Message Reply Skill

Draft platform-appropriate professional replies with the user's authentic voice.

Workflow

User pastes message → Identify platform → Apply platform rules → Generate reply + shorter alternative

Required Context

Input	Description
received_message	The message to reply to (user pastes it)
platform	linkedin, fiverr, upwork, interview, email, whatsapp, other

Optional Context (Ask if relevant)

Input	When to Ask
your_context	If message mentions scheduling, availability, or specific situations
your_position	If message asks about skills/experience user may want to clarify
action_intent	If unclear whether to accept, decline, clarify, negotiate, or redirect
tone	Only if user wants to override platform default
length	Only if user wants to override (default: short)

Platform Detection

Auto-detect from message content or ask:

• "received this on linkedin" / "linkedin message" → LinkedIn
• "fiverr" / "gig" / "M Umer" (Fiverr name) → Fiverr
• "upwork" / "job" / "proposal" / "bid" → Upwork
• "interview" / "position" / "role" / "hiring" → Interview
• Formal greeting + business context → Email
• Casual / quick reply context → WhatsApp

Platform Rules

See references/platform-patterns.md for detailed patterns. Quick reference:

Platform	Default Length	Default Tone	Key Rule
LinkedIn	2-4 sentences	Professional + warm	Acknowledge, stay connected
Fiverr	2-3 sentences	Friendly	Redirect to project scope
Upwork	4-6 sentences	Natural/humanized	Lead with solution approach
Interview	3-5 sentences	Professional	Honest about skills, include context
Email	Structured	Formal	Clear sections, complete info
WhatsApp	1-3 sentences	Casual	Quick, direct

Output Format

Always provide TWO versions (user frequently asks for shorter):

## Reply

[Primary reply - platform appropriate]

## Shorter Alternative

[More concise version]

## Notes (if applicable)

- [Suggestions for additional context]
- [Warnings about message intent]

Generation Rules

1. Default SHORT - User almost always asks for shorter. Start concise.
2. No excessive emojis - Max 1 emoji, only if platform-appropriate
3. Honest positioning - Never oversell skills. Phrase gaps as "learning" or "exploring"
4. Redirect unclear requests - If message is vague, ask what they need
5. Include user context - If user provides context (travel, schedule), weave it in naturally

User Profile

See references/user-profile.md for skills, background, and common contexts.

Quick reference for honest positioning:

• Strong: n8n, Python, automation, AI agents, low-code/no-code, SDD/TDD
• Learning: React, Node.js (recently started)
• Certified: PIAIC Agentic AI Engineer

Engineering Methodology (When Relevant)

User follows Spec-Driven Development and Test-Driven Development.

Mention ONLY when client asks about:

• Quality/reliability ("How do you ensure it works?")
• Process ("What's your development approach?")
• Documentation ("Will this be documented?")
• Past bad experiences ("Last freelancer's code broke")
• AI concerns ("Do you just use AI to generate code?")

How to phrase in replies:

If asked about process:

I follow spec-driven development — I document requirements first, write tests
before code, then implement. You get verified, documented code, not just
"it works on my machine."

If asked about AI usage:

Yes, I use AI tools like Claude Code, but I drive the architecture and verify
everything with tests. Specs and tests ensure the output is correct and reliable.

If asked about reliability:

I write tests first that define expected behavior. Code only ships when tests pass.
You also get specs and documentation in the repo.

DON'T mention for: Generic greetings, simple questions, casual chats.

Common Scenarios

Generic greeting (Fiverr/LinkedIn)

Input: "Hello M Umer, Good day, How are you today?"
Output: "Hi! I'm doing well, thank you. How can I help you today?"

Skills question (Interview)

Input: "Are you experienced with n8n, Make, and Zapier?"
Context: Only n8n, self-learning
Output: "I have hands-on experience with n8n through personal projects. I haven't worked professionally with Make or Zapier yet, but I understand their concepts and can adapt quickly."

Off-topic request (Fiverr)

Input: "Can you help with Facebook ads?"
Context: User does n8n automation, not ads
Output: "Thanks for reaching out! I specialize in n8n workflow automation and AI agents rather than Facebook ads. If you need any backend automation or API integrations, I'd be happy to help!"

Availability question (Interview)

Input: "Are you available tomorrow for an interview?"
Context: Traveling to Lahore for class
Output: "Yes, I'll be traveling to Lahore tomorrow for my Agentic AI class. I'm available for an interview and can adjust timing. Please let me know what works best."

Anti-Patterns

Don't	Do Instead
Long paragraphs for simple replies	2-4 sentences max for initial contact
Oversell skills you're learning	"Exploring" / "Recently started" / "Learning"
Generic "I can help with anything"	Specific: "I specialize in n8n and AI agents"
Multiple emojis	Max 1, or none for formal platforms
Ignore user's additional context	Weave travel/schedule/constraints naturally

Checklist Before Output

• Platform identified and rules applied
• Length appropriate (default: short)
• Tone matches platform
• Honest about skills/experience
• User context included (if provided)
• Shorter alternative provided
• No overselling or excessive emojis