Research Prospect

Prospect Research

Research a prospect company and contact to produce a structured intelligence brief. This skill gathers information from the company's website and the contact's LinkedIn profile, then cross-references with your company-profile.md to find personalization hooks.

When to Use

• Before writing any outreach (cold email, LinkedIn message)
• Before a demo or meeting (this skill is called automatically by demo-prep)
• When the user says "research [name] at [company]"
• When you need to understand a prospect before generating any output

Required Inputs

Ask the user for:

Company: [Name or website URL]
Contact: [Name and title — optional for company-only research]

If they only provide a name and company, that's enough. Research the rest.

Process

1. Load Company Profile

Read company-profile.md from the workspace root. You need this to:

• Identify shared connections between the user and the prospect
• Match the prospect's pain points to the user's product
• Determine if this company fits the user's ICP

If company-profile.md doesn't exist, tell the user: "Run the setup-company skill first to set up your profile. This makes research much more useful."

2. Research the Company

Navigate to the company's website and gather:

| What to Find | Where to Look | |-------------|---------------| | What they do | Homepage hero section, About page | | Who they sell to | Pricing page, case studies, customer logos | | Company size | About page, LinkedIn company page, job postings | | Sales model | Pricing page structure, "Book a Demo" vs "Start Free Trial" | | Recent news | Blog, press page, recent funding announcements | | Tech stack signals | Job postings, integrations page | | Demo / trial motion | CTA buttons, pricing page, "how it works" |

Also check for (see [research checklist](references/research-checklist.md)):

• Whether they have a "Book a Demo" button (relevant for demo automation)
• Whether they're hiring for roles related to the user's product
• Recent product launches or pivots
• Customer testimonials that reveal their value prop

3. Research the Contact

Look up the contact on LinkedIn (use browser tools) and gather:

| What to Find | Why It Matters | |-------------|---------------| | Current role and tenure | How long they've been there — new hires are more open to change | | Department | Maps to which angle to use in outreach | | Previous companies | Shared connection opportunities | | Education | Shared school connections | | Recent posts or activity | Conversation hooks, what they care about publicly | | Responsibilities | What they likely own and what metrics they're measured on |

4. Find Personalization Hooks

Cross-reference the contact's background with the user's company-profile.md:

| If You Find... | Hook to Use | |----------------|-------------| | Same school (any level) | "Fellow [school] alum" | | Same previous company | "Ex-[company] here too" / "My co-founder is ex-[company]" | | Same city or region | Geographic connection | | Mutual LinkedIn connections | "We both know [name]" | | They posted about a relevant topic | Reference the post directly | | They recently changed roles | "Congrats on the new role" | | Their company just raised funding | "Congrats on the raise" |

Priority order: Shared school > shared company > shared connection > recent event > geographic.

5. Assess ICP Fit

Based on research, assess how well this prospect matches the user's ICP from company-profile.md:

| Signal | What It Means | |--------|--------------| | Role matches target roles | Strong fit | | Company size matches ICP | Strong fit | | Industry matches ICP | Strong fit | | They have signals from ICP (e.g., "hiring SDRs") | Strong fit | | None of the above | Weak fit — note this in the output |

Output Format

# Prospect Research: [Contact Name] @ [Company Name]
**Researched:** [Date]

## Company Intel
| Field | Details |
|-------|---------|
| **Company** | [Name] |
| **Website** | [URL] |
| **Industry** | [Industry] |
| **Size** | [Estimate — employees, stage, funding] |
| **What They Do** | [2-3 sentences] |
| **Who They Sell To** | [Their customers] |
| **Sales Model** | [Inbound / outbound / PLG / enterprise] |
| **Recent News** | [Funding, launches, hires — if any] |

## Contact Intel
| Field | Details |
|-------|---------|
| **Name** | [Full name] |
| **Title** | [Job title] |
| **Department** | [Sales, Engineering, L&D, Product, etc.] |
| **Tenure** | [How long in current role] |
| **Background** | [Previous companies, education] |
| **Recent Activity** | [Posts, articles, comments — if any] |
| **Likely Priorities** | [What they probably care about based on role] |

## Personalization Hooks
- **Strongest hook:** [The best connection point]
- **Additional hooks:** [Other connection points]

## ICP Fit
[Strong / Medium / Weak] — [1 sentence explaining why]

## Relevant Pain Points
Based on their business and role, they likely care about:
1. [Pain point 1 — mapped to user's product]
2. [Pain point 2 — mapped to user's product]
3. [Pain point 3 — mapped to user's product]

Instructions

1. Always research before generating. Never fabricate company details or contact background. If you can't find something, say so.
2. Use browser tools for LinkedIn. Navigate to LinkedIn and search for the person. Don't guess at their background.
3. Be specific about hooks. "Fellow IIT Delhi alum" is useful. "You both work in tech" is not.
4. Flag weak ICP fit. If the prospect doesn't match the user's ICP, say so clearly. Don't force a fit.
5. Keep it scannable. Tables and bullet points, not paragraphs. The user will skim this in 30 seconds.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: tough-tongue
• Source: tough-tongue/demo-prep-skills
• License: MIT

Install and usage instructions live in the source repository linked above.