agentleFS
BenPremBenPrem/business-brain/.claude/skills/gbp-local-seo/SKILL.md

gbp-local-seo

Skill1 starsChanged 3 months agoMITView on GitHub

Audit and optimize Google Business Profiles and local SEO for local-service clients — completeness scoring, ranked fixes, 2x/week GBP post plans, NAP/citation sweeps, no-paid-tool rank checks. Triggers on "GBP", "Google Business Profile", "local SEO", "map pack", "local rankings". NOT for review responses or review-request campaigns — use review-manager.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    All outputs → clients/<slug>/deliverables/local-seo/ (mkdir -p first), date-stamped YYYY-MM-DD-*.md

    line 11
  2. Distinctive:

    Add the booking link → the site's scheduling questionnaire

    line 38
  3. Distinctive:

    Start the post cadence (section 3)

    line 41
  4. Distinctive:

    No fabricated claims, locations, or customer quotes.

    line 53
  5. Distinctive:

    Check via web search with the location modifier IN the query.

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

GBP + Local SEO Manager

Drive map-pack visibility for local-service clients. Without GBP API access this skill PREPARES everything — audits from public data, post copy, ranked fix lists — as paste-ready deliverables the operator (or the client) executes in the GBP UI. Items marked [automation] activate only once API access exists.

Paths

  • All outputs → clients/<slug>/deliverables/local-seo/ (mkdir -p first), date-stamped YYYY-MM-DD-*.md
  • Canonical business facts (name, address, phone, services, service areas) come from the client record at clients/<slug>/ — never guess them

1. GBP audit — completeness score (100 pts)

Pull what's publicly visible (web search + fetching the Maps listing and brand searches). Score:

ItemPtsPass bar
Primary category correct + specific15e.g. "Window installation service", not generic "Contractor"
2-5 secondary categories10Each maps to a real service page on the site
Services listed WITH descriptions10Every core service, ~300-char descriptions using local keywords
Photos: ≥25 total, ≥3 added in last 30 days15Real job photos — exterior/team/at-work mix
Hours complete (incl. holiday hours)5—
Attributes set5Family-owned, onsite services, payment types, etc.
Q&A seeded (≥5 owner-posted)10Answer the questions prospects actually ask on calls
Booking/appointment link present10Points at the site's scheduling questionnaire
Website link UTM-tagged5?utm_source=google&utm_medium=organic&utm_campaign=gbp-listing
Description: 750 chars, keyword + city rich, no promo fluff5—
Posts: ≥1 in last 7 days10See cadence below

Output YYYY-MM-DD-gbp-audit-<profile>.md with score, per-item pass/fail, and evidence. Below 70 = "needs work" in client-facing framing. This audit doubles as a retainer proof artifact and a sellable service — pricing is value-based; the operator sets the number.

2. Optimization actions — impact-ranked order

  1. Fix the primary category (biggest single ranking lever)
  2. Fill services + descriptions (feeds "provides: X" justifications in the map pack)
  3. Photo backlog: batch 10-15 real job photos, then 3+/month steady
  4. UTM-tag the website link (otherwise GBP traffic hides as "direct" in analytics)
  5. Add the booking link → the site's scheduling questionnaire
  6. Seed Q&A (owner asks + answers; mine questions from the client's actual customer calls)
  7. Secondary categories, attributes, holiday hours
  8. Start the post cadence (section 3)

Deliver as a checklist the operator or client works through in the GBP UI. [automation: push via API when access exists.]

3. Recurring GBP posts — 2/week per profile

One profile = one track; sibling businesses never share a post. Rotation:

  • Job showcase (weekly): 1-2 real photos of that client's actual completed work + 2-3 sentences (what/where-ish/outcome) + a "Get a free quote" CTA on a UTM link (utm_campaign=gbp-post)
  • Offer / seasonal / FAQ-answer (weekly, rotate): service spotlight, regional-weather angle, or an answered customer question

Hard rules:

  • Real photos only. Source from the client's asset folders or photos the client sends. NEVER present stock or AI imagery as a completed job.
  • Regulated offers = gate. Posts touching financing or other regulated claims need the operator's compliance sign-off; default is not to mention financing in GBP posts at all.
  • No fabricated claims, locations, or customer quotes. City names only where the job actually was.

Output a month of drafts to deliverables/local-seo/YYYY-MM-gbp-posts-<profile>.md (post text + which photo file + CTA URL) for pasting into the UI. GBP-only — site-wide social content belongs to social-media-manager/content-calendar.

4. Citation / NAP consistency sweep

  1. Confirm the canonical NAP (exact name, address format, phone) from the client record, or ask the operator. Never guess.
  2. Web-search loop per directory: "[business name]" [city] and "[phone number]" against Google Maps, Apple Maps, Bing Places, Yelp, Facebook, BBB, Angi, HomeAdvisor, Houzz, Nextdoor, Yellowpages.com, Foursquare, MapQuest, and the local chamber of commerce. Also run a bare phone-number search to catch stragglers/duplicates.
  3. Record per directory: listed? NAP exact-match? wrong/old data? duplicate listings?
  4. Output YYYY-MM-DD-nap-sweep-<slug>.md: findings table + prioritized fix list (wrong phone > wrong address > name variants > missing listings) with the edit/claim URL for each. Fixes are manual claims/edits the operator or client submits.

5. Local rank tracking — no paid tools

  • Query set per profile: 5-10 "[service] [city]" terms derived from the client's service and service-area pages
  • Check via web search with the location modifier IN the query; record whether the client appears, roughly where, and who holds the top-3 map pack
  • Be honest about precision: search results are not geolocated like a phone standing in the client's city, there is no geo-grid, and personalization makes single checks noisy. Treat results as directional trend data (present/absent, top-3 vs page-2), never exact positions — and say so in the deliverable.
  • Log to deliverables/local-seo/rank-log.md (append a dated block per check; monthly cadence). If the operator needs defensible precision for a client report, flag that a real geo-grid tool is the upgrade path — a decision, not a silent workaround.

6. Review velocity

Reviews are a top-3 map-pack factor but NOT this skill's job. Route responses, request timing, and reputation reports to review-manager. In the audit, record only review count + rating + recency as scoring context and note "run review-manager" if velocity is weak.

7. Write-back (non-negotiable)

After every audit, sweep, post batch, or rank check: log an activity note in the client record (date + 1-2 sentence summary + deliverable path) and update the matching <TASK SYSTEM> task. Do the writes and verify the files/task actually changed — never claim writes that didn't happen.

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.