Guide · Agent skill
How to use the Lead Generation Expert Knowledge agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The Lead Generation Expert Knowledge skill (lead-hand-skill) is a set of written instructions that teaches an AI agent a structured method for B2B lead generation. Once it is installed, the agent follows the playbook when you ask it to build a prospect list: defining an ideal customer profile, researching companies and decision-makers on the public web, scoring leads, removing duplicates and exporting the results. This guide explains what the skill does, how to install it, how to trigger it, and what to check before you rely on its output.
Lead Generation Expert KnowledgeRightNow-AI/openfang · Apache-2.0 Download the skill Quick start
- 1Download and unzip. Download the skill ZIP from this page and unzip it to get HAND.toml and SKILL.md.
- 2Place the skill. Put the folder in ~/.claude/skills/ for personal use, in a project's .claude/skills/ folder, or upload it under Settings → Capabilities → Skills in the Claude apps.
- 3Describe your ideal customer. State the industry, company size, geography and buyer title you want to reach.
- 4Ask for research and enrichment. Request the companies or decision-makers you need and which fields to collect.
- 5Review the scores. Check the 0-100 score and grade for each lead and the reasons behind high or low values.
- 6Verify and export. Confirm contact details against their sources, then ask for CSV, JSON or a markdown table.
What is an agent skill, and what does this one make the agent do?
An agent skill is a folder of instructions that an AI agent loads when a task matches it. This skill is reference knowledge only: its main file, SKILL.md, is a prompt-only playbook and it contains no scripts of its own. It does not give the agent new tools by itself, so the agent's web research relies on whatever browsing or search capability your setup already provides.
The playbook covers the full lead workflow: writing an ideal customer profile (industry, company size, geography, technology, budget signals, decision-maker and pain points), searching for target companies and people, enriching records with contact and company details, scoring each lead on a 0-100 rubric, merging duplicates, and producing output as CSV, JSON or a markdown table.
How do you install the lead-hand-skill?
Download the ZIP file for the skill from this page and unzip it. The package contains two files: HAND.toml and SKILL.md. Then choose one of these locations:
- Personal install: unzip into ~/.claude/skills/ so the skill is available across your projects.
- Project install: unzip into the .claude/skills/ folder inside a specific project so it applies only there.
- Claude apps: upload the skill in the Claude apps under Settings → Capabilities → Skills.
How do you trigger the skill with a natural request?
You do not need a special command. Describe the lead task in plain language and mention the kind of customer you want. The skill's guidance is built around an ideal customer profile, so giving the industry, company size, location and buyer title in your first message gives the agent what it needs to apply the scoring rubric.
- Build an ICP for a payroll software company selling to 50 to 500 employee firms in the UK, then find 20 target companies.
- Find heads of operations at logistics companies in Germany and score each lead using the 0-100 rubric.
- Take this list of 40 contacts, remove duplicates by company domain and fuzzy name match, and output a markdown table ranked by score.
- Export the enriched leads as JSON with a field for where each detail came from.
What does the output look like?
Results can be returned in three formats defined in the skill. CSV includes columns such as Name, Title, Company, Company URL, LinkedIn, Industry, Size, Score, Discovered and Notes. JSON uses a similar set of fields with nested enrichment data. A markdown table gives a short ranked view with a key signal for each lead.
The scoring rubric adds up to 100 points across ICP match (30), growth signals (20), enrichment quality (20), recency (15) and accessibility (15). Scores map to grades: 80-100 is A (hot), 60-79 is B (warm), 40-59 is C (cool) and 0-39 is D (cold).
Name,Title,Company,Company URL,LinkedIn,Industry,Size,Score,Discovered,NotesWhat are the safety and compliance rules?
The skill sets clear limits. It says to use only publicly available information, respect robots.txt and rate limits, and record where each detail came from. It forbids scraping behind logins or paywalls, bypassing CAPTCHAs or anti-scraping measures, fabricating contact details, storing sensitive personal data such as financial or health information, and sending unsolicited communications on your behalf. Email format patterns are listed only as reference, and the skill notes they must not be used to send unsolicited email.
- It declares no required dependencies and does not run scripts.
- It does use the network, because web research is its core task.
- It says lead data should stay in local files and should not be sent elsewhere.
- It gives only general ethics guidance, not legal advice.
What are the limits you should plan for?
The skill does not send outreach emails and will not do so for you. It is also not a substitute for a data provider, and the skill itself says the agent can still make mistakes. Check every email address, phone number and funding claim against its original source before acting on it. If you work in a market covered by contact-data law such as GDPR or CAN-SPAM, get legal advice; this skill does not provide it.
- Verify contact details against the cited source before use.
- Treat scores as a consistent ranking method, not a measure of buying intent.
- Mark or remove leads with no activity for more than 90 days, as the skill recommends for stale data.
What people use it for
- Building a list of 30 SaaS companies in a region that have recently raised funding or are hiring.
- Scoring an existing contact list with one consistent rubric so sales reps prioritise the same way.
- Removing duplicate company and contact rows exported from several sources before a CRM import.
- Drafting a structured ICP document that a small team can review before starting outreach.
Frequently asked questions
Is the lead-hand-skill free?
The skill is published under the Apache-2.0 licence, which is stated on this page. The source does not list prices for the skill, so check the licence file in the repository for the terms that apply to your use.
What do I need to install it?
The skill lists no required dependencies. You need an agent environment that supports skills, such as the Claude apps or Claude Code with a skills folder, and the ability to unzip the package.
Where do I put the skill folder in Claude Code?
Unzip it into ~/.claude/skills/ for a personal install, or into .claude/skills/ inside a project for a project-only install. Then ask the agent for a lead task that matches its description.
Does the skill send emails or scrape LinkedIn?
No. The skill forbids sending unsolicited communications and forbids scraping behind logins. It uses only public information, so any LinkedIn detail should come from a public profile.
How does it compare to a paid lead database?
This skill is a method and set of rules, not a data source. It structures research you do from public web pages, while paid databases supply their own contact records. Choose it when you want no data tool purchase and can verify results yourself.
Why are some of the agent's results wrong?
The skill warns that the agent can make mistakes, including on contact details. Compare each name, title, email and funding fact with its original source, and discard any detail you cannot confirm.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
This skill suits sales, business development and marketing people, or small teams, who want a consistent, documented method for building prospect lists from public web sources without buying a data tool. It is a poor fit if you need automated outreach, bulk personal data collection, or guidance on legally regulated contact-data rules, since it forbids unsolicited messages and gives only general ethics guidance.
Download the skill