SToneX
fd4453ea17
Preserve public page sources and portable maintenance tools while moving hosted records and pricing evidence out of the public tree. Require explicit external pricing evidence and remove hosted database access helpers. Update the catalog, ads conversion, archive, super-admin, data-model, MCP OAuth, API, SEO and skill context fragments. Merge the companion private import before this change.
12 KiB
Facts registry — every number allowed on a landing page
Rule: if a figure is not in this table, it may not appear on a page. Each page ends with a Numbers used on this page table naming its fact keys, so re-verification is mechanical.
Verified 2026-08-17 unless stated. Re-verify with:
curl -s https://treg.to/catalog/platforms | python3 -c "..." # counts
treg catalog get <endpoint_id> # price + measured stats
Product and pricing
| Key | Value | Source | Verified |
|---|---|---|---|
F-01 |
2,600+ tools · 40+ providers | treg.to hero (2,617 / 42); rounded down so it holds under every count | 2026-08-17 |
F-02 |
$1.00 free credit once per new verified account | treg.to; llms.txt; application/signup.py; config.py promo_grant_micro = 1_000_000 |
2026-09-10 |
F-03 |
0% additional fee — treg.to does not mark up provider call prices | treg.to ("0% additional fee — we earn on volume pricing with vendors, not on you") | 2026-08-17 |
F-04 |
Credentials injected server-side; response relayed verbatim; every call audited | llms.txt |
2026-08-17 |
F-05 |
Your own key always wins over treg's, and those calls are never metered | llms.txt; treg.to |
2026-08-17 |
F-06 |
treg.to does not choose a provider or fail over for you, by design | llms.txt line 182 |
2026-08-17 |
F-07 |
5xx, timeouts and network errors are never billed; Idempotency-Key prevents paying twice for a retry |
llms.txt |
2026-08-17 |
F-08 |
Sign-in doors: GitHub, Google, email one-time code. No card required to create a team or spend the grant | treg.to sign-in modal; ledger.grant fires at org creation |
2026-08-17 |
F-09 |
Installer registers the MCP server into Claude Code, Cursor and opencode; other MCP clients connect via OAuth; anything that runs a shell can use the CLI | llms.txt |
2026-08-17 |
F-10 |
100% open source, AGPL — github.com/superdesigndev/treg | treg.to footer | 2026-08-17 |
F-11 |
Price is visible before the call (catalog get / estimated_cost_usd) |
llms.txt |
2026-08-17 |
F-12 |
Out of balance returns HTTP 402 with balance_micro, estimated_cost_micro and a top-up URL |
llms.txt |
2026-08-17 |
F-13 |
A team can pin a capability to one provider: treg org pin <capability> --provider <p> |
llms.txt CLI reference |
2026-08-17 |
Why
F-01is a rounded claim, not an exact one. Three surfaces currently disagree: treg.to's hero says 2,617 endpoints / 42 providers,llms.txtsays "~2,600 endpoints across ~48 providers", andGET /catalog/platformssums to 2,363 endpoints across 47 providers over 80 platforms. 2,600+ tools / 40+ providers is Jason's call (2026-08-17) and is safe on the provider half under every count.One loose end on the tool half:
/catalog/platformsis the only surface below 2,600, so if anyone audits the claim that endpoint is what they will hit. Worth checking whether it under-reports (it may exclude own-key or unpriced endpoints) before this number goes into a Google Ads asset — a claim in an ad has to be defensible on the page it points at.Prefer "tools" over "endpoints" in all public copy: per
CLAUDE.md, a tool is what an agent calls, and it is the word a visitor understands without knowing what an API endpoint is.
Subscription list prices (the "instead of" anchors, all from treg.to's catalog grid)
| Key | Value |
|---|---|
F-20 |
Semrush $139/mo · Serpstat $69/mo · SpyFu $39/mo · SerpApi $75/mo · SE Ranking $65/mo · Moz $99/mo · Majestic $50/mo · SEOTesting $40/mo · Supermetrics $69/mo |
F-21 |
Hunter $34/mo · Lusha $49/mo · Apollo $59/seat · Crunchbase $99/mo · Diffbot $299/mo |
F-22 |
X API $200/mo · Postiz $29/mo |
F-23 |
Optmyzr $249/mo · Revealbot $99/mo · Madgicx $55/mo · Adalysis $99/mo · Tailwind $25/mo · BrightLocal $39/mo |
F-24 |
treg.to's own summed example: "$371/mo of seats → cents per call" |
Measured endpoint data — treg catalog get, 2026-08-17
Columns: cost · success rate (sample size) · median latency. — means nobody has called it
enough to measure, which is itself worth showing.
Keyword volume — capability google.keywords.volume, 5 providers (F-30)
| Provider | Cost | Success (n) | Median |
|---|---|---|---|
| Serpstat | $0.0005 / keyword returned | 100% (77) | 1.2 s |
| SE Ranking | $0.00179 / keyword returned | 92.3% (20) | 1.7 s |
| DataForSEO | $0.09 flat / call, 1–1,000 keywords | 98.8% (165) | 3.9 s |
| Google Ads | free with a connected account | — (8) | — |
| Semrush | 10 API units / row; own key only | — (0) | — |
Google organic results — capability google.serp.organic (F-31)
Serpstat $0.0005 · DataForSEO $0.002 · ScrapeCreators $0.00188 · SerpApi $0.015 · Semrush own-key only.
Other SEO endpoints (F-32)
Keyword ideas: DataForSEO $0.012 · SE Ranking $0.00179 · Serpstat $0.0005. Bulk keyword difficulty: DataForSEO $0.01236. Backlinks / authority: Majestic $0.0008 (own key) · SE Ranking $0.002685 · Moz $0.006667 · DataForSEO $0.024. Backlinks capability has 76 matching endpoints.
Find a work email — capability people.email.find (F-40)
| Provider | Cost | Success (n) | Median |
|---|---|---|---|
| Hunter | $0.0245, charged only when an email is found | 100% (582) | 1.4 s |
| LeadMagic | $0.025, free on a miss | 100% (86) | 8.3 s |
| LeadMagic (from a profile URL) | $0.125, free on a miss | 100% (15) | 1.6 s |
Verify an email (F-41)
LeadMagic $0.00625 — 100% over 963 calls, 6.1 s median; the most-called enrichment endpoint on the platform. Hunter $0.01225.
Every email at a company (F-43)
hunter.companies.emails — 100% over 597 calls, 0.4 s median. Returns the known addresses at a domain in
one call. Not previously represented on any page; added to p2 on 2026-08-17.
Enrich a person (F-42)
LeadMagic $0.025 · Lusha $0.2496 · Coresignal $0.392.
Search companies — capability companies.search, 11 providers (F-50)
| Provider | Cost | Success (n) | Median |
|---|---|---|---|
| Akta | free | 100% (198) | 3.2 s |
| Coresignal | free (ids only; the cost lands on collect) | — (1) | — |
| Hunter Discover | free | 100% (24) | 1.7 s |
| The Companies API | $0.0019 / company returned; free with simplified=true |
100% (338) | 0.3 s |
| Lusha | $0.004992 / up to 25 results | — (0) | — |
| LeadMagic | $0.025 / company returned | 70% (10) | 3.2 s |
| Apollo | $0.026 / page returned | — (0) | — |
| Diffbot | $0.0299 / record returned | 100% (5) | 0.7 s |
| PDL | $0.38 / record returned | — (0) | — |
| Crunchbase | own key only, no per-call price | — (2) | — |
200× spread for the same job, and the cheapest is free. This table is the single most defensible asset in this whole cluster.
Company signals and funding (F-51)
Lusha company signals $0.1248 · Lusha signal types (list) free · LeadMagic company funding $0.10 · LeadMagic company search $0.025 · Apollo company enrich $0.026 · Coresignal company enrich $0.392 · ScrapeCreators LinkedIn company page $0.00188.
Ad libraries (F-60)
| Job | Provider | Cost | Success (n) | Median |
|---|---|---|---|---|
| Meta ad library search | ScrapeCreators | $0.00188 / call | 100% (166) | 3.1 s |
| Meta ad library search | Apify | $0.005 / ad returned | 100% (16) | 12.9 s |
| Meta ad library search | Meta (own key) | free; identity verification required | — (1) | — |
| Google Ads Transparency | DataForSEO | $0.0006 (async batch) | — | — |
| Google Ads Transparency | ScrapeCreators | $0.00188 | — | — |
| Google Ads Transparency | SerpApi | $0.015 | — | — |
| TikTok ad library | ScrapeCreators | $0.00188 / call | 100% (16) | 3.5 s |
| TikTok ad library | Apify | $0.003 / result | — | — |
| TikTok ad library | tikhub | $0.001 | none of 7 recorded successful — returns 405 on its documented shape | — |
| LinkedIn ads search | ScrapeCreators | $0.00188 | — | — |
| LinkedIn ads search | tikhub | $0.004 | — | — |
Social platforms — endpoints per platform, GET /catalog/platforms (F-70)
TikTok 163 · Instagram 150 · YouTube 148 · X 147 · LinkedIn 72 · Facebook 58 · Reddit 35 · Threads 17 · Pinterest 5. (Douyin 291, Weibo 75, Zhihu 56, Bilibili 49, Kuaishou 46, Xiaohongshu 40 are also live and are the strongest coverage in the catalog — worth a China-facing variant later, not on these pages.)
Social endpoint prices (F-71)
Reddit search: tikhub $0.001 · ScrapeCreators $0.00188 · JustOneAPI $0.01476. Most tikhub social endpoints are $0.001 per call, including a TikTok profile.
Category totals (F-80)
Social 1,501 endpoints · SEO/AEO 367 · Advertising 205 · Enrichment 88 (companies 47, people 41) · E-commerce 81 · Reviews & Apps 55 · Market data 40 · Community 16 · Developer 10.
Public endpoint observations
These historical per-endpoint figures were exposed by the public catalog. Refresh before reuse.
Measured endpoint stats — use these, they beat the old sample sizes (F-92)
| Endpoint | Calls | ok | p50 |
|---|---|---|---|
tikhub.x.twitter-web-fetch-search-timeline |
3,311 | 1.00 | 2.6 s |
dataforseo.x.serp-google-maps-live-advanced |
2,646 | 1.00 | 1.9 s |
dataforseo.google.serp.organic |
2,075 | 1.00 | 5.5 s |
leadmagic.people.email.verify |
963 | 1.00 | 6.1 s |
dataforseo.x.serp-google-news-live-advanced |
690 | 1.00 | 9.0 s |
hunter.companies.emails |
597 | 1.00 | 0.4 s |
hunter.people.email.find |
582 | 1.00 | 1.4 s |
| Google Trends (dataforseo) | 462 | 0.98 | 6.7 s |
tikhub.x.tiktok-app-v3-fetch-hashtag-search |
436 | 1.00 | 1.3 s |
tikhub.tiktok.search.videos |
389 | 1.00 | 3.3 s |
tikhub.x.reddit-app-fetch-dynamic-search |
384 | 1.00 | 1.0 s |
tikhub.tiktok.user.profile |
365 | 1.00 | 0.4 s |
hunter.people.enrich |
356 | 1.00 | 0.3 s |
scrapecreators.x.v1-google-search |
343 | 1.00 | 1.6 s |
thecompaniesapi.companies.search |
339 | 1.00 | 0.3 s |
scrapecreators.x.v2-instagram-reels-search |
300 | 1.00 | 3.9 s |
scrapecreators.reddit.search.posts |
283 | 0.98 | 8.5 s |
tikhub.tiktok.user.videos |
267 | 1.00 | 0.5 s |
apollo.people.enrich |
263 | 1.00 | 0.3 s |
Endpoints that FAIL — never send a reader to one of these (F-93)
| Endpoint | Calls | State |
|---|---|---|
tikhub.x.linkedin-web-search-people |
313 | every call 4xx |
diffbot extract-analyze |
73 | every call 4xx |
tikhub.x.linkedin-web-search-jobs |
26 | every call 4xx |
tiktok-ads trends-hashtag |
22 | every call 4xx |
lusha decision-makers |
13 | every call 4xx |
x.x.search-posts-recent (official X API) |
208 | ok 0.62 — fails 2 calls in 5 |
x.x.post.create (official X API) |
38 | ok 0.64 |
spyfu.google.domain.competitors |
36 | ok 0.81 |
x.x.search-posts-recentat ok 0.62 againsttikhubat 1.00 over 3,311 calls is the single best proof point in this dataset — the official API is the unreliable option, and we measured it.
The caveat that limits all of the above
The stats pool every org deliberately: there is no per-tenant breakdown, so the observed calls could be forty
teams or four. Our own superdesign org contributes under 40 of them, so the demand is genuinely
external — but its concentration is unknown. Treat this as directional for targeting, not as proof of a
market. The team count needs TREG_ADMIN_TOKEN and one /admin/* call.
Claims that are NOT verified — do not use
- Any customer name, quote, logo or case study. None exist.
- Any security certification (SOC 2, ISO, GDPR compliance statement). None claimed anywhere.
- Any uptime, latency or accuracy guarantee. The catalog reports what has been measured; that is an observation, not an SLA.
- "Access to every provider / every platform." Coverage is specific and varies by platform.
- Automatic routing, fallback or failover. See
F-06— it does not exist and must not be implied. - Anything about how many people or teams use treg.to. Not instrumented.