> ## Documentation Index
> Fetch the complete documentation index at: https://closedloop.sh/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Skills Overview

> Pre-built skills for Claude Code, grouped by the team they serve: product, engineering, sales, marketing, leadership, and customer success.

Turn every customer conversation into structured product insights, inside Claude Code. These skills connect to your ClosedLoop AI data, so the answer comes from real call transcripts, tickets, and feedback rather than from a summary someone wrote by hand. They are grouped below by the team that gets the most out of them, but nothing stops you running any of them.

## Install

<Steps>
  <Step title="Connect the MCP">
    If you haven't already, add the ClosedLoop AI MCP to Claude Code:

    ```bash theme={null}
    # US workspace
    claude mcp add --transport http closedloop-ai https://mcp.closedloop.sh

    # EU workspace
    claude mcp add --transport http closedloop-ai https://eu.mcp.closedloop.sh
    ```

    Restart Claude Code, select `closedloop-ai` from `/mcp`, and authorize.
  </Step>

  <Step title="Install the skills plugin">
    In Claude Code, add the community marketplace, then install the skills:

    ```
    /plugin marketplace add anthropics/claude-plugins-community
    /plugin install closedloop-skills@claude-community
    ```
  </Step>

  <Step title="Use a skill">
    ```
    /closedloop-deep-dive dynamic pricing
    ```
  </Step>
</Steps>

## For product

<CardGroup cols={2}>
  <Card title="Deep Dive" icon="magnifying-glass" href="/docs/skills/deep-dive">
    Deep-dive into any topic against all your product insights. Reads every matching insight, context record, and conversation transcript.
  </Card>

  <Card title="PM Prep" icon="lightbulb" href="/docs/skills/pm-prep">
    Discovery brief for PMs. Learning goals, knowledge gaps, data-grounded questions, and whether pain is a market signal or outlier.
  </Card>

  <Card title="Scope Guard" icon="shield-halved" href="/docs/skills/scope-guard">
    Check a spec against what customers actually said, before engineering time goes into it. Every requirement gets a verdict.
  </Card>

  <Card title="Synthetic Customer" icon="user" href="/docs/skills/synthetic-customer">
    Talk to any customer or segment as an AI persona grounded in real call transcripts, feedback, and CRM data.
  </Card>

  <Card title="Competitor Gap" icon="swords" href="/docs/skills/competitor-gap">
    Competitive intelligence from customer conversations: threat ranking, trend analysis, feature gaps, and your advantages.
  </Card>

  <Card title="Product Ambassadors" icon="users" href="/docs/skills/product-ambassadors">
    Find design partners, beta users, expert reviewers, and champions for a feature area, with one specific ask for each.
  </Card>
</CardGroup>

## For engineering

<CardGroup cols={2}>
  <Card title="Technical Spec" icon="file-code" href="/docs/skills/spec">
    A spec in Google design doc format, grounded in what customers said and the patterns already in your repo. Every decision cites a quote or a line number.
  </Card>

  <Card title="Bug Blast Radius" icon="bug" href="/docs/skills/bug-blast-radius">
    How important is this bug? Who reported it, how badly, what those customers are worth, and when it was last reported.
  </Card>
</CardGroup>

## For sales

<CardGroup cols={2}>
  <Card title="Sales Playbook" icon="bullhorn" href="/docs/skills/sales-playbook">
    Talk tracks from real customer quotes. General or micro-targeted to a specific customer with @ syntax.
  </Card>

  <Card title="Sales Proof" icon="handshake" href="/docs/skills/sales-proof">
    A proof packet for one prospect concern: the talk track, the verbatim proof points, and the permission caveat.
  </Card>

  <Card title="Reference Customers" icon="phone" href="/docs/skills/reference-customers">
    Who could take a private reference call, with the topics to cover, the topics to avoid, and the route in.
  </Card>
</CardGroup>

## For marketing

<CardGroup cols={2}>
  <Card title="Customer Proof" icon="quote-left" href="/docs/skills/customer-proof">
    Marketing-ready proof: website quotes, case-study candidates, ROI evidence, adoption stories, and reference candidates.
  </Card>

  <Card title="Case Study Scout" icon="book-open" href="/docs/skills/case-study-scout">
    Who has enough evidence for a real story, what proof is still missing, and the interview questions to ask them.
  </Card>
</CardGroup>

## For leadership

<CardGroup cols={2}>
  <Card title="Weekly Brief" icon="newspaper" href="/docs/skills/weekly-brief">
    Weekly intelligence brief that reads ALL evidence and synthesizes into a 40-line brief you can scan in 60 seconds.
  </Card>
</CardGroup>

## For customer success

<CardGroup cols={2}>
  <Card title="CSM Prep" icon="clipboard-check" href="/docs/skills/csm-prep">
    90-second pre-call brief for CSMs. Headline, landmines, top concerns in their own words, open threads, and shared pain context.
  </Card>

  <Card title="Ship Notify" icon="paper-plane" href="/docs/skills/ship-notify">
    Find every customer who asked for something you shipped, generate personalized outreach per tier.
  </Card>
</CardGroup>

## For everyone

<CardGroup cols={2}>
  <Card title="Watch" icon="bell" href="/docs/skills/watch">
    Describe an alert in plain language and get notified in Slack, Teams, or email when it crosses your threshold.
  </Card>

  <Card title="Send Feedback" icon="comment-dots" href="/docs/skills/send-feedback">
    Tell the ClosedLoop AI team when a result is wrong, missing, stale, or confusing, without leaving the conversation.
  </Card>
</CardGroup>

## Requirements

* [Claude Code](https://claude.ai/code)
* [ClosedLoop AI](https://closedloop.sh) account with connected data sources
* ClosedLoop AI MCP configured (see Step 1 above)

<Card title="MCP Tools Reference" icon="wrench" href="/docs/mcp-server/tools">
  Reference for the MCP tools that power these skills
</Card>