Hire an AI CRM records specialist

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A fix list sorted by how much each issue matters
• Missing field and stale date flags per deal
• Stage mismatch notes with the evidence
• Links to each record

How it does the job well

• Use the CRM's own field names and definitions from the export.
• Find duplicates, missing required fields and stale stages first.
• Propose changes as a list with record, field, old value, new value and source.
• Keep fixes small and reversible, and log each one.
• Ask how the team defines each stage before judging pipeline health.
• Research the buyer and their company before writing. One specific, true detail beats ten generic lines.
• Lead with the buyer's problem and outcome, not your features.

Mistakes it avoids

• Merging records without checking they are the same company or person.
• Filling fields with guesses.
• Changing live records before the exact changes are approved.
• Making claims about results, prices or terms that are not approved.
• Long first messages that ask for too much.
• Updating pipeline numbers on hope rather than evidence.

What a good result looks like

An example, not your task: Your forecast call is tomorrow and you do not trust the pipeline data.

• Summary. Counts of critical issues, attention issues and clean deals, so you see how bad it is first.
• Critical flags. Deals with a close date in the past or a blank or zero amount.
• Attention flags. Deals with a stale next step, no activity in 14 days or more, or only one contact.
• Fix list per deal. Each deal with a record link, stage and amount, its issues, and suggested values with their source.
• Bulk actions. Totals such as how many deals need dates pushed and how many need contacts added.

How the work is checked

• Every proposed change has a source note.
• Counts before and after are shown for each issue type.
• The checklist separates approved changes from open questions.
• Personal details are verifiable and correct.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Chris Voss and Aaron Ross would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• CRM hygiene auditor. Share your CRM deals. Get a fix list of missing fields, stale dates and stage mismatches. Changes nothing.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id crm-records-specialist only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/crm-records-specialist/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• Your open deals from the CRM or an export
• Your own stage names and required fields (optional)
• Which owner or team to check (optional)

Tools and access

• A scoped CRM pipeline export with its field definitions, or an authorized read connection
• Source notes for any proposed field changes
• Claude, ChatGPT or Gemini with web search on: Filling fields means looking things up. Without web search it can only tidy what you paste.

Functions and deliverables

• CRM hygiene auditor: Share your CRM deals. Get a fix list of missing fields, stale dates and stage mismatches. Changes nothing.
  • A fix list sorted by how much each issue matters
  • Missing field and stale date flags per deal
  • Stage mismatch notes with the evidence
  • Links to each record
  Source: Anthropic.
• CRM record filler: Use CRM records and available sources to draft missing fields with sources and confidence labels.
  • A before and after table for each record
  • A source and a confidence level for every new value
  • A review list of values it was not sure about
  • Possible duplicates flagged for you to merge
  • A change log you can use to update your CRM
  Source: Burhan Sebin.
• CRM deal updater: Update a deal in your CRM by saying what changed. See the before and after, accept it, and get the saved record link.
  • Before and after view of each field
  • The update saved after you accept
  • A record link to check the result
  • A note on any blank or unclear values
  Source: Anthropic.
• Deal notes and CRM updater: Share a sales call, email or meeting. Get CRM updates and follow-up drafts to review.
  • Logged activity per deal
  • Updated contacts and deal fields
  • Follow-up drafts for quiet deals
  • Next-step queue with owners
  Source: Anthropic.
• Brand profile builder: Share your website. Get a reusable profile covering your products, customers, voice and approved claims, with sources.
  • A folder of eight plain files, from a start file for your AI to a claims list and a change log
  • A tag on every fact, and every claim marked approved, pending or banned
  • A two-sentence diagnosis of your biggest gap, and a list of what's missing and who can answer it
  • A short slice for your next step, if you name one
• Chris Voss: Prepare a difficult conversation using Chris Voss’s recorded approach. Summarise the other side, ask calm questions and test how an agreement will work.
  • Your problem worked on the way Chris Voss would: their questions, their method, their push-back
  • The answer or the work, in Chris Voss's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Aaron Ross: Plan sales outreach and split team roles using Aaron Ross’s public work. Build a steadier flow of sales leads.
  • Your problem worked on the way Aaron Ross would: their questions, their method, their push-back
  • The answer or the work, in Aaron Ross's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• Your forecast call is tomorrow and you do not trust the pipeline data.
• You inherited a set of accounts and need to know what is missing.
• You suspect deals are sitting in the wrong stage.
• You want a monthly clean-up list for your own opportunities.

Limits to check before starting

• You want to fix records in bulk right now: this audit only reads and lists fixes, so apply them by hand or ask your AI to update one deal at a time.
• You need to clean contact lists, lead data or account records: ask your CRM admin or use your CRM's own duplicate and data tools.
• You have no open deals to check, or no export from your CRM: pull your pipeline export first, because there is nothing to audit without it.
• An export supports a file review. Live schema, history and saved changes require separate connection checks.
• Suggested values remain proposals until the exact changes are accepted.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Open Claude, ChatGPT or Gemini and paste in this skill. If your CRM is connected, your AI can read it. If not, upload your pipeline export.
• Say which deals to check: your own, one stage, or a close-date range. If you leave this out, your AI will ask.
• Add your own stage names and required fields if you have them (optional).
• Read the summary first, then the fix list. Check each suggested value against the record.
• Tell your AI which fixes to apply. It only writes after you approve, and only if your CRM allows writes. Otherwise, make the changes by hand.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/crm-records-specialist for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.