find-seo-agency#

Drive the ServiceGraph API ( https://api.servicegraph.co) to find, shortlist, and enrich US SEO agencies via the pro_services dataset. The catalog has thousands of US firms tagged with service_provided:seo under industry:marketing_agency.

Always pin both industry:marketing_agency and service_provided:seo. SEO sub-flavors (technical, local, link-building, on-page, ecommerce, B2B, etc.) are not separate tags — the catalog has one seo tag — so sub-flavor specialization is inferred via keyword search across firm text.

Any HTTP client works (curl, fetch, requests). Examples below use curl.

Sibling skills — defer when scope is broader#

If the user wants a multi-service marketing engagement (SEO plus PPC plus content plus social, or a "full-service digital agency"), defer to find-marketing-agency — that skill covers the same catalog with a broader filter and won't over-constrain on SEO.

If the user wants strictly web/app development (build a site, ship a feature), defer to find-web-developer / find-software-developer.

MCP server (preferred for authed calls)#

If your harness has the ServiceGraph MCP server loaded (tools containing servicegraph), prefer those — OAuth 2.1 + PKCE keeps the token in the harness sandbox. Otherwise use the REST flow below.

API surface (dataset id: pro_services)#

Every endpoint requires the bearer ( Authorization: Bearer vk_…). No anonymous tier.

EndpointCostUse it forGET /v1/datasets/pro_services/fields[?include_values=1]freeConfirm seo is in the service_provided value list.GET /v1/datasets/pro_services/check?filter=…freeValidate filter.POST /v1/datasets/pro_services/translate-intentfree{intent} → DSL filter + sanity count.GET /v1/datasets/pro_services/search?filter=…&limit=freeBrief firm cards + per-row unlock hint + total.GET /v1/datasets/pro_services/:apexfreeOne row brief; detail only if unlocked.POST /v1/datasets/pro_services/unlocks10 credits / firm{apexes:[...]} ≤100; atomic; 30-day TTL on detail.GET /v1/me/creditsfreeBalance.

Cost model. Discovery / validation / search / brief reads are free. Detail (url, phone, email, social, address, full platforms map) costs 10 credits per firm and lasts 30 days.

Auth#

vk_* API keys minted in the dashboard. Keep the token out of the LLM context — never read .env* into your context; dispatch via shell.

1. Try the call first through a shell wrapper that sources .env.local: bash

   ( set -a; [ -f .env.local ] && . ./.env.local; set +a;
     curl -sS -H "Authorization: Bearer $SERVICEGRAPH_API_KEY" \
          'https://api.servicegraph.co/v1/datasets/pro_services/fields' )
   
   

2. On 401 prompt the user:

   "Open https://servicegraph.co/profile/api-keys, create a key, and add SERVICEGRAPH_API_KEY=vk_… to .env.local here (or export it). Tell me when done. Please don't paste the key into chat."

3. Retry after the user signals ready.

Filter DSL#

GitHub-search-style.

text

filter   := orExpr
orExpr   := andExpr ("OR" andExpr)*
andExpr  := notExpr (("AND")? notExpr)*    # whitespace = implicit AND
notExpr  := ("NOT" | "-") notExpr | atom
atom     := "(" filter ")" | predicate
predicate:= IDENT op valueOrList | bareword
op       := ":" | "=" | ">=" | "<=" | ">" | "<"
valueOrList := value ("," value)*
value    := IDENT | NUMBER | tagAtEvidence
tagAtEvidence := IDENT "@" ("low"|"medium"|"high")
bareword := IDENT | NUMBER          # → keyword:<bareword>

Four rules that bite: AND binds tighter than OR (use parens); comma list = OR within one predicate; negation is -x or NOT x; bareword = keyword search (quote multi-word phrases).

SEO-flavored examples (validate yours with /check):

text

industry:marketing_agency service_provided:seo
industry:marketing_agency service_provided:seo@high state:TX
industry:marketing_agency service_provided:seo technical
industry:marketing_agency service_provided:seo local
industry:marketing_agency service_provided:seo ecommerce
industry:marketing_agency service_provided:seo b2b saas
industry:marketing_agency service_provided:seo@high rating>=4 review_count_total>=20 has:clutch
industry:marketing_agency service_provided:seo "core web vitals"

Sub-flavor → keyword mapping (catalog has one seo tag):

User asks forAdd as keywordTechnical SEOtechnicalLocal SEOlocalLink-buildinglink-building or linksOn-page SEOon-page or onpageOff-page SEOoff-page or offpageEcommerce SEOecommerceB2B SEOb2bShopify SEOshopifyCore Web Vitals / page speed"core web vitals"

Identifying firms — apex#

Firms are identified by their apex domain ( searchpilot.com, not www.searchpilot.com/about).

Recipes#

A. SEO agency in a state (the baseline)#

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo+state:TX&limit=10
# Present, get pick of 3. "Unlocking 3 = 30 credits, 30-day TTL."
POST /v1/datasets/pro_services/unlocks
  { "apexes": ["firm-a.com", "firm-b.com", "firm-c.com"] }

B. Technical SEO for a SaaS company#

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo@high+technical+b2b+saas&limit=10

C. Link-building + on-page for ecommerce#

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo+link-building+on-page+ecommerce&limit=10

If the breakdown is sparse, drop on-page — agencies that do SEO at all usually do both.

D. Local SEO for a multi-location business#

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo+local+state:NJ&limit=10

The catalog tags one state per firm (HQ); local-SEO firms can serve NJ without being headquartered there. If results are narrow, drop state:NJ and use geography_served:national_US,multi_state_regional.

E. Indirect intent — "organic traffic flat"#

User: "Our organic traffic has been flat for 6 months — we need an SEO partner."

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo&limit=10

Or use the translator:

text

POST /v1/datasets/pro_services/translate-intent
  { "intent": "SEO partner — our organic traffic has been flat for 6 months" }

If the user gave a vertical or location elsewhere, add it. Otherwise present the top-10 and ask for constraints.

F. Quality threshold + third-party signals#

Be cautious — TX seo@high collapses sharply with rating gates or Clutch alone. Layer @high evidence + non-solo size first; add the rating gate only if pool is still large:

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo@high+-company_size_signal:solo+state:TX&limit=10

For users insisting on third-party signals:

text

GET /v1/datasets/pro_services/search?filter=industry:marketing_agency+service_provided:seo@high+rating>=4+review_count_total>=20+has:clutch&limit=10

G. BYO apex list — enrich domains#

User pastes 8–20 SEO firm domains:

1. GET /v1/datasets/pro_services/:apex per domain — free brief (404 = not in catalog, no charge). Note: not every firm with seo in their domain is tagged service_provided:seo.
2. User picks N to fully enrich. POST /unlocks = 10×N credits, atomic, detail returned.
3. Re-runs within 30-day TTL are free.

Gotchas#

• Always pin both industry:marketing_agency AND service_provided:seo. Without the industry pin, service_provided:seo matches SEO services from IT firms or design shops too. Without the service pin, you'd return all marketing agencies regardless of SEO specialty.
• SEO sub-flavors are NOT separate tags. Use barewords for sub-flavor — they become keyword substring matches in firm text.
• Defer to find-marketing-agency for multi-service scope. SEO plus paid plus content plus social as one engagement is a full-service marketing ask.
• Defer to find-web-developer for "fix our site speed / Core Web Vitals" if they want it BUILT. SEO agencies advise on CWV; they don't refactor your front-end.
• Rating/Clutch gates are sparse for SEO. TX seo@high (~243 firms) collapses below k=20 with rating>=4 OR has:clutch alone. Prefer @high + non-solo as the quality proxy and add rating only if the base pool is still wide.
• "SEM" / "PPC" / "paid search" / "Google Ads" are NOT SEO. Defer to find-marketing-agency. SEM-vs-SEO is a frequent terminology mix-up.
• Multi-word phrases must be split or quoted. core web vitals parses as three AND'd keywords; "core web vitals" is one phrase.
• Briefs DO include apex, name, location, ratings. They DON'T include url, phone_primary, email_primary, legal_name, address_full, full platforms — those require an unlock.
• not_found / not_in_dataset 404 = not in pro_services. Skip; not charged.
• Unlock is atomic. N apexes either all charge (up to 10×N credits) or none on 402.
• Within-TTL re-views are free ( was_cached:true).

Errors#

JSON envelope: {"error": {"code": "...", "message": "..."}}.

StatusCodeWhat to do400filter_parse_error``position included; fix and re-validate with /check.400kind_in_filterStrip any kind: from filter.400field_not_in_datasetDrop the disallowed field.400invalid_apexRe-normalize.401unauthorized / invalid_audienceRe-prompt for fresh vk_….402insufficient_credits``needed and balance; nothing charged.404not_found / not_in_datasetSkip; not charged.429rate_limitedHonor Retry-After.

End-to-end example#

User: "Three technical SEO consultancies for a B2B SaaS company, ideally with at least a 4-star rating and a Clutch listing."

text

GET /v1/datasets/pro_services/fields?include_values=1
GET /v1/datasets/pro_services/check?filter=industry:marketing_agency+service_provided:seo@high+technical+b2b+saas+rating>=4+has:clutch
GET /v1/datasets/pro_services/search?filter=...&limit=10
# Present briefs. "Unlocking 3 = 30 credits, 30-day TTL."
POST /v1/datasets/pro_services/unlocks
  { "apexes": ["firm-a.com", "firm-b.com", "firm-c.com"] }
GET /v1/me/credits