Skip to content
J·BJon BishopFractional growth marketer

Data platforms · llms.txt file breakdown

Count llms.txt

A directly retrieved Markdown index for Count's documentation, with 709 links across 5 detected headings.

The actual file

Source Markdown

The saved llms.txt text, displayed as source. The file belongs to its publisher; the live endpoint may change.

Open saved text ↗

Showing the first 18000 characters. Load the complete file to read the remainder.

# Count

> Count is a collaborative platform for AI-powered analytics and research. Connect it to Claude, ChatGPT, or any assistant that supports MCP, and it splits an AI answer into individual claims, draws the reasoning chain behind each one, and flags hallucinations and assumptions, so you can check the work before you share it.

AI assistants make things up: an outdated fact, an invented citation, or true sentences that add up to an unsupported conclusion. Count shows why an answer is what it is, so people can trust it and share it. Count is also a full analytics canvas: it connects to data warehouses and business apps, runs SQL and Python, and has its own agent.

- Website: https://count.co
- Full text of the pages below: https://count.co/llms-full.txt

## What is Count?

- [What is Count?](https://count.co/what-is-count)
  Count is a collaborative platform for AI-powered analytics and research. Connect it to Claude or ChatGPT and it shows the reasoning chain under every claim, so you can check the answer before you share it.
- [What is a reasoning chain?](https://count.co/what-is-count#reasoning-chain)
  A reasoning chain is a map of how an AI reached a claim. Count draws it as a graph under the claim: the prompt, what the model already believed, each search, query, or calculation, and the source the claim rests on. When a step is wrong or missing, the chain shows where.
- [How do I check whether ChatGPT’s or Claude’s analysis is right?](https://count.co/what-is-count#check-ai-answers)
  Check the claims, not the whole answer. Split the response into the individual statements it depends on, then confirm what each one rests on: a source you can open, a query you can re-run, or an assumption you can test. Count does this for you and links every claim to its evidence.
- [What is Count’s MCP server?](https://count.co/what-is-count#mcp-server)
  Count's MCP server connects your AI assistant to Count. Add it once to Claude, ChatGPT, or any app that supports the Model Context Protocol, then ask Count to audit an answer from inside the conversation. Count replies with a link to a canvas holding every claim and its reasoning chain.
- [What else does Count do?](https://count.co/what-is-count#canvas)
  The canvas is where the audit lives, and it is a full analytics workspace. Count connects to your data warehouse and business apps, runs SQL and Python, and has its own agent. Teams use it for exploratory analysis, reports, dashboards, and decks, all checked and edited together in real time.

## Product

- [Count](https://count.co)
  Count is a collaborative platform for AI-powered analytics and research, where every answer earns your trust.

## Pricing

Simple pricing for AI you can trust. Viewers free, and unlimited AI credits included.

Full pricing details: https://count.co/pricing

### Free - $0
See what your team and AI can find together. No card required.
- 1 project, 3 canvases
- Up to 5 users
- Daily AI credit limits (limited models)
- 1 database/MCP connection, web search, file uploads
- 2 automations
- Community support

### Pro - $49/editor / month
Everything a small team needs to dig into their business.
Everything in Free, plus:
- Unlimited AI credits*
- Unlimited canvases, projects, reports & automations
- 3 database/MCP connections
- Shared agent skills & context
- 4GB compute capacity (Python VM & dataset extract size)
- Slack integration and MCP client
- Email, Slack or Teams support

### Enterprise - Let’s talk
Run Count at scale across your organization.
Everything in Pro, plus:
- Unlimited editors
- Choose your own models or bring your own API keys
- Unlimited database connections
- Governed semantic layer & admin change approvals
- Up to 32GB compute capacity
- REST API
- Workspace telemetry
- User groups, row-level security & embedded analytics
- SSO, SCIM, HIPAA, EU/US data residency


## Comparisons

How Count compares to other approaches:

- [Count vs AI Workspaces](https://count.co/compare/count-vs-ai-workspaces)
  Count vs AI workspaces like Dust and Langdock: agents, skills, collaboration, usage limits and checking AI answers, compared for research teams.
- [Count vs ChatGPT](https://count.co/compare/chatgpt)
  Count vs ChatGPT: check ChatGPT's research claim by claim. See which claims are sourced, assumed or unsupported, without re-prompting for each one.
- [Count vs Chatbots](https://count.co/compare/count-vs-chatbots)
  Count vs AI chatbots: keep asking Claude or ChatGPT, then see the reasoning chain behind every claim and flag hallucinations before you share it.
- [Count vs Claude](https://count.co/compare/claude)
  Count vs Claude: keep using Claude, then check every claim in Count. See sources, assumptions and logic gaps without prompting for them each time.
- [Count vs Dust](https://count.co/compare/dust)
  Count vs Dust: agents, skills, collaboration, Frames vs live reports, credits vs unlimited AI, and checking every claim an AI agent makes.
- [Count vs Gemini](https://count.co/compare/gemini)
  Count vs Gemini: how Gemini's Double-check compares with Count's reasoning chains for checking AI research, plus collaboration and usage limits.
- [Count vs Hex](https://count.co/compare/hex)
  Count vs Hex: Hex for heavy Python and ML; Count for AI answers you can check, with a reasoning chain under every claim and lower warehouse costs.
- [Count vs Langdock](https://count.co/compare/langdock)
  Count vs Langdock: EU governance vs checked answers. Agents, collaboration, usage limits and claim-by-claim checking of AI research compared.
- [Count vs Looker](https://count.co/compare/looker)
  Is Count the best Looker alternative? Governed reporting vs AI answers you can check: reasoning chains, warehouse cost, MCP support, LookML import.
- [Count vs Microsoft Copilot](https://count.co/compare/microsoft-copilot)
  Count vs Microsoft 365 Copilot: go beyond file citations. Check AI research claim by claim and work through it together on one shared canvas.
- [Count vs Tableau](https://count.co/compare/tableau)
  Count vs Tableau: established dashboards vs AI answers you can check, with a reasoning chain under every claim and stakeholders in the analysis.
- [Count vs Thoughtspot](https://count.co/compare/thoughtspot)
  Count vs ThoughtSpot: search-style AI answers vs AI analysis you can check, with a reasoning chain under every claim. Where each fits, plus pricing.

## Integrations

Count connects to the following data sources and tools:

- [Apollo.io](https://count.co/integration/apollo): Bring your Apollo.io sales intelligence data into Count. Track prospects, outreach performance, and pipeline metrics with natural language queries.
- [Asana](https://count.co/integration/asana): Connect Asana to Count and analyze your project data effortlessly. Ask questions about tasks, team performance, and deadlines in plain English.
- [Attio](https://count.co/integration/attio): Connect Attio CRM to Count and explore your customer data with AI-powered analytics. Track deals, contacts, and sales performance instantly.
- [Chargebee](https://count.co/integration/chargebee): Bring your Chargebee billing data into Count. Analyze revenue, subscriptions, and churn metrics with natural language - setup takes minutes.
- [Customer.io](https://count.co/integration/customerio): Explore your Customer.io marketing data with Count's AI analytics. Track campaigns, user engagement, and conversion metrics effortlessly.
- [GitHub](https://count.co/integration/github): Connect GitHub to Count and analyze your development data. Track pull requests, commits, and team productivity with natural language queries.
- [Google Ads](https://count.co/integration/google-ads): Bring your Google Ads data into Count for instant insights. Analyze campaigns, spend, and performance metrics with AI-powered natural language.
- [Google Analytics](https://count.co/integration/google-analytics): Connect Google Analytics to Count and explore your website data effortlessly. Ask questions about traffic, conversions, and user behavior.
- [Granola](https://count.co/integration/granola): Connect Granola to Count and analyze your meeting data with natural language. Track call outcomes, engagement, and sales insights instantly.
- [HubSpot](https://count.co/integration/hubspot): Bring your HubSpot CRM data into Count. Ask questions about deals, contacts, and marketing performance in plain English - quick setup.
- [Intercom](https://count.co/integration/intercom): Connect Intercom to Count and explore your customer support data. Track conversations, response times, and satisfaction with natural language.
- [Jira](https://count.co/integration/jira): Bring your Jira project data into Count. Analyze sprints, issues, and team performance with AI-powered analytics and natural language queries.
- [Klaviyo](https://count.co/integration/klaviyo): Connect Klaviyo to Count and explore your email marketing data. Track campaigns, customer segments, and revenue with natural language analytics.
- [Linear](https://count.co/integration/linear): Bring your Linear project data into Count. Analyze issues, cycles, and team velocity with AI-powered insights and natural language queries.
- [Monday.com](https://count.co/integration/monday): Connect Monday.com to Count and explore your project data effortlessly. Track boards, tasks, and team progress with natural language analytics.
- [Notion](https://count.co/integration/notion): Connect Notion to Count and analyze your workspace data with natural language. Track page views, content performance, and team activity effortlessly.
- [PostHog](https://count.co/integration/posthog): Bring PostHog data into Count for AI-powered product analytics. Ask questions about user behavior, funnels, and feature adoption in plain English.
- [Pylon](https://count.co/integration/pylon): Connect Pylon to Count and explore customer support metrics with natural language. Track ticket volumes, response times, and satisfaction instantly.
- [Ramp](https://count.co/integration/ramp): Connect Ramp to Count and explore your expense data with AI. Track spending patterns, budget performance, and financial KPIs with natural language queries.
- [Salesforce](https://count.co/integration/salesforce): Bring Salesforce CRM data into Count. Ask questions about deals, pipeline performance, and customer relationships in plain English. Setup takes minutes.
- [Shopify](https://count.co/integration/shopify): Connect Shopify to Count for AI-powered eCommerce insights. Track sales, conversion rates, and customer behavior with natural language queries.
- [Slack](https://count.co/integration/slack): Connect Slack to Count and analyze team communication patterns. Track message volume, channel activity, and engagement metrics with natural language.
- [Stripe](https://count.co/integration/stripe): Connect Stripe to Count and explore your payment data with natural language. Track revenue, churn, and subscription metrics with AI-powered analytics.

## Blog

- [The trust trap: how showing AI reasoning helps but also hinders](https://count.co/blog/the-trust-trap)
  Chas Nelson explains why AI reasoning can mislead—and how checkable reasoning chains help teams challenge outputs and build trust.
- [How We Talked AI Into a UFO Cover-Up](https://count.co/blog/why-does-ai-turn-a-hunch-into-a-finding)
  See how loaded prompts create AI feedback loops—and learn to verify sources, inspect reasoning and prevent polished hallucinations from driving decisions.
- [Why Does Claude Become Confidently Wrong?](https://count.co/blog/why-does-claude-become-confidently-wrong)
  Learn why Claude becomes confidently wrong in long chats—and how to spot hallucinations, verify citations and prevent conversational drift from misleading your research.
- [How AI's confidence bubble affects decision-making](https://count.co/blog/confidence-bubble)
  Oliver Hughes explains how AI’s confidence bubble distorts decisions—and why visible, traceable reasoning helps teams verify claims and act with trust.
- [How Bad Are AI Hallucinations Really?](https://count.co/blog/how-bad-are-ai-hallucinations-really)
  Explore 8 ways AI answers fail—from fabrication to stale knowledge—and learn how to verify claims before using them in high-stakes work.
- [The hidden AI hallucination: when right facts make a wrong decision](https://count.co/blog/the-hidden-ai-hallucination)
  Chas Nelson explains how AI can use true facts to reach wrong conclusions—and how to spot flawed reasoning before it drives business decisions.
- [The trust gap: why AI works for code but not (yet) for the rest of your work](https://count.co/blog/the-ai-trust-gap)
  Chas Nelson explains why AI thrives in coding but struggles in knowledge work—and how reviewing its reasoning can close the trust gap.
- [Meet the new Count Agent ](https://count.co/blog/meet-the-new-count-agent)
  Oliver Hughes shares how the new Count Agent creates polished charts, auto layouts, and web-powered analysis to make deep research faster and more accessible.
- [Python in Count has had a major upgrade](https://count.co/blog/python-in-count-has-had-a-major-upgrade)
  Run powerful Python in Count: install any package, handle big data, automate workflows, and let the AI agent write, run, and schedule your Python analyses.
- [Three new ways to teach, manage, and improve Count's agent.](https://count.co/blog/new-context-features)
  Oliver Hughes explains three new Count features that refine agent context, automate updates via API, and use telemetry to continuously improve answers.
- [MCPs as data sources](https://count.co/blog/mcps-as-data)
  Oliver Hughes explains how MCP servers become first-class data sources in Count, unifying qualitative and quantitative insights across your entire stack.
- [Introducing @Count for Slack: your data in the conversation](https://count.co/blog/introducing-count-for-slack-your-data-in-the-conversation)
  Bring live data into Slack conversations. Ask @Count in any channel and get instant answers, charts, and a canvas to go deeper—without leaving the thread.
- [Semantic interoperability](https://count.co/blog/semantic-interoperability)
  Oliver Hughes explains how Count auto-converts LookML, Snowflake, and OSI models into a unified semantic layer, eliminating rewrites and keeping metrics in sync.
- [Introducing Count's API and MCP Server](https://count.co/blog/introducing-api-mcp)
  David Tomasoni-Major shares how Count’s API and MCP server automate workflows and bring trusted, governed data into the tools your teams use every day.
- [Why AI analysis tools need to produce artifacts, not transcripts](https://count.co/blog/why-ai-analysis-tools-need-to-produce-artifacts-not-transcripts)
  David Tomasoni-Major explains why AI should create shareable analysis artifacts—dashboards, reports, and canvases—instead of endless text transcripts.
- [How to successfully deploy an analytics agent in 2 weeks or less](https://count.co/blog/how-to-successfully-deploy-an-analytics-agent)
  Holly Traynor shares a practical three-phase plan to deploy a trusted analytics agent in two weeks or less.
- [Giving AI space to breathe](https://count.co/blog/introducing-the-count-agent)
  Oliver Hughes explains how Count’s new AI agent delivers transparent, collaborative analytics at scale with powerful compute and governance for trusted decisions
- [How to integrate Google Analytics with Count for better marketing dashboards](https://count.co/blog/how-to-integrate-google-analytics-with-count)
  David Tomasoni-Major explains how to turn GA4 data in your warehouse into flexible, self-serve marketing dashboards using Count’s semantic layer.
- [10 Best Business Intelligence Tools for Self-Serve, Data Accessibility, and no-code analysis [2025]](https://count.co/blog/10-best-business-intelligence-tools-for-self-serve-data-accessibility-and-no-code-analysis-2025)
  10 best BI tools compared on true self-service, data accessibility, and no-code analysis—see which platforms actually let non-technical users generate real insights.
- [10 Best AI Analytics Agents for BI [2005]](https://count.co/blog/10-best-ai-analytics-agents-for-bi-2005)
  Compare 10 leading AI BI agents on trust, security, auditability, collaboration, and analysis depth—see which tools actually deliver reliable, safe AI analytics.
- [The three most important jobs for analytics teams in an AI-first world](https://count.co/blog/the-three-most-important-jobs-for-analytics-teams-in-an-ai-first-world)
  Discover the three critical roles analytics teams must play to stay essential and drive better decisions in an AI‑first world.
- [The types of canvas that matter most](https://count.co/blog/the-three-types-of-canvas-that-matter-most)
  Unlock operational clarity with Count's dynamic canvases: Metric Maps, Collaborative Canvases, and Explorations. Transform data into actionable insights and foster team collaboration.
- [Chatbots aren’t the answer](https://count.co/blog/chatbots-arent-the-answer)
  Why chatbots and dashboards fail for analytics—and why a collaborative, flexible, transparent canvas is the right interface for AI-first data work.
- [10 Best Business intelligence Tools for Collaborative Data Analysis and Business Improvement [2025]](https://count.co/blog/best-business-intelligence-tools-for-collaborative-data-analysis-and-business-improvement-2025)
  Oliver Hughes explains how collaborative BI tools like Count turn data analysis into a real team sport, with live co-editing, rich context, and stakeholder-friendly workflows.
- [What Will “Analyst 2.0” Look Like?](https://count.co/blog/what-will-analyst-2-0-look-like)
  Discover the future of data analysis with "Analyst 2.0": bridging insights to action, enhancing decision-making, and redefining roles for impactful, data-driven success.
- [The Tableau Era is Over](https://count.co/blog/the-tableau-era-is-over)
  Tableau's era ends; embrace data canvases for storytelling and collaboration. Move beyond dashboards to empo

Measured from the saved file

Linked document

UTF-8 text size
94.7 KB97014 bytes
Token count
21723o200k_base encoding
Markdown links
738738 distinct URLs
Fenced code blocks
01016 lines in the file

Size measures the captured UTF-8 text, not compressed transfer size. Tokens are counted with a named tokenizer; other models can count differently. Neither measure establishes context compatibility or answer quality.

Project H1
Count

Opening blockquote

Count is a collaborative platform for AI-powered analytics and research. Connect it to Claude, ChatGPT, or any assistant that supports MCP, and it splits an AI answer into individual claims, draws the reasoning chain behind each one, and flags hallucinations and assumptions, so you can check the work before you share it.

Context architecture

How the content is divided

9 H2 sections organize 738 parsed Markdown links. 0 links point to .md or .markdown paths; 2 link occurrences use another hostname. Linked destinations have not been recursively validated.

H2 sectionLinksUTF-8 bytesTokens
What is Count?Line 1051750399
ProductLine 23114230
PricingLine 2801181283
ComparisonsLine 68122512576
IntegrationsLine 97234570921
BlogLine 125122343067246
WebinarsLine 3783585071943
LearnLine 451220959
OptionalLine 458 · Optional5384295610078

Nested indexes

1 distinct links point to other llms.txt paths. These are routes to more indexes, not proof their contents were included in this file.

Link counts exclude parsed fenced and inline code and resolve relative destinations against the final endpoint. Section sizes cover the text below each H2 until the next H2.

Embedded material

Code, installation and API examples

Fenced blocks
0 closed
Declared languages
No fenced blocks detected
Package-install commands
0 detected in fenced or inline code
Authentication examples
0 fenced lines match API-key, Authorization, Bearer, OAuth or access-token patterns
HTTP request examples
0 fenced lines match request-method patterns
Authentication links
0 links have authentication-related labels

Embedded example text and links to further documentation are counted separately. A code block does not establish that a sample runs or that an authentication flow succeeds.

Publisher prioritization

The Optional section

1 H2 section named “Optional” contain 538 parsed links. 200 links sit outside those sections. That puts 72.9% of the file’s parsed links under Optional.

The community proposal uses Optional by convention for secondary links that can be skipped for a shorter context. This label does not establish that everything outside it is critical, or that the material inside it is deprecated.

Machine-readable routes

MCP references and configuration text

MCP-related text was detected in the captured file. 3 links have MCP-related labels or paths. 0 fenced lines match mcpServers, mcp_servers or an mcp add command.

These are text and link observations. No MCP server connection, authentication or tool call was tested.

Index and bundle

llms-full.txt companion

Verified text companion

Text companion verified

https://learn.count.co/llms-full.txt ↗
Discovery
Linked from this llms.txt
HTTP response
200 · text/plain; charset=utf-8
Text size
462.1 KB · 473145 bytes (measured)
Token count
97752 o200k_base tokens
Retrieval scope
Complete body retrieved

Text companion verified

https://count.co/llms-full.txt ↗
Discovery
Conventional sibling path checked
HTTP response
200 · text/plain; charset=utf-8
Text size
101.6 KB · 104007 bytes (measured)
Token count
23216 o200k_base tokens
Retrieval scope
Complete body retrieved

A verified text response establishes a companion at the checked URL. It does not prove the bundle contains all product documentation, that generation is automated, or that it fits an assistant’s context window.

Change evidence

What can be said about freshness

This capture establishes a baseline. Update frequency cannot be determined from one observation.

Linked release and migration material

4 links have release, changelog, deprecation or migration terms in their labels or paths. These provide places to investigate product changes separately.

File changes can reveal documentation edits or navigation changes. They do not by themselves measure product release velocity, accuracy, agent adoption or AI visibility.

Capture identity and measurement method
Requested endpoint
https://count.co/llms.txt
Final endpoint
https://count.co/llms.txt
HTTP status
200
Content type
text/plain; charset=utf-8
Text retrieval
Firecrawl literal text capture
Tokenizer
tiktoken 0.14.0 · o200k_base
SHA-256 of saved UTF-8 text
427e64fedef05a6d5215c4cf4acda0485c6ee89c807322c8b3c1bec4987cc3c1

Counts use a deterministic parser, not a spec-compliance grade. Unusual Markdown and indented code can fall outside its recognized patterns. See the llms.txt proposal and tokenizer implementation.

Related tools and resources

Contact Jon