Acronyms and terms

Read this once before the test. Every lesson assumes these definitions.

Listing year

How BotCentral bills: $10 per host per year, debited from the customer's own bc_live_ key when the proven card is written. Edits inside the year and all reads are free; a year nobody renews lapses (listed, unproven) until a publish with a funded key renews it. CiteFleet sends the key only while CITEFLEET_BOTCENTRAL_BILLING is on.

Origin / property

The public website you are trying to get indexed, e.g. https://acme-dating.com. A property is that origin inside CiteFleet.

Grok Dispatcher

The in-app assigner. It creates playbook tasks and pins each one to a specialist bot. Not an xAI subscription and not a ChatGPT plugin.

Bot / callsign

A named agent with one job (Aether, Helios, Nimbus…). The callsign is the short name on badges (AETHER).

Playbook

The fixed runbook of jobs every property follows. Campaign tasks are live instances of playbook steps. Includes the BotCentral listing door.

SPASingle-Page Application

A site that serves one HTML shell and routes in JavaScript. Crawlers that do not send Accept: text/html may get a JSON 404 unless a fallback serves HTML for every public URL.

SPA fallback

The single-page-app bug where real routes return JSON 404 to crawlers that omit Accept: text/html. Aether owns this check.

Accept header

An HTTP header saying what format the client wants. Googlebot and many preview crawlers often omit Accept: text/html. The auditor fetches both ways.

robots.txt

File at /robots.txt that tells crawlers what they may fetch. Must allow AI bots and declare the sitemap.

sitemap.xml

List of public URLs you want indexed. Helios submits it to Google; Nimbus submits it to Bing.

GSCGoogle Search Console

Google’s webmaster door. Verify the domain, submit the sitemap, request indexing. Feeds Gemini and AI Overviews.

URL Inspection

GSC tool that fetches one URL as Googlebot. Used to confirm a 404 is gone after an SPA fallback fix.

AI Overviews

Google’s AI answer boxes on search results. They read the Google index, so GSC + crawlable HTML matter.

IndexNow

A ping protocol. You host a public key file and POST changed URLs to api.indexnow.org so Bing (and Yandex, Naver, Seznam, Yep) recrawl quickly.

Bing Webmaster Tools

Bing’s webmaster door. Feeds ChatGPT search and Copilot. Can be imported from GSC via OAuth.

OAI-SearchBotOpenAI Search Bot

ChatGPT’s search crawler. Allow it in robots.txt. Discovery still depends on Bing plus public links.

GPTBot

OpenAI crawler used more for training/collection than for the live ChatGPT search box. Still allow it unless you have a reason not to.

PerplexityBot

Perplexity’s crawler. No submit form. Allow in robots.txt and earn inbound links.

ClaudeBot

Anthropic crawler. Claude answers also lean on Brave Search. No webmaster portal — links matter.

Googlebot / Bingbot

Official Google and Bing crawlers. If they cannot fetch HTML 200s, nothing downstream (Gemini, ChatGPT-via-Bing) works.

Cite vs index

Indexed = the URL exists in Google/Bing. Cited = an AI answer names or links you. Mentions and press drive citations.

KPIKey Performance Indicator

The four Command tiles: workspace, properties, bots assigned, open tasks.

llms.txt

Optional public file that explains the product in plain text for AI assistants. Helpful, not a substitute for GSC/Bing.

BotCentral

The public catalog at botcentral.org. Assistants read GET /v1/search and GET /v1/site/{domain}. CiteFleet is the only publisher. The Card Inspector is the human page; the machine card is the JSON bots fetch.

Machine card

The BotCentral 1.0 JSON at /v1/site/{domain}. Not a website. Use the inspector at /site/{domain} to read it as a page.

Kill switch

Freezes acts (publish, mentions, spend, autopilot) without stopping observe (audit, monitor, reconcile). Global or per door. Open Monitor to freeze or thaw.

Reconcile / drift

A check that origin, sitemap, BotCentral card, and campaign ticks still agree. Drift means they do not. A tick without a proof URL is invalid.

Origin pack

The files CiteFleet writes into a customer GitHub repo: public/robots.txt, public/sitemap.xml, public/llms.txt, public/.well-known/botcentral.txt, and the IndexNow key file. They only go live after that repo is deployed to the customer domain. CiteFleet does not host them.

Origin proofbotcentral-verify=citefleet-app

How BotCentral knows the origin opted in: that line in an apex DNS TXT record on the domain, or the same line served as plain text at /.well-known/botcentral.txt. CiteFleet checks both with the registry's own rules before publishing (Verify proof) so a missing proof is reported with the exact line to add.

Webhook / deploy hook

Automatic listing. A repository webhook (GitHub) or a signed POST from any CI (deploy hook) tells CiteFleet a deploy happened; CiteFleet re-checks the proof for a few minutes and then lists or refreshes the card. Each property has its own secret, shown once when generated.

User account

A CiteFleet login (Google, X, or email and password). The console, every action, and workspace data sit behind a signed-in session. Customer webhooks and /health do not — they authenticate by signature or stay public.