agentleFS
prosyo-aiprosyo-ai/gtm-skill/skills/reply-handler/SKILL.md

reply-handler

Skill13 starsChanged 21 days agoMITView on GitHub

Use this skill when outbound replies need triage and answers: \"sort these replies\", \"draft responses to my inbox\", \"what do I say to this reply\", \"handle replies from my campaign\". Classifies each reply (interested, question, objection, referral, not now, not interested, out of office, wrong person, unsubscribe), sets priority and the next action, drafts a response for each that moves toward a meeting or closes cleanly, and updates what the CRM needs to know. Drafts only: a human approves before anything is sent.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this when replies arrive from a campaign.

    line 16
  2. Distinctive:

    Read references/reply-playbook.md for a draft pattern for each class.

    line 39
  3. Distinctive:

    MUST suppress unsubscribes and hostile replies across all channels immediately.

    line 69
  4. Distinctive:

    MUST get approval before sending any drafted reply.

    line 70
  5. Distinctive:

    MUST stop all sequences for anyone who replied.

    line 71
  6. Distinctive:

    NEVER argue with a "not interested".

    line 72
  7. Distinctive:

    NEVER invent pricing, features or availability in an answer.

    line 73

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Use this when replies arrive from a campaign. Speed decides how many positive replies become meetings: interested prospects answered within an hour book far more often than those answered the next day.

Inputs

  • Each reply with the message it answers.
  • Offer, a booking link or the sender's availability, and approved answers to common questions.
  • Who approves outgoing replies.

The play

1. Classify every reply

ClassSignalPriority
Interested"Sure", "Tell me more", "Let's talk", asks for timesNow
QuestionAsks about price, how it works, integrationsNow
Referral"Talk to Sam"Today
Objection"We already use X", "No budget", "Not a fit because..."Today
Not now"Reach out in Q3", "Busy until after launch"Today, then schedule
Not interested"No thanks", "Not interested"Today, close politely
Wrong person"I don't handle this"Today, ask who does
Out of officeAuto-replyPause; resume after return date
Unsubscribe or hostile"Remove me", "Stop"Now: suppress everywhere

Read references/reply-playbook.md for a draft pattern for each class.

2. Draft the response

  • Answer what they asked first, in their length. A one-line reply gets a short answer.
  • Interested: confirm, offer two specific times or the booking link, and one line on what the call will cover.
  • Question: answer directly; then the meeting question.
  • Objection: acknowledge, one clarifying question, respect the answer.
  • Referral: thank them, then contact the referred person naming who suggested it.

3. Set the next action

Each reply gets: classification, draft, next step, follow-up date and the CRM update (status, reason, notes).

4. Hand over for approval

Group drafts by priority. Unsubscribes are processed immediately (suppression is not a sales message, so it needs no approval to stop sending).

Output format

<prospect> | <class> | priority | draft reply | next step + date | CRM update

What good looks like

  • Interested and question replies have drafts ready within minutes.
  • Drafts answer the exact question before asking for the meeting.
  • "Not now" replies get a dated follow-up, not a lost note.
  • The common mistake is answering every positive reply with a calendar link and nothing else, which reads like a bot and drops conversion.

Rules

  • MUST suppress unsubscribes and hostile replies across all channels immediately.
  • MUST get approval before sending any drafted reply.
  • MUST stop all sequences for anyone who replied.
  • NEVER argue with a "not interested".
  • NEVER invent pricing, features or availability in an answer.

Example prompts

  • "Here are 37 replies from yesterday's campaign. Classify them and draft responses."
  • "They replied 'what does it cost?'. Draft an answer that gets to a meeting."
  • "Set up reply categories and templates for our SDR team."

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.