similarweb-traffic — Claude Code skill: what it does, how to install
similarweb-traffic fetches domain traffic estimates from SimilarWeb: monthly visits, acquisition channels, top countries, and search terms without an API key.
MIT 227 installs ★ 750 updated 2026-10-21
npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a claude-codeWhat similarweb-traffic does
The skill runs a self-contained Node.js script querying SimilarWeb's undocumented public endpoint at https://data.similarweb.com/api/v1/data?domain=<domain>. The script mimics a standard browser user-agent header, strips URL prefixes like https:// or www., and outputs either a formatted Markdown summary or a normalized JSON object with --json.
The resulting payload covers global and category ranks, bounce rates, visit durations, pages per visit, historical visit series, and channel breakdowns including search, direct, paid, and generative AI referrals. Node.js with native fetch support is the only runtime requirement; external npm packages are not needed.
Because this integration relies on an unauthenticated extension endpoint, SimilarWeb can adjust response schemas or throttle access without warning. Low-traffic websites will typically yield synthetic zeroes with an HTTP 200 status code rather than a clear error message.
What it helps with
- estimate monthly visits and audience engagement metrics for a competitor domain
- break down traffic acquisition across organic search, direct entries, and generative AI
- inspect top visitor countries and primary search keywords for a specific URL
- benchmark multiple sites side-by-side using normalized JSON output
Example prompts for the agent
- Check monthly visit estimates and traffic sources for stripe.com
- Compare traffic metrics between notion.so and linear.app
- Pull geographic distribution and top search terms for figma.com
How to install similarweb-traffic
1Skills CLI (Vercel)
npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a claude-code2Manually: put the files in place
~/.claude/skills/similarweb-traffic/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/similarweb-traffic/
1Skills CLI (Vercel)
npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a codex2Manually: put the files in place
.agents/skills/similarweb-traffic/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a cursor2Manually: put the files in place
.cursor/skills/similarweb-traffic/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install OpenClaudia/openclaudia-skills skills/similarweb-traffic2Manually: put the files in place
.github/skills/similarweb-traffic/SKILL.md1Skills CLI (Vercel)
npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a gemini-cli2Platform CLI
gemini skills install https://github.com/OpenClaudia/openclaudia-skills --path skills/similarweb-trafficManually: put the files in place
.devin/skills/similarweb-traffic/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about similarweb-traffic
How to install similarweb-traffic in Claude Code?
Run npx skills add OpenClaudia/openclaudia-skills --skill similarweb-traffic -a claude-code in your terminal.
Does this skill require a paid SimilarWeb subscription?
No subscription or token is needed. The script queries the public web extension endpoint using browser headers directly.
Why does the script show all zeroes for a domain?
SimilarWeb returns HTTP 200 with zeroed metrics whenever its database holds insufficient volume observations for smaller websites.
Source text
SKILL.md · MIT
---
name: similarweb-traffic
description: Fetch website traffic estimates (monthly visits, traffic sources, top countries, keywords, engagement, ranks) for any domain from SimilarWeb. Use when the user asks about a domain's traffic, monthly visits, traffic sources, audience countries, or wants to compare/benchmark sites against competitors.
---
# SimilarWeb Traffic
Fetch traffic-overview estimates for any domain — monthly visits, traffic-source breakdown, top countries, top keywords, engagement metrics, and global/category ranks.
## When to Use This Skill
Use this skill when the user:
- Asks how much traffic a domain gets ("what's the traffic for X")
- Wants monthly visit trends for a site
- Needs a traffic-source breakdown (organic vs direct vs paid vs GenAI, etc.)
- Wants to know a site's audience countries
- Wants to compare/benchmark several domains against each other or competitors
## Usage
No API key required. Self-contained Node script (uses built-in `fetch`, no `npm install`).
```bash
node scripts/fetch-similarweb-traffic.js <domain>
```
Examples:
```bash
# Formatted markdown summary (default)
node scripts/fetch-similarweb-traffic.js chatslide.ai
# Raw normalized JSON (for further processing / comparisons)
node scripts/fetch-similarweb-traffic.js stripe.com --json
```
The domain can be passed bare (`example.com`) or as a URL (`https://www.example.com/pricing`) — it gets normalized (strips scheme, `www.`, and path).
### Comparing multiple domains
Run the script once per domain with `--json` and combine the results into a table. Add a short delay between calls to be polite to the endpoint.
## Output
Default mode prints a markdown summary: a metrics table (monthly visits, ranks, bounce, pages/visit, time on site), a monthly-visits trend line, traffic-source breakdown, top countries, and top keywords.
`--json` mode prints a normalized object:
```json
{
"domain": "stripe.com",
"globalRank": 327,
"categoryRank": 21,
"category": "Finance",
"snapshotMonth": "2026-05",
"visits": 122394040,
"bounceRate": 0.55,
"pagesPerVisit": 3.26,
"timeOnSite": 165.9,
"estimatedMonthlyVisits": [{ "date": "2026-03-01", "visits": 118067753 }],
"trafficSources": { "searchOrganic": 0.48, "direct": 0.23, "...": 0 },
"topCountries": [{ "countryCode": "US", "value": 0.087 }],
"topKeywords": ["..."]
}
```
`visits` is the most recent complete month. Rates (`bounceRate`, all `trafficSources`, `topCountries[].value`) are 0–1 fractions.
## API Details
- Endpoint: `https://data.similarweb.com/api/v1/data?domain=<domain>` — SimilarWeb's **undocumented public data endpoint** (the one their browser extension uses). No auth, no key, free.
- Requires a browser `user-agent` header (the script sends one) — default `curl`/bot agents are blocked.
- Synchronous, returns immediately.
## Important Notes
- **Estimates, not ground truth.** SimilarWeb models traffic; numbers can be off, especially for mid/low-traffic sites. For a domain you own, your own Google Analytics / Search Console data is authoritative — prefer those when available.
- **Unknown domains return zeros, not an error.** SimilarWeb responds HTTP 200 with all-zero/synthetic data when it has no data for a domain. If every visit count is 0, treat it as "no SimilarWeb data."
- **Unofficial endpoint.** It can be rate-limited, change shape, or require auth without notice; this is a gray-area use of SimilarWeb's data outside their paid API.
- Exit codes: `2` network error, `3` non-200 HTTP response.
© OpenClaudia · MIT · OpenClaudia/openclaudia-skills