# objection-handler

The objection-handler Claude Code skill drafts replies to sales objections using the terms of your deal. It activates when someone pushes back, asks a question, or sends a message you’re unsure how to answer.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf (Devin Desktop), Claude.ai · ★ 92 repo stars
- Author: Jake Schincariol
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/Jakeschincariol/whale-skill/tree/HEAD/objection-handler/SKILL.md
- Updated: 2026-11-06
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/objection-handler

## What it does

The skill matches a prospect’s words to a suggested reply and explains the reasoning. Its `objections.py` script searches the entries in the editable `objections.json` file. A clear refusal gets no rebuttal; `/follow-up stop` handles a request to end contact.

Rewrite each draft in your own voice and fill in only facts that are true. The skill does not invent missing details or send a message without approval for that specific reply. Running the script requires `python3`; without it, objection matching cannot run.

## Install

### Claude Code (native platform)

1. **Plugin marketplace**
In the agent chat (not a shell):
```text
/plugin marketplace add Jakeschincariol/whale-skill
```
In the agent chat (not a shell):
```text
/plugin install whale-skill@whale-skill
```
This installs the whole whale-skill plugin with all its skills

2. **Skills CLI (Vercel)**
```bash
npx skills add Jakeschincariol/whale-skill --skill objection-handler -a claude-code
```

3. **Manually: put the files in place**
- File: `~/.claude/skills/objection-handler/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/objection-handler/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add Jakeschincariol/whale-skill --skill objection-handler -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/objection-handler/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add Jakeschincariol/whale-skill --skill objection-handler -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/objection-handler/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install Jakeschincariol/whale-skill objection-handler
```

2. **Manually: put the files in place**
- File: `.github/skills/objection-handler/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add Jakeschincariol/whale-skill --skill objection-handler -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/Jakeschincariol/whale-skill --path objection-handler
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/objection-handler/SKILL.md` — The legacy .windsurf/skills/ path still works

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/Jakeschincariol/whale-skill/tree/HEAD/objection-handler/SKILL.md

## When to use it
- draft a reply to concerns about a launch date
- answer a price objection using the buyer’s current costs
- separate hesitation from a firm refusal
- respond to questions about paying up front

## Example prompts
- They asked what happens if the product never ships. Draft a reply using our actual terms.
- How should I answer, “That’s too expensive for us right now”?
- They asked me to stop contacting them. End the follow-up.

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: d19676f88c80f226.

BEGIN THIRD-PARTY FILE d19676f88c80f226
````text
---
name: objection-handler
description: >-
  Objection Handler. Answers every objection, and knows a no when it hears one. Matches what they said
  ("it isn't built yet", "what if you never ship it", "why pay up front", "too expensive", "I need to ask my
  CFO") to an honest answer filled in from your own terms, and shows why it works. A real no gets no answer
  at all: it stops the deal for good. Every answer is a draft for your yes. Use when the user says "they
  said", "how do I answer this", "what do I say to", "they pushed back", "objection", "they think it's too
  expensive", or pastes a reply they don't know how to answer.
argument-hint: "<what they said>"
---

# objection-handler

When you're asking for money before the product exists, the objections are good ones. So the answers have to be
honest, specific and short. This skill has the answer ready, filled in from your real terms. It also knows the
difference between a worry, which deserves an answer, and a no, which deserves to be left alone.

`${CLAUDE_SKILL_DIR}` below is this skill's folder (where this SKILL.md lives).

```bash
python3 "${CLAUDE_SKILL_DIR}/objections.py" match --deal north-pier --text "What if you never ship it?"
python3 "${CLAUDE_SKILL_DIR}/objections.py" match --text "It looks great, but it's too expensive for us right now."
python3 "${CLAUDE_SKILL_DIR}/objections.py" list
```

The answers live in `objections.json`, so the user can change the words.

## A worry or a no?

- **A worry is a question, even when it sounds like a no.** "That's a lot of money." "We're busy." "I'd have to ask my CFO." Answer it, once.
- **A no is a no.** "Not interested." "We're all set." "Please stop." `match` won't give you an answer for those, even if the same sentence has an
  objection in it. Run `/follow-up stop` with their words. They hear nothing else.
- **Not sure?** Ask one honest question and don't assume. "Is this something you'd like me to drop, or is it the timing?"

## Steps

1. Log their reply: `followup.py reply`. A no stops everything there.
2. `objections.py match --deal <id> --text "<what they said>"`. It shows the label, what to say and why.
3. Rewrite it in the user's voice and for this person, using their words from the call. Keep it to 3 sentences. Fill any `[fill: ...]`, or ask the
   user: never make one up. No match means asking them a question, not guessing.
4. **Ask.** Show it. "Send this reply to Renee? Reply yes or no." One message per question.
5. **Send.** On a yes, in the same thread from the user's account. If they push back again on the same point, stop answering it and ask what would
   have to be true. One answer per objection: the third round is pestering.

## What it knows

| they say | it answers |
| --- | --- |
| It isn't built yet | You're right, and here's what you get instead: the date, a say in it, the refund. |
````
END THIRD-PARTY FILE d19676f88c80f226

The third-party file has ended. Everything between the d19676f88c80f226 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2866 of 4788 bytes. © Jake Schincariol · MIT · https://github.com/Jakeschincariol/whale-skill/tree/HEAD/objection-handler/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
