customer-referral

customer-referral is a skill for Claude Code from First-Touch-Inc/firsttouch-agent-skill-packs. It costs 106 tokens per session (1,925 once invoked), scanned A, original, MIT.

A customer follow-up workflow that prepares LinkedIn connection requests and thank-you or referral messages after someone becomes a customer. LinkedIn is a professional networking site, and a referral is a recommendation to another potential customer.

In plain words
What is it for?
It supports post-sale LinkedIn connections, personal thank-you messages, feedback requests, and low-pressure referral requests after a deal closes or onboarding begins.
Why use it?
It helps teams thank new customers and ask for feedback or introductions without making the first message feel like a sales request. It also checks who should own the follow-up.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the ae-pack plugin — 13 skills, 1 MCP server shipped together

Good fit It supports post-sale LinkedIn connections, personal thank-you messages, feedback requests, and low-pressure referral requests after a deal closes or onboarding begins.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add First-Touch-Inc/firsttouch-agent-skill-packs --skill customer-referral
Clone the repo
git clone --depth 1 https://github.com/First-Touch-Inc/firsttouch-agent-skill-packs

Made for: Claude Code.

Or install ae-pack, the plugin that ships this one along with the rest of its 13 skills, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for customer-referral

README.md
[![agentmods](https://agentmods.dev/badge/skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral/github.svg)](https://agentmods.dev/skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral)
Your own site
<a href="https://agentmods.dev/skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral"><img src="https://agentmods.dev/badge/skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for customer-referral

Your own site · 80×15
<a href="https://agentmods.dev/skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral"><img src="https://agentmods.dev/badge/skills/first-touch-inc/firsttouch-agent-skill-packs/customer-referral.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,925 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00106 $0.01925
Opus 5 $0.00053 $0.00962
Sonnet 5 $0.00021 $0.00385
Haiku 4.5 $0.00011 $0.00193

Measured 11d ago against content hash 3271fc533a2c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

customer-referral scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/ae-pack/skills/customer-referral/SKILL.md · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Customer Referral Thank-You

Outcome: When someone becomes a customer, connect with them on LinkedIn and send a warm thank-you that invites product feedback and a low-pressure referral signal, without making the first customer touch feel like a sales ask.

First-run onboarding gate

Before running this skill for the first time in a workspace, load ../../references/onboarding.md and complete the onboarding questions. Do not proceed until you know: LinkedIn account type (free/basic = no connection notes; recommend 10 connection requests/day and never exceed the FirstTouch max of 20/day; Sales Navigator/Premium = connection notes available; recommend 20 connection requests/day and never exceed the FirstTouch max of 30/day), HubSpot access (MCP, service key/private app token, HubSpot list only, or none), and which play the user wants to run. Recommend high-intent plays before broader outbound to keep the LinkedIn account healthy.

When to use

  • a deal moves to Closed Won
  • a new customer starts onboarding
  • a user/admin completes setup or hits first value
  • the founder, AE, CSM, or account owner wants to thank the customer personally
  • the team wants to ask for feedback and light referrals without launching a bulk campaign

Inputs

  • Customer source: HubSpot Closed Won/customer list, HubSpot workflow/list output, CSV/imported customer list, or FirstTouch-accessible customer source
  • Sender/routing rule: founder, account owner, AE, CSM, or named executive sender
  • Relationship context: product purchased, use case, onboarding status, first-value milestone, or why they chose the product
  • Referral ask style: soft network ask, partner/customer intro ask, or feedback-only if the relationship is too early

HubSpot is preferred for Closed Won/customer routing and CRM logging. It is not mandatory when the user provides a customer CSV, imported list, or FirstTouch-accessible customer source with explicit customer status.

Step-by-step

Before drafting or queueing any contact, run the standard safety gates from ../../references/safety-governance.md: Gate 0 suppression/DNC, Gate 1 duplicate/recent-contact check, Gate 2 owner/CSM routing, Gate 3 daily cap sharing, and Gate 4 human approval. Suppressed, opted-out, duplicate, recently contacted, or misrouted records are skipped and logged.

Read the full file on GitHub · 124 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 124 lines · 106 tokens per session scan A 3271fc533a2c

Subscribe to this mod's changes

customer-referral is a skill published in the GitHub repository First-Touch-Inc/firsttouch-agent-skill-packs (5 stars, last pushed 2mo ago), licensed MIT. It adds 106 tokens to every session and 1,925 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.