# Trademark Watch: USPTO New Filings & Similar Mark Alerts (`chaptersoft/trademark-watch`) Actor

Get alerted when someone files or publishes a US trademark that looks or sounds like yours. Fuzzy + phonetic similarity, Nice classes, opposition deadlines, since-last-run, webhook alerts. Official USPTO data, no key. Misses are free.

- **URL**: https://apify.com/chaptersoft/trademark-watch.md
- **Developed by:** [Jeremy Lam](https://apify.com/chaptersoft) (community)
- **Categories:** Lead generation, Developer tools, Automation
- **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks
- **User rating**: No ratings yet

## Pricing

from $8.00 / 1,000 trademark matches

This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events.
Since this Actor supports Apify Store discounts, the price gets lower the higher subscription plan you have.

Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event

## What's an Apify Actor?

An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes.
In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours,
and optionally produces a well-defined JSON output, datasets with results, or files in key-value store.
In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server.

Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md.

## How to integrate an Actor?

If asked about integration, you help developers integrate Actors into their projects.
You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready.

Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run.

For examples already wired to this Actor's own input schema, see the [API](#api) section below.

Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`).

# README

## Trademark Watch: USPTO New Filings & Similar Mark Alerts

Find out when someone **files or publishes a U.S. trademark that looks or sounds like yours**, early enough to oppose it. Give it your marks (and, if you like, Nice classes). It searches the **official USPTO trademark register** for new applications or marks published for opposition, scores every candidate against your mark with **fuzzy, phonetic and pseudo-mark similarity**, and returns only the close ones. Each one comes with its owner, classes, goods and services, status, **opposition deadline** and links to the official record.

- 💸 **$10 per 1,000 matches** ($9 on Starter, $8 on Scale and above). **No start fee, no monthly rental.** Candidates below your threshold, your own filings, filings already delivered, no-match rows and errors are **free**.
- 🔔 **Built to run on a schedule:** **since-last-run** mode remembers what it already delivered, so each filing is reported and charged **once**. An optional **Slack / Discord / Teams / Zapier webhook** alerts you only when something new turns up.
- 🧠 **Similarity a trademark examiner would recognise:** identical, look-alike (edit distance + Jaro-Winkler), **sound-alike** (phonetic key: `NYKE`≈`NIKE`, `KWIK`≈`QUICK`, `XEROX`≈`ZEROX`), contains-your-mark (`PVQUICK`), shared dominant word (`SOL GLOW`), and **USPTO's own pseudo-mark readings** (`KWIK TENT` → "QUICK TENT"). Each row has a 0–100 score, a match type and plain-English reasons.
- ⚖️ **Opposition-ready:** watch by **filing date** (the earliest warning) or by **publication for opposition**, including publications USPTO has *already scheduled* in the coming weeks. You get the publication date, the 30-day deadline, the days left, and whether the window is open now.
- 🎯 **Low noise:** Nice-class filter (global or per mark: `"Starbucks | 30, 43"`), live marks only, exclude your own company, a **risk level** (high / medium / low) that combines score and class overlap, and one row per filing even when it matches several of your marks.
- 🏢 **Owner watch:** also list competitors in `owners` to get **every new filing by those companies**.
- 🔓 **No API key, no login, no proxy.** It uses the public USPTO Trademark Search service, politely rate-limited.

### Who it's for

- **Brand owners and founders:** a weekly watch on your own marks instead of a $300+/year watch service.
- **Trademark attorneys and IP agencies:** one saved task per client (`stateKey` per client), with opposition deadlines for the docket.
- **Clearance before filing:** `"watchFor": "all-dates"` runs a quick knockout search for a new name across the live register.
- **E-commerce sellers:** catch look-alike filings in your product classes before they register (and before Amazon Brand Registry trouble).
- **AI agents and automations:** structured JSON with a stable schema, one call per watch list (MCP-friendly).

### Input examples

**Weekly watch on your marks (the typical schedule):**

```json
{
  "marks": ["Starbucks | 30, 43", "Frappuccino"],
  "watchFor": "new-filings-and-publications",
  "sinceLastRun": true,
  "excludeOwners": ["Starbucks Corporation"]
}
```

**Everything published for opposition in your classes over the next 60 days:**

```json
{ "marks": ["Glow"], "niceClasses": ["3", "5"], "watchFor": "published-for-opposition", "lookbackDays": 7, "lookaheadDays": 60 }
```

**Competitor filings, plus a Slack alert:**

```json
{ "owners": ["Nike, Inc.", "Lululemon"], "lookbackDays": 30, "notifyWebhookUrl": "https://hooks.slack.com/services/…" }
```

**Clearance / knockout search for a new name (all dates, strict):**

```json
{ "marks": ["Zentrova | 9, 42"], "watchFor": "all-dates", "minSimilarity": 85, "maxResults": 50 }
```

| Field | What it does | Default |
|---|---|---|
| `marks` | Your marks, one per line. Optional classes per mark: `Mark | 25, 35` or `Mark [25]`. Up to 500 marks per run. | – |
| `watchFor` | `new-filings` (filing date), `published-for-opposition` (Official Gazette date), `new-filings-and-publications` (both), or `all-dates` (knockout search). | `new-filings` |
| `niceClasses` | Only filings in these classes (1–45). Per-mark classes override it. | all |
| `lookbackDays` / `dateFrom` / `dateTo` | Date window. Explicit dates win over lookback. | 7 days |
| `lookaheadDays` | Include scheduled publications up to N days ahead. | 60 |
| `sinceLastRun` | Deliver (and charge) each filing only once across scheduled runs. State is kept in a named key-value store (`trademark-watch-state`) and is pruned after 180 days. | off |
| `minSimilarity` | 50–100. 80 is a good balance; 90+ is near-identical only; 70 is a wide net. | 80 |
| `owners` / `excludeOwners` | Watch every filing by a company / ignore your own filings. | – |
| `liveOnly` | Skip dead (abandoned, cancelled) marks. | on |
| `maxResults` / `maxResultsPerTerm` | Caps on saved (charged) rows, best-ranked first. | 1000 / none |
| `maxCandidatesPerTerm` | How many USPTO candidates are scored per mark (20–2000). | 300 |
| `notifyWebhookUrl` | https webhook for a short alert when there are new matches. | – |
| `jurisdictions` | `US` today (see roadmap). | `["US"]` |

### Output

One dataset row per matching filing (`"type": "match"` or `"owner-filing"`, charged), plus one free `"no-match"` row per watched term with nothing new, so your list always reconciles.

```json
{
  "type": "match",
  "watchedTerm": "Kwik",
  "similarityScore": 90,
  "matchType": "shared-word-pseudo-mark",
  "matchReasons": ["shared-word (via pseudo-mark \"KWIK SWAP\")", "contains-your-mark"],
  "riskLevel": "high",
  "classOverlap": [],
  "serialNumber": "99862674",
  "mark": "KWIKSWAP",
  "markPseudoText": ["QUICK SWAP", "KWIK SWAP"],
  "markDrawingType": "Standard character mark",
  "markImageUrl": "https://tsdr.uspto.gov/img/99862674/large",
  "niceClasses": [35],
  "goodsAndServices": [{ "niceClass": 35, "description": "Advertising and marketing; …" }],
  "filingDate": "2026-06-03",
  "filingBasis": ["Intent to use (Section 1(b))"],
  "status": "PUBLICATION/ISSUE REVIEW COMPLETE",
  "live": true,
  "publishedForOppositionDate": "2026-11-10",
  "oppositionDeadline": "2026-12-10",
  "daysUntilOppositionDeadline": 44,
  "oppositionWindowOpen": false,
  "owner": "GUANGDONG TUOBITUO TECHNOLOGY CORP., LTD",
  "ownerCitizenship": "China",
  "ownerAddress": "NO.4 KNIVES & SCISSORS DISPLAYING PARK, …, CHINA",
  "attorney": "Manali Shah",
  "sourceUrl": "https://tsdr.uspto.gov/#caseNumber=99862674&caseType=SERIAL_NO&searchType=statusSearch",
  "documentsUrl": "https://tsdr.uspto.gov/documentviewer?caseId=sn99862674"
}
```

Rows are sorted by risk, then score, then date. The dataset has **Overview**, **Opposition deadlines** and **Goods & services** views. The key-value store `OUTPUT` record has the windows searched, per-term counts (candidates, matches, already delivered, own filings excluded), notes, any unprocessed terms, and the webhook result.

#### How the score works (watching `NIKE`, `QUICK`, `GLOW`)

| Match type | Example | Typical score |
|---|---|---|
| identical | `N.I.K.E`, `ChapterSoft` vs `CHAPTER SOFT` | 100 |
| identical via USPTO pseudo-mark | `QUIX` (USPTO reads it as "QUICK") | 97 |
| sound-alike | `NYKE`, `NIKA`, `ZEROX` vs `XEROX`, `KWIK` vs `QUICK` | 85–92 |
| contains-your-mark | `PVQUICK`, `HEREWEGLOW BEAUTY` | 82–94 |
| shared-word | `SOL GLOW`, `NIKE AIR` | 82–90 |
| look-alike | `GOOGEL`, `SPOTIFLY` | 80–90 |
| below default threshold | `HIKE`, `LIKE`, `STARBURST`, `ROLLS` vs `ROLEX` | < 80 (free, not returned) |

**Risk level:** *high* = score ≥ 90 and a shared class (or no classes given); *medium* = score ≥ 80; *low* = a strong score but in unrelated classes.

### Pricing

Pay per event, no start fee:

| Event | Free plan | Starter | Scale and above |
|---|---|---|---|
| Trademark match (a similar filing, or a filing by a watched owner) | $10 / 1,000 | $9 / 1,000 | $8 / 1,000 |
| Candidates below threshold, already delivered, own filings, no-match, errors | free | free | free |

A weekly watch on 5 marks typically finds 0–20 matches, so it costs **$0.00–$0.20 a week**. Set **Maximum cost per run** to cap spend: the Actor saves the best-ranked matches up to the limit, stops cleanly, and (with since-last-run) picks up the rest next time.

### Tips and limits

- **Generic words are noisy:** a watch on `GLOW` finds dozens of filings a week. Add classes (`Glow | 3`), raise `minSimilarity`, or use `maxResultsPerTerm`.
- **Design marks:** USPTO indexes only the word element and its pseudo-mark. Pure logos without words can be found only via `owners`. `markImageUrl` shows the drawing.
- **Data freshness:** USPTO loads new applications into Trademark Search daily, usually a few days after filing. Since-last-run re-scans a 14-day overlap (7 days for publications), so late-loaded records aren't missed and aren't charged twice.
- **Opposition deadline** = publication date + 30 days (the initial period). Extensions are requested via ESTTA; check TSDR (`sourceUrl`) before relying on a date.
- **Speed:** about 1 USPTO request per second per mark (polite rate limit), so 100 marks take about 2–3 minutes.
- This is a monitoring tool, not legal advice. Similarity scores approximate a likelihood-of-confusion review; a trademark attorney should assess any conflict.

### Data source and fair use

Data comes from the **USPTO Trademark Search** system (tmsearch.uspto.gov), the official public U.S. trademark register. U.S. government records are public domain. The Actor uses no login and no proxy, sends one request at a time (≥ 1 s apart) with a descriptive User-Agent, and never loads TSDR document pages. Not affiliated with or endorsed by the USPTO.

### Use with AI agents and MCP

- Call it from Claude, Cursor or any MCP client through the Apify MCP server (`https://mcp.apify.com`) as `chaptersoft/trademark-watch`, or over HTTP with `POST https://api.apify.com/v2/acts/chaptersoft~trademark-watch/run-sync-get-dataset-items`.
- Minimal agent input: `{"marks":["YourBrand | 25"],"watchFor":"new-filings","sinceLastRun":true}`.
- Pricing is per event with no start fee, so an agent can run it on a fixed budget. Set **Maximum cost per run** to cap spend.
- Output rows are flat JSON with a stable schema, and the `OUTPUT` record in the default key-value store gives a one-object run summary.
- Works with Apify schedules, webhooks and the Zapier, Make and n8n integrations for no-code alerts.

### FAQ

**Do I need a USPTO account or API key?** No. It reads the public USPTO Trademark Search service with no login, key or proxy.

**How is this different from a paid trademark watch service?** You set it up yourself, run it on an Apify schedule and pay only for matches ($10 per 1,000), with no annual fee. It covers U.S. filings today; EU, UK and WIPO are on the roadmap.

**Will I be charged again for the same filing next week?** Not with `sinceLastRun` on. Each filing is delivered and charged once; repeats are free.

**Can it find logos?** Only through their word element or USPTO's pseudo-mark. For pure design marks, watch the company with `owners`.

**Can I watch for several clients?** Yes. Save one task per client and give each its own `stateKey`, so their alert memory stays separate.

**Is the score legal advice?** No. It approximates a likelihood-of-confusion check to flag filings worth a look. Have a trademark attorney assess any real conflict.

### Roadmap

- **EUIPO (EU trade marks)** via EUIPO's official API, **UK IPO**, **WIPO Madrid** and **Canada (CIPO)** open data. TMview is deliberately not scraped, because EUIPO's terms reserve its data against automated extraction.
- Design-code (logo) watch and goods/services keyword watch.

Missing something you need? Open an issue on the Actor page.

# Actor input Schema

## `marks` (type: `array`):

Your brand names / word marks. We find new USPTO filings that look alike, sound alike, contain your mark or share its dominant word. Optional per-mark Nice classes after a bar: "Starbucks | 30, 43".

## `watchFor` (type: `string`):

New filings = applications filed in the date window (alerts arrive the day after filing). Published for opposition = marks published (or scheduled to publish) in the Official Gazette, with your 30-day opposition deadline. All dates = one-off knockout / clearance search across the whole register.

## `niceClasses` (type: `array`):

Only report filings in these international (Nice) classes, e.g. \["25", "35"]. Empty = all 45 classes. Per-mark classes in "Marks to watch" override this.

## `lookbackDays` (type: `integer`):

Date window when no explicit dates are given: filings from the last N days (publications: from N days ago up to “Look ahead” days ahead). Use 1–7 for a daily/weekly schedule.

## `lookaheadDays` (type: `integer`):

Publications are scheduled in the Official Gazette weeks in advance. Include publications up to this many days in the future so you see them before the 30-day opposition window opens.

## `sinceLastRun` (type: `boolean`):

Remembers what was already delivered (in a named key-value store in your account, "trademark-watch-state") and only returns — and charges for — filings you have not seen yet. Ideal for a daily or weekly schedule.

## `minSimilarity` (type: `integer`):

Report filings scoring at least this much against a watched mark. 90+ = near-identical or sound-alike, 80 = one-letter changes / shared dominant word (recommended), 70 = broad net.

## `owners` (type: `array`):

Also report every new filing by these applicants (e.g. a competitor's legal name). Charged like a match.

## `excludeOwners` (type: `array`):

Skip filings by these owners so your own applications never show up (and are never charged).

## `dateFrom` (type: `string`):

YYYY-MM-DD. Overrides "Look back".

## `dateTo` (type: `string`):

YYYY-MM-DD. Default: today (publications: today + 60 days).

## `liveOnly` (type: `boolean`):

Skip abandoned, cancelled and expired records.

## `maxResults` (type: `integer`):

Hard cap on saved (charged) matches, highest risk first. Your "Maximum cost per run" is always respected too.

## `maxResultsPerTerm` (type: `integer`):

Keep at most this many (best-ranked) matches per watched mark, so one generic word can't use up the run. 0 = no per-term limit.

## `maxCandidatesPerTerm` (type: `integer`):

How many USPTO search hits are fetched and scored per watched term (best first). Raise for very common words or long windows. Candidates that do not match are free.

## `includeNoMatchRows` (type: `boolean`):

One free "no-match" row per watched term with nothing new, so every term is accounted for.

## `stateKey` (type: `string`):

Name for the since-last-run memory. Default: derived from your marks/owners/classes, so changing the watch list starts fresh.

## `jurisdictions` (type: `array`):

Register(s) to watch. Today: US (USPTO). EUIPO, UK IPO, Canada and WIPO Madrid are on the roadmap.

## `notifyWebhookUrl` (type: `string`):

Send a short alert (top 10 matches with links and opposition deadlines) to this https:// webhook when the run finds new matches. Works with Slack and Discord incoming webhooks, Microsoft Teams/Zapier/Make, or your own endpoint (JSON body: text, content, summary). Nothing is sent when there are no matches. You can also use Apify integrations (email, Slack, Google Sheets) on the run.

## Actor input object example

```json
{
  "marks": [
    "Starbucks",
    "Glow | 3",
    "Kwik Fix"
  ],
  "watchFor": "new-filings",
  "niceClasses": [
    "25",
    "35"
  ],
  "lookbackDays": 30,
  "lookaheadDays": 60,
  "sinceLastRun": false,
  "minSimilarity": 80,
  "liveOnly": true,
  "maxResults": 1000,
  "maxResultsPerTerm": 0,
  "maxCandidatesPerTerm": 300,
  "includeNoMatchRows": true,
  "jurisdictions": [
    "US"
  ]
}
```

# Actor output Schema

## `matches` (type: `string`):

Overview table: risk, score, watched term, matching mark, serial, owner, classes, filing date, status, opposition deadline, link.

## `opposition` (type: `string`):

Publication date, 30-day opposition deadline and days left for each match.

## `fullJson` (type: `string`):

Every field including goods/services by class, owner address, attorney, basis and match reasons.

## `summary` (type: `string`):

Date window, per-term candidate and match counts, totals, charged rows, notes, unsaved matches and any terms left unprocessed.

# API

You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup.

## JavaScript example

```javascript
import { ApifyClient } from 'apify-client';

// Initialize the ApifyClient with your Apify API token
// Replace the '<YOUR_API_TOKEN>' with your token
const client = new ApifyClient({
    token: '<YOUR_API_TOKEN>',
});

// Prepare Actor input
const input = {
    "marks": [
        "Starbucks",
        "Glow | 3"
    ],
    "watchFor": "new-filings",
    "lookbackDays": 30
};

// Run the Actor and wait for it to finish
const run = await client.actor("chaptersoft/trademark-watch").call(input);

// Fetch and print Actor results from the run's dataset (if any)
console.log('Results from dataset');
console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach((item) => {
    console.dir(item);
});

// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs

```

## Python example

```python
from apify_client import ApifyClient

# Initialize the ApifyClient with your Apify API token
# Replace '<YOUR_API_TOKEN>' with your token.
client = ApifyClient("<YOUR_API_TOKEN>")

# Prepare the Actor input
run_input = {
    "marks": [
        "Starbucks",
        "Glow | 3",
    ],
    "watchFor": "new-filings",
    "lookbackDays": 30,
}

# Run the Actor and wait for it to finish
run = client.actor("chaptersoft/trademark-watch").call(run_input=run_input)

# Fetch and print Actor results from the run's dataset (if there are any)
print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
for item in client.dataset(run.default_dataset_id).iterate_items():
    print(item)

# 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start

```

## CLI example

```bash
echo '{
  "marks": [
    "Starbucks",
    "Glow | 3"
  ],
  "watchFor": "new-filings",
  "lookbackDays": 30
}' |
apify call chaptersoft/trademark-watch --silent --output-dataset

```

## MCP server setup

```json
{
    "mcpServers": {
        "apify": {
            "type": "http",
            "url": "https://mcp.apify.com/?tools=fetch-actor-details,chaptersoft/trademark-watch"
        }
    }
}
```

The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer <APIFY_API_TOKEN>` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations).

## OpenAPI specification

Download the OpenAPI definition: https://api.apify.com/v2/actors/twC5wpc4vHA9Bqdqo/builds/TecbgnBHB9JGokP6o/openapi.json