsApp.

Core Value

Empowers agents to detect intent, assign priority, and create action plans from WhatsApp messages stored in `.md` files, routing outgoing actions to `/Pending_Approval/` for human-in-the-loop approval, utilizing file monitoring and intent classification patterns from `references/intents.md`.

Capabilities Granted for whatsapp-triage

Automating workflow triage for CEOs
Generating weekly briefings from WhatsApp messages
Routing urgent actions to pending approval folders

! Prerequisites & Limits

  • Requires file system access to `/Needs_Action/` and `/Pending_Approval/` directories
  • All WhatsApp sends require human-in-the-loop approval
  • Limited to processing `.md` files
Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

whatsapp-triage

Install whatsapp-triage, an AI agent skill for AI agent workflows and automation. Works with Claude Code, Cursor, and Windsurf with one-command setup.

SKILL.md
Readonly

WhatsApp Triage

Process WHATSAPP_*.md files from /Needs_Action/. Detect intent, assign priority, create action plans, and route all outgoing actions to /Pending_Approval/. All WhatsApp sends require HITL approval — never act directly.

Workflow

  1. Read the WHATSAPP_*.md file from /Needs_Action/
  2. Detect intent — see references/intents.md for classification patterns
  3. Assign priority based on urgent keywords (already set by watcher):
    • high: urgent, asap, invoice, payment, emergency, important
    • medium: help, problem, price, quote, service inquiry
    • low: general conversation, thank you messages
  4. Create action plan in /Plans/PLAN_WHATSAPP_{timestamp}.md using templates from references/action-templates.md
  5. Route sensitive messages to /Pending_Approval/ for human review:
    • Payment/invoice messages (always)
    • Urgent/emergency messages (always)
    • All reply drafts (always — never send without approval)
  6. Log action to /Vault/Logs/YYYY-MM-DD.json
  7. Move source WHATSAPP_*.md to /Done/ when plan is complete
  8. Output <promise>TASK_COMPLETE</promise>

Urgent Keywords (from watcher)

The WhatsApp watcher automatically flags messages containing these keywords and marks them as priority: high:

  • urgent
  • asap
  • invoice
  • payment
  • emergency
  • important

These are defined in watchers/whatsapp_watcher.py:18:

python
1URGENT_KEYWORDS = ["urgent", "asap", "invoice", "payment", "emergency", "important"]

Action Decision Tree

WHATSAPP_*.md arrives in /Needs_Action/
    │
    ├─ Contains "payment" or "invoice"? → Generate invoice plan → /Pending_Approval/ (high priority)
    │
    ├─ Contains "urgent" or "emergency"? → Draft urgent response → /Pending_Approval/ (high priority)
    │
    ├─ Contains "price" or "quote"? → Draft pricing response → /Pending_Approval/ (medium priority)
    │
    ├─ Service inquiry? → Draft informative response → /Pending_Approval/ (medium priority)
    │
    └─ General message? → Determine if response needed → /Pending_Approval/ or archive

WHATSAPP_*.md File Format

Files created by the watcher have this structure:

markdown
1--- 2type: whatsapp 3source: whatsapp_web 4received: {iso_timestamp} 5priority: high 6status: pending 7msg_id: {hash_id} 8--- 9 10# WhatsApp Urgent Message 11 12**Received:** {date_time} 13**Priority:** High 14 15## Message Preview 16{message_text_with_context} 17 18## Matched Keyword In 19{line_that_triggered_urgent_flag} 20 21## Suggested Actions 22- [ ] Reply to sender 23- [ ] Generate invoice if requested 24- [ ] Escalate if payment-related

Plan.md Template

markdown
1--- 2type: plan 3created: {iso_timestamp} 4source_file: WHATSAPP_{timestamp}.md 5source_type: whatsapp 6priority: {high|medium|low} 7status: pending 8--- 9 10# Plan: WhatsApp — {detected_intent} 11 12## Context 13**Received:** {date} 14**Priority:** {priority} 15**Matched Keywords:** {keywords} 16 17## Objective 18{one-line goal — e.g., "Draft reply to urgent client inquiry about project status"} 19 20## Steps 21- [ ] {step 1} 22- [ ] {step 2} 23- [ ] {step 3} 24... 25 26## Completion Rule 27Move this file and source WHATSAPP_*.md to /Done/ when {completion condition}.

Security Rules

  • NEVER send WhatsApp messages directly — always write drafts to /Pending_Approval/
  • ALWAYS route payment/invoice messages to /Pending_Approval/ for human review
  • ALWAYS route urgent messages to /Pending_Approval/ for immediate human attention
  • ALWAYS log every action to /Vault/Logs/YYYY-MM-DD.json
  • NEVER delete WHATSAPP_*.md files — move to /Done/ instead
  • NEVER confirm payment receipt without explicit human verification

Approval File Format

When routing replies to /Pending_Approval/, use this format:

markdown
1--- 2type: whatsapp_reply 3action: send_whatsapp 4to: {contact_name} 5phone: {phone_number} 6source_file: WHATSAPP_{timestamp}.md 7priority: {high|medium|low} 8created: {iso_timestamp} 9status: awaiting_approval 10--- 11 12# WhatsApp Reply: {contact_name} 13 14**To:** {contact_name} ({phone_number}) 15**Priority:** {priority} 16 17## Message to Send 18 19{reply_text} 20 21--- 22 23## Original Message Context 24**Received:** {date} 25**Matched Keywords:** {keywords} 26**Content:** {original message preview} 27 28--- 29 30## To Approve 31Move this file to /Approved folder. 32 33## To Reject 34Move this file to /Rejected folder or delete. 35 36--- 37*Awaiting human approval before sending*

Invoice Approval Format

For invoice generation requests:

markdown
1--- 2type: invoice 3action: generate_invoice 4client: {client_name} 5source_file: WHATSAPP_{timestamp}.md 6priority: high 7created: {iso_timestamp} 8status: awaiting_approval 9--- 10 11# Invoice Request: {client_name} 12 13**Client:** {client_name} 14**Service/Period:** {description} 15**Amount:** ${amount} 16 17## Invoice Details 18 19**To:** {client_name} 20**Email:** {client_email} 21**Amount:** ${amount} 22**Due Date:** {date} 23**Bank Details:** {from Company_Handbook.md} 24 25## Description 26{service description} 27 28--- 29 30## Original WhatsApp Request 31**Received:** {date} 32**Content:** {original message} 33 34--- 35 36## To Approve 37Move this file to /Approved folder to generate and send invoice. 38 39## To Reject 40Move this file to /Rejected folder. 41 42--- 43*Awaiting human approval before generating invoice*

Intent Detection Guide

See references/intents.md for detailed intent classification patterns:

  1. Invoice Request (high priority, always approve)
  2. Payment Confirmation/Query (high priority, always approve)
  3. Urgent Query/Problem (high priority)
  4. Pricing Query (medium priority)
  5. Service Inquiry (medium priority)
  6. Appointment/Scheduling (medium priority)
  7. General/Informational (low priority)

Action Templates

See references/action-templates.md for complete plan templates:

  • Reply Plan Template
  • Invoice Generation Plan Template
  • Urgent Escalation Template
  • Payment/Financial Escalation Template
  • Pricing Query Plan Template
  • General Query Plan Template

Example Scenarios

Scenario 1: Invoice request

Input: WHATSAPP_20260305_120000.md
Content: "Hi, can you send me the invoice for this month?"
Matched Keywords: "invoice"
Priority: high
Action: Create invoice generation plan
Output: PLAN_WHATSAPP_20260305_120000_invoice.md in /Plans/
Route: INVOICE_{client}_{timestamp}.md to /Pending_Approval/

Scenario 2: Urgent problem

Input: WHATSAPP_20260305_140000.md
Content: "Urgent! The website is down, clients can't access it"
Matched Keywords: "urgent"
Priority: high
Action: Draft urgent response plan
Output: PLAN_WHATSAPP_20260305_140000_urgent.md in /Plans/
Route: WHATSAPP_REPLY_{timestamp}.md to /Pending_Approval/

Scenario 3: Pricing inquiry

Input: WHATSAPP_20260305_160000.md
Content: "How much do you charge for web design?"
Matched Keywords: "how much", "charge"
Priority: medium
Action: Draft pricing response using /Vault/Accounting/Rates.md
Output: PLAN_WHATSAPP_20260305_160000_pricing.md in /Plans/
Route: WHATSAPP_REPLY_{timestamp}.md to /Pending_Approval/

Scenario 4: Payment confirmation

Input: WHATSAPP_20260305_180000.md
Content: "I just transferred the payment, transaction ID: 12345"
Matched Keywords: "payment", "transferred"
Priority: high
Action: Escalate to human for verification (NEVER auto-confirm)
Output: PLAN_WHATSAPP_20260305_180000_payment.md in /Plans/
Route: Copy WHATSAPP_*.md to /Pending_Approval/ immediately

Scenario 5: Thank you message

Input: WHATSAPP_20260305_200000.md
Content: "Thanks for the quick response!"
Matched Keywords: "thanks"
Priority: low
Action: Archive (no response needed)
Output: Move WHATSAPP_*.md to /Done/, log action

Logging Format

Every WhatsApp action must be logged to /Vault/Logs/YYYY-MM-DD.json:

Triage log:

json
1{ 2 "timestamp": "2026-03-05T14:30:00Z", 3 "action_type": "whatsapp_triage", 4 "source_file": "WHATSAPP_20260305_143000.md", 5 "intent": "invoice_request", 6 "priority": "high", 7 "action_taken": "created_invoice_plan", 8 "plan_file": "PLAN_WHATSAPP_20260305_143000_invoice.md", 9 "routed_to_approval": true 10}

Reply draft log:

json
1{ 2 "timestamp": "2026-03-05T14:30:00Z", 3 "action_type": "whatsapp_draft_created", 4 "source_file": "WHATSAPP_20260305_143000.md", 5 "recipient": "Client Name", 6 "intent": "urgent_query", 7 "priority": "high", 8 "approval_file": "WHATSAPP_REPLY_20260305_143000.md" 9}

Archive log:

json
1{ 2 "timestamp": "2026-03-05T14:30:00Z", 3 "action_type": "whatsapp_archived", 4 "source_file": "WHATSAPP_20260305_143000.md", 5 "intent": "general", 6 "priority": "low", 7 "reason": "no_response_needed" 8}

Completion Signal

Always end with: <promise>TASK_COMPLETE</promise>

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is whatsapp-triage?

Ideal for Task Management Agents requiring automated workflow triage and CEO briefing capabilities via WhatsApp. whatsapp-triage is a local-first autonomous AI employee that detects intent, assigns priority, and creates action plans based on WhatsApp messages.

How do I install whatsapp-triage?

Run the command: npx killer-skills add muhammadwaheedairi/Personal-AI-Employee/whatsapp-triage. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for whatsapp-triage?

Key use cases include: Automating workflow triage for CEOs, Generating weekly briefings from WhatsApp messages, Routing urgent actions to pending approval folders.

Which IDEs are compatible with whatsapp-triage?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for whatsapp-triage?

Requires file system access to `/Needs_Action/` and `/Pending_Approval/` directories. All WhatsApp sends require human-in-the-loop approval. Limited to processing `.md` files.

How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add muhammadwaheedairi/Personal-AI-Employee/whatsapp-triage. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use whatsapp-triage immediately in the current project.

Related Skills

Looking for an alternative to whatsapp-triage or another community skill for your workflow? Explore these related open-source skills.

View All

widget-generator

Logo of f
f

f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

149.6k
0
AI

flags

Logo of vercel
vercel

flags is a Next.js feature management skill that enables developers to efficiently add or modify framework feature flags, streamlining React application development.

138.4k
0
Browser

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI