llms.txt
Compare with other sites' llms.txt files →# Shruwd
> Shruwd measures how a brand appears in AI-generated answers (Google AI Overviews,
> Google AI Mode and ChatGPT), sets that beside first-party AI-crawler server logs,
> diagnoses why the brand is or is not cited, and recommends specific fixes that
> can be re-checked to prove movement.
Two things are worth knowing before reading anything else, because they change how
every number in this product should be interpreted:
- A metric computed from fewer than ten responses is never shown as a number. The
API returns `{ state: 'insufficient_data' }`, which is not zero and is not a
loading state. Every point estimate carries a 95% interval.
- A metric has not "moved" until two gates pass: a significant change whose 95%
confidence interval excludes zero AND an absolute change of at least 5
percentage points, with a Benjamini-Hochberg correction across the prompt set.
## Getting started
- [Overview](https://shruwd.io/docs/getting-started/overview.md): Shruwd tells you how AI assistants describe your brand, why you are left out, and whether your fix changed anything.
- [Quickstart](https://shruwd.io/docs/getting-started/quickstart.md): Add a brand, add prompts, and get your first measurement of how AI answers describe you.
- [Reading your results](https://shruwd.io/docs/getting-started/reading-your-results.md): Understand mention rate, share of voice and prominence — and when a number is solid enough to act on.
## What you track
- [Brands](https://shruwd.io/docs/tracking/brands.md): A brand is a domain and the prompts, competitors and findings attached to it.
- [Prompts](https://shruwd.io/docs/tracking/prompts.md): Prompts are the questions asked on each engine every cycle. Choosing them well is the highest-leverage setup decision.
- [Competitors](https://shruwd.io/docs/tracking/competitors.md): Competitors set the field your share of voice is measured against. Matching is exact, so the spellings you add are what counts.
- [Connecting Search Console](https://shruwd.io/docs/tracking/connecting-search-console.md): Connect Google Search Console so Shruwd knows which of your pages people actually search for.
## Crawler logs
- [Choosing a log path](https://shruwd.io/docs/crawler-logs/choosing-a-log-path.md): Four ways to send AI-crawler hits to Shruwd. Which one is right depends on whether a CDN answers for your site.
- [Platform log drains](https://shruwd.io/docs/crawler-logs/platform-log-drains.md): Paste an endpoint and a token into your hosting platform log-drain settings. No code to deploy.
- [Cloudflare Worker](https://shruwd.io/docs/crawler-logs/cloudflare-worker.md): Deploy a small Cloudflare Worker that forwards AI-crawler hits to Shruwd, including requests your CDN serves from cache.
- [Cloudflare Logpush](https://shruwd.io/docs/crawler-logs/cloudflare-logpush.md): Point a Logpush job at Shruwd to send your zone request logs with no code to deploy.
- [Server middleware](https://shruwd.io/docs/crawler-logs/server-middleware.md): Forward AI-crawler hits from a Nuxt or Nitro origin — only when no CDN answers for your site.
- [Crawlers tracked](https://shruwd.io/docs/crawler-logs/crawlers-tracked.md): The AI crawlers Shruwd recognises, how each hit is verified against vendor IP ranges, and why some are shown but never counted.
- [Ingest not arriving](https://shruwd.io/docs/crawler-logs/ingest-not-arriving.md): Work through the causes of a crawler log drain that stays on awaiting first logs.
## Findings and fixes
- [How findings work](https://shruwd.io/docs/findings/how-findings-work.md): A finding names a cause, a page and a fix — and tracks whether your change moved the number.
- [Rechecks](https://shruwd.io/docs/findings/rechecks.md): What happens after you mark a fix applied, and how Shruwd decides whether it worked.
- [Findings reference](https://shruwd.io/docs/findings/findings-reference.md): Every diagnostic Shruwd runs, grouped by family, with what triggers each one and what it asks you to do.
## Metrics
- [Metric definitions](https://shruwd.io/docs/metrics/metric-definitions.md): Exactly what mention rate, share of voice, prominence and citation share count — and when each one is undefined rather than zero.
- [Confidence and sample size](https://shruwd.io/docs/metrics/confidence-and-sample-size.md): Why every Shruwd rate carries a range, and why a metric with fewer than ten responses behind it is withheld.
- [When a number has moved](https://shruwd.io/docs/metrics/when-a-number-has-moved.md): Shruwd calls a change movement only when it passes two gates and survives a correction for testing many prompts at once.
## Account
- [Plans and limits](https://shruwd.io/docs/account/plans-and-limits.md): What Crawl, Starter and Pro include, and what happens when you reach a limit.
- [Team and roles](https://shruwd.io/docs/account/team-and-roles.md): Invite people to a brand, choose what each of them can change on it, and remove access when they leave.
- [Billing](https://shruwd.io/docs/account/billing.md): Change plan, track usage against your allowance, and understand what happens when a payment fails.
- [Email notifications](https://shruwd.io/docs/account/email-notifications.md): The weekly digest and the alerts Shruwd sends, who receives them, and how to unsubscribe.
## Build on Shruwd
- [API overview](https://shruwd.io/docs/api/api-overview.md): Base URL, authentication, scopes, rate limits and the error shape shared by every Shruwd API endpoint.
- [Brands](https://shruwd.io/docs/api/brands.md): Create and manage brands — the domain and the prompts, competitors and findings attached to it.
- [Prompts](https://shruwd.io/docs/api/prompts.md): Manage the fixed set of questions Shruwd asks each engine every cycle.
- [Entities](https://shruwd.io/docs/api/entities.md): Manage competitors and the exact aliases that count as a mention, plus the suggestions drawn from real answers.
- [Measurement](https://shruwd.io/docs/api/measurement.md): Trigger measurement cycles and read visibility, trends, crawler activity, the sources the answers cite and the individual answers behind them.
- [Findings](https://shruwd.io/docs/api/findings.md): Read the diagnoses behind your numbers and transition them as you act on each one.
- [TypeScript SDK](https://shruwd.io/docs/api/typescript-sdk.md): A typed, dependency-free TypeScript client for the Shruwd API, with retries and structured errors built in.
- [MCP server](https://shruwd.io/docs/api/mcp-server.md): The Shruwd MCP server: hosted at https://shruwd.io/mcp for Claude, ChatGPT and Grok, or local over npx for Claude Desktop, Claude Code and Cursor.
- [Use in Claude](https://shruwd.io/docs/api/use-in-claude.md): Add Shruwd to Claude as a custom connector, sign in, and ask Claude about your brand in AI answers: visibility, findings, crawlers.
- [Use in ChatGPT](https://shruwd.io/docs/api/use-in-chatgpt.md): Install the Shruwd plugin in ChatGPT, sign in, and ask ChatGPT about your brand in AI answers: visibility, findings, crawlers.
- [Use in Grok](https://shruwd.io/docs/api/use-in-grok.md): Add Shruwd to Grok as a custom connector, sign in, and ask Grok about your brand in AI answers. Or call it from the xAI API.
- [For AI agents](https://shruwd.io/docs/api/for-ai-agents.md): Every page of these docs is available as markdown, plus a curated index, the whole corpus in one file, and a ready-made setup procedure.
## Machine interfaces
- [Full documentation, one file](https://shruwd.io/llms-full.txt): every page concatenated
- [Agent procedure](https://shruwd.io/skill.md): how to set a brand up end to end
- [Pricing](https://shruwd.io/pricing.md): every plan's price and limits, as markdown
- [OpenAPI description](https://shruwd.io/openapi.json): the REST API describes itself
- MCP server, hosted: `https://shruwd.io/mcp` (Streamable HTTP). Sign in with Shruwd (OAuth), or send
`Authorization: Bearer sh_live_...`. Setup for Claude, ChatGPT and Grok: https://shruwd.io/docs/api/mcp-server
- [MCP manifest](https://shruwd.io/.well-known/mcp.json): the endpoint, and how to run `shruwd-mcp` locally
- `@shruwd/sdk` on npm: the typed TypeScript client
- `shruwd-mcp` on npm: the same MCP server, local over stdio
## Notes
- Any docs page is available as raw markdown by appending `.md` to its URL.
- How Shruwd measures, and what it will not claim: https://shruwd.io/methodology
- What Shruwd tracks, the four numbers, and when a change counts as real: https://shruwd.io/features/ai-visibility-tracking
- Findings: why AI answers leave a brand out, the page and change to make, and the recheck that says whether it worked or why it cannot tell: https://shruwd.io/features/findings
- AI crawler logs: which AI crawlers really reach a site, which are turned away, verified from the site's own logs, on every plan: https://shruwd.io/features/ai-crawler-logs
- API, SDK and MCP: run Shruwd from code or from an agent, on every plan: https://shruwd.io/features/api
- The AI crawlers Shruwd identifies, and how each is verified: https://shruwd.io/ai-crawlers
- Shruwd and Profound, side by side, with the date Profound's pricing page was last read: https://shruwd.io/alternatives/profound
- Shruwd and Peec AI, side by side, with the date Peec's pricing page was last read: https://shruwd.io/alternatives/peec-ai
- Product changes are listed at https://shruwd.io/changelog, newest first. Longer pieces are at https://shruwd.io/blog.
- The API base is `https://shruwd.io/api/v1`, authenticated with `Authorization: Bearer sh_live_...`.
Keys are minted in the dashboard and cannot administer the account.