agentleFS
harborrankharborrank/harborrank-skills/skills/lead-teardown/SKILL.md

lead-teardown

Skill2 starsChanged yesterdayMITView on GitHub

For agencies and freelancers who sell SEO to local businesses: prospect a trade in a city from Google Maps, shortlist the best leads, research each one (three-point Maps pack check, top competitors, site, contact), and produce a one-page teardown plus outreach emails to pitch with. Use when the user wants local SEO leads, prospects or clients, a sales teardown of a local business, or cold outreach to businesses such as roofers, dentists, or electricians.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 8 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If neither tool is available, or whoami fails because the user is not signed in, tell the user plainly: "HarborRank isn't connected, so I …

    line 16
  2. Distinctive:

    Do not stop there. Continue in fallback mode (below).

    line 27
  3. Very specific:

    If creditsRemaining will not cover the planned calls, give the user the estimate and ask before spending.

    line 33
  4. Very specific:

    If a tool fails for insufficient credits, stop making paid calls, tell the user, and finish with free tools and the data you already have.

    line 35
  5. Very specific:

    If the user asks to upgrade, change plan or buy credits, call get_upgrade_link (no credits) and give them the link.

    line 39
  6. Very specific:

    When whoami returns mode: "self-hosted", there are no plan gates.

    line 40
  7. Distinctive:

    Use this mode only when Step 0 finds no connection.

    line 44
  8. Distinctive:

    Read their sites: services listed, service area, city pages, calls to action, and obvious gaps.

    line 49
  9. Very specific:

    Call the result a shortlist, never a teardown.

    line 52
  10. Distinctive:

    Start with the label Web-evidence only: no HarborRank data.

    line 60
  11. Very specific:

    If the project's SEO folder has seo-context.md (see seo-context), read it first for the site, market, audience, competitors and goals, and …

    line 66
  12. Distinctive:

    Use only figures that appear in that lead's teardown, written exactly as the teardown has them.

    line 95
  13. Distinctive:

    Follow-up, about 60 words, in the same thread: one different gap from the teardown.

    line 97
  14. Distinctive:

    Add the sender's name, postal address, and a one-line opt-out to both emails.

    line 100
  15. Very specific:

    When the user sends: update_lead with the owner name and email when known, and status contacted.

    line 102
  16. Distinctive:

    Never invent a number. Every figure in the summary must come from the teardown or a tool result.

    line 124

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

HarborRank Lead Teardown

Goal

Turn a trade and a city into a short list of local businesses worth pitching, each with a one-page teardown built from real data: where they show up in Google Maps from three points in their service area, what the top three profiles have that they do not, what their website earns, and the first three fixes. The teardown is the sales document; the agency sends it, the business owner reads their own numbers.

Step 0: Check connection

Before any other step, confirm the HarborRank MCP tools are loaded. Look for get_domain_overview or list_leads among your tools, including deferred tools you can load (the name may carry a prefix such as mcp__harborrank__). If they are there, call whoami: it uses no credits, confirms the user is signed in, and returns creditsRemaining for budgeting the run. Do not call a research tool to test the connection.

If neither tool is available, or whoami fails because the user is not signed in, tell the user plainly: "HarborRank isn't connected, so I can't pull live HarborRank data yet." Then tell them how to connect it:

  • Claude Code with the HarborRank plugin: run /mcp, choose harborrank, and sign in.
  • Claude Code without the plugin: run claude mcp add --transport http --scope user harborrank https://app.harborrank.com/mcp, then /mcp to sign in.
  • Claude on the web, desktop or Cowork: open Customize, then Connectors, search for HarborRank, choose Connect, and sign in. New accounts start with a 7-day free trial.
  • Cursor with the HarborRank plugin: open Cursor Settings, then MCP, and sign in to harborrank.
  • Gemini CLI with the HarborRank extension: run /mcp auth harborrank.
  • Other clients: add https://app.harborrank.com/mcp as a remote MCP server. Setup for each client is at https://harborrank.com/docs/mcp.

Tell them what happens when they connect: they sign in to HarborRank with Google or email and start a 7-day free trial. Then they come back to this conversation and say "connected", and you pick up the task where it stopped.

Do not stop there. Continue in fallback mode (below). If the user connects HarborRank mid-task, run Step 0 again and switch to the full workflow.

Plan limits

On hosted HarborRank, research tools spend credits. New accounts start with a 7-day free trial (1,000 research credits, one project); paid plans start at $29/month with 15,000 credits. Search Console, saved keywords, rank tracker reads, and site crawls use no credits. A keyword search costs about 50 credits and a domain overview about 80. Full backlinks, lead research (research_lead), and Lighthouse checks need a paid plan; on Free, the backlinks tools return a preview (totals and the top 3 referring domains, about 50 credits) with an upgrade link.

  • If creditsRemaining will not cover the planned calls, give the user the estimate and ask before spending. Use the free tools first.
  • If a tool replies that a feature "is included in paid plans", tell the user once which feature needs a paid plan and pass on the upgrade link from the reply, skip that tool for the rest of the run, and finish with the other evidence. Do not retry it.
  • If a tool fails for insufficient credits, stop making paid calls, tell the user, and finish with free tools and the data you already have. If whoami warns that credits are low, say so before a large run.
  • If a HarborRank tool says the free trial hasn't started, or that credits ran out, give the user the link from that reply exactly as written, as a clickable link, and wait for them to say they're done before calling HarborRank again. Don't switch to answering without HarborRank data unless they ask you to.
  • Two extras are included in Pro and Agency: local grids larger than 5x5 (a 7x7 run_local_grid) and more than 3 tracked competitors per rank tracker (add_rank_tracker_competitors). If a tool replies that something needs Pro or Agency, tell the user once what Pro adds, pass on the Pro link from the reply exactly as written, and carry on with what their plan includes (a 5x5 grid, the first 3 competitors). Do not retry it.
  • Mention Pro unprompted only when the task needs it: a whole-town map, or watching more than 3 competitors. Say it once per run, in one sentence, with the link from get_upgrade_link (plan: "pro").
  • If the user asks to upgrade, change plan or buy credits, call get_upgrade_link (no credits) and give them the link. They pay on HarborRank by card; never say the upgrade is done until whoami shows the new plan.
  • When whoami returns mode: "self-hosted", there are no plan gates.

Fallback mode (no HarborRank connection)

Use this mode only when Step 0 finds no connection. It works from web search, page reading, and files the user provides.

Can still do

  • Find businesses in the trade and city with web search. Record each one's name, website, and public rating and review count, with the source.
  • Read their sites: services listed, service area, city pages, calls to action, and obvious gaps.
  • Compare them with the sites that appear at the top of web search for the main trade-and-city query.
  • Apply the shortlist rules (has a website, 20 or more reviews at 4.3 or better, not a chain).
  • Call the result a shortlist, never a teardown. The pitch only works when the owner sees their own numbers, and those come from research_lead.

Must not estimate

Traffic, search volume, keyword difficulty, CPC, keyword counts, rankings, backlink counts, Maps pack positions, and Q&A counts. Write unknown for each. A web search shows which pages appeared for a query when you searched, not their Google ranking: report it as "appeared in web search", never as a position.

Output

  • Start with the label Web-evidence only: no HarborRank data.
  • Name the source (URL or file) of every figure you report.
  • End, after any recommended next workflow, with one sentence that names the numbers this report is missing: "Connect HarborRank (start with a 7-day free trial) to pull these businesses' Maps listings; the three-point pack check and the one-page teardown need a paid plan." Do not add more sales copy than that.

Required inputs

If the project's SEO folder has seo-context.md (see seo-context), read it first for the site, market, audience, competitors and goals, and ask only for what it doesn't cover.

  • projectId
  • A trade or Google Business category (e.g. roofing_contractor, dentist, electrician)
  • A city (city mode) or a coordinate and radius (coordinate mode)
  • Optional: how many leads to prospect (default 20), whether to run the AI Overview check (costs more)

HarborRank MCP tools

  • list_business_categories: find the exact category slug.
  • prospect_leads: pull the Maps listings for the category in the city into the project as leads (name, site, rating, reviews, coordinates, photo count, categories, hours).
  • list_leads: see what is already in the project and its outreach status.
  • research_lead: the paid step. For one lead it runs the money-keyword SERP checks, the three-point pack check, the top-three profile comparison with Q&A counts, the domain footprint, contact discovery, a 10-page crawl, and optionally the AI Overview check, then writes the teardown. Requires a paid plan on hosted HarborRank.
  • get_lead_teardown: the Markdown one-pager for a researched lead.
  • update_lead: record the owner's name, email, notes and outreach status.
  • get_local_serp_results, get_google_business_questions: for a deeper look at one market when the teardown raises a question.
  • run_local_grid: where one business shows in Google Maps for one keyword from a square of points around it, saved in the project as a heatmap. A 5x5 grid is 25 Maps searches; 7x7 (Pro and Agency) covers a whole town and makes the clearest picture for a pitch. Paid plans.
  • create_share_report: publish the finished work as a read-only page at a shareable link (see "Shareable report" below). Only after the user agrees.

Workflow

  1. Confirm the category slug with list_business_categories and the city with the user. One trade, one city per run.
  2. prospect_leads with the category and city. Report how many were found, added and skipped.
  3. Shortlist before spending credits. From list_leads, prefer businesses that have a website, 20 or more reviews at 4.3 or better (a real business that will value the pitch), and sit below the obvious leaders. Skip chains and businesses with no reviews. Tell the user which leads you will research and the approximate cost (research_lead costs about 200 credits each without the AI check).
  4. On the Free plan, prospect_leads and list_leads work but research_lead does not: hand over the shortlist with the data list_leads returned and say that teardowns need a paid plan. Otherwise, research_lead for each shortlisted lead, sequentially. Do not run the AI check unless the user asks; it is the expensive part.
  5. get_lead_teardown for each. Read it before handing it over: if a section is thin (no pack points because the listing had no coordinates, no competitors because the SERP was empty) say so rather than sending a half document.
  6. Optional, for the one or two leads the user will pitch first: offer a run_local_grid scan for the lead's primary keyword, using the name, coordinates and place ID from list_leads. It shows where the business disappears across town, which the three-point check only samples. State the point count before running it. If the user is on Pro or Agency and the lead serves a whole city, suggest 7x7.
  7. Summarize for the user: for each lead, one line with the sharpest gap (the number and the competitor's number), and whether a contact email was found.
  8. Offer to draft outreach: for each lead, a first email and a follow-up to send four days later. If the user accepts:
    • Use only figures that appear in that lead's teardown, written exactly as the teardown has them. If a point has no figure in the teardown, leave it out.
    • First email, about 120 words: the business name in the subject, the sharpest gap (their number and the leader's), the first fix, and an offer to send the full teardown. Ask one low-pressure question.
    • Follow-up, about 60 words, in the same thread: one different gap from the teardown. Do not repeat the first email.
    • Pricing, packages, and guarantees belong in the email or not at all, never in the teardown. Use only the user's own offer; if they have not given one, leave a [your offer] placeholder. Never promise rankings or results.
    • Address the owner by name only if the teardown found it. If no email was found, say so and do not guess an address.
    • Add the sender's name, postal address, and a one-line opt-out to both emails; ask the user for the address if you do not have it.
    • One first email and one follow-up per lead. Do not draft a third message; a lead that has not replied after the follow-up is done.
  9. When the user sends: update_lead with the owner name and email when known, and status contacted. After the follow-up goes out, status followed_up; if the lead answers, replied.

Output format

Per lead:

  • Business, city, rating and reviews
  • Pack position at the three points for the primary keyword (and the grid's top-3 share when one ran)
  • The one gap with the biggest gap between their number and the leader's
  • Contact found: yes/no (and confidence)
  • Link or path to the teardown

Then the list of leads you skipped and why, and the offer to draft outreach.

  • Next workflow: end with one recommended next workflow, the reason for it, and the input it starts from. Default: seo-project-setup for any lead that replies and becomes a client, so the work starts with goals, positioning, and Search Console. If no lead has replied yet, recommend lead-teardown for the next trade or city instead.
  • Check back: end with one dated follow-up: the date (today plus 5 days, written out, for example "Monday, November 9") and the exact question to ask then: "Which of my leads need a follow-up email?" Say in one line why it needs that wait: most replies to cold outreach come within a few days, and the leads' statuses stay in HarborRank (list_leads). Add it to the Follow-ups list in the SEO folder's README.md (create the list if it is missing) so the next session finds it. One follow-up only; don't add reminders or repeat it. If scheduled tasks are available (Cowork and the Claude apps on paid plans), offer once to make this a weekly scheduled task instead, named for the site, so the check runs by itself; the user reviews and confirms the schedule.
  • Where the output goes: the teardowns stay in HarborRank on each lead (get_lead_teardown). Save a copy of each teardown with its email drafts as outreach/leads/<business-slug>.md. Save it in the project's SEO folder (see seo-project-setup); if there is none, ask once whether to create one or where to save instead. If the user prefers a doc or a sheet and a docs or spreadsheet tool is available, put reports in a doc and tables in a sheet instead. Tell the user where it was saved.
  • Shareable report: offer to publish each teardown the user wants to send as a shareable report with create_share_report: a read-only page at a link the prospect can open without signing in, which reads better in a cold email than an attachment. Use the teardown's findings and the fixes you'd make, never the lead's private contact details. Publish only after the user says yes; give them the link to paste into the outreach email. delete_share_report removes it.

Guardrails

  • Research charges credits; always state the count and cost before the batch.
  • Never invent a number. Every figure in the summary must come from the teardown or a tool result.
  • The teardown is written for the business owner. Do not add agency pricing or promises to it; put those in the email.
  • Respect the project's outreach rules: one first email and one follow-up, no third message without a reply.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.