Data platforms · llms.txt file breakdown

docs.airbyte.com llms.txt llms.txt

Airbyte is an open source platform designed for building and managing data pipelines, offering extensive connector options to facilitate data movement from various sources to destinations efficiently and effectively.

Open original llms.txt ↗

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.

# docs.airbyte.com llms.txt

> Airbyte is an open source platform designed for building and managing data pipelines, offering extensive connector options to facilitate data movement from various sources to destinations efficiently and effectively.

- [docs.airbyte.com llms.txt](/index.md)

## search

- [Search the documentation](/search.md)

## ai-agents

Airbyte Agents is a data and context layer for AI agents. It gives your agents real-time access to business data through open source, type-safe connectors, managed credentials, and low-latency search. Use it as a cloud platform or import connectors directly into your own agents.

- [Airbyte Agents](/ai-agents.md): Airbyte Agents is a data and context layer for AI agents. It gives your agents real-time access to business data through open source, type-safe connectors, managed credentials, and low-latency search. Use it as a cloud platform or import connectors directly into your own agents.

### admin

Manage your Airbyte Agents account. This section covers billing, including plans, payments, usage monitoring, invoices, and subscription changes. On the Team and Custom plans, you can also manage users and configure single sign-on. If your network restricts inbound traffic, see IP allow list.

- [Account and administration](/ai-agents/admin.md): Manage your Airbyte Agents account. This section covers billing, including plans, payments, usage monitoring, invoices, and subscription changes. On the Team and Custom plans, you can also manage users and configure single sign-on. If your network restricts inbound traffic, see IP allow list.

#### billing

Airbyte Agents bills you based on a unit of measurement called agent operations (AOs). All plans come with a specific number of free AOs, and most plans allow you to exceed this limit for an additional cost. Your plan determines:

- [Billing and pricing](/ai-agents/admin/billing.md): Airbyte Agents bills you based on a unit of measurement called agent operations (AOs). All plans come with a specific number of free AOs, and most plans allow you to exceed this limit for an additional cost. Your plan determines:

#### ip-allowlist

Airbyte Agents uses the Airbyte Cloud IP addresses, plus additional IP addresses for Agents-specific workloads. If your organization restricts inbound traffic by IP address, allow the following addresses in your firewall.

- [IP allow list for Agents](/ai-agents/admin/ip-allowlist.md): Airbyte Agents uses the Airbyte Cloud IP addresses, plus additional IP addresses for Agents-specific workloads. If your organization restricts inbound traffic by IP address, allow the following addresses in your firewall.

#### profile

Your Profile page contains your personal account details, your organization's name, and the API credentials Airbyte Agents uses to authenticate programmatic access. Go to the Profile page to review these details or make changes.

- [Manage your user profile](/ai-agents/admin/profile.md): Your Profile page contains your personal account details, your organization's name, and the API credentials Airbyte Agents uses to authenticate programmatic access. Go to the Profile page to review these details or make changes.

#### sessions

The Sessions page shows a read-only history of every agent interaction in your workspace, along with token usage and the connectors each interaction used. Use it to audit what your agents did, troubleshoot unexpected results, and understand what's driving your agent operations (AOs).

- [Review sessions](/ai-agents/admin/sessions.md): The Sessions page shows a read-only history of every agent interaction in your workspace, along with token usage and the connectors each interaction used. Use it to audit what your agents did, troubleshoot unexpected results, and understand what's driving your agent operations (AOs).

#### sso

Single sign-on lets your organization's members sign in to Airbyte Agents through your identity provider instead of using individual passwords. SSO is available on the Team and Custom plans.

- [Single sign-on (SSO)](/ai-agents/admin/sso.md): Single sign-on lets your organization's members sign in to Airbyte Agents through your identity provider instead of using individual passwords. SSO is available on the Team and Custom plans.

#### tool-calls

Airbyte Agents logs every tool call your agents make. Review this log to understand what your agents are doing, troubleshoot failures, and investigate unexpected activity or cost spikes.

- [Review tool calls](/ai-agents/admin/tool-calls.md): Airbyte Agents logs every tool call your agents make. Review this log to understand what your agents are doing, troubleshoot failures, and investigate unexpected activity or cost spikes.

#### users

The Users page lets organization administrators invite people, assign roles, and manage who has access to the organization. User management is available on the Team and Custom plans.

- [Users](/ai-agents/admin/users.md): The Users page lets organization administrators invite people, assign roles, and manage who has access to the organization. User management is available on the Team and Custom plans.

### concepts

This section covers the building blocks of Airbyte Agents: the models, resources, and platform behaviors you work with across every interface.

- [Core concepts](/ai-agents/concepts.md): This section covers the building blocks of Airbyte Agents: the models, resources, and platform behaviors you work with across every interface.

#### agent-operations

An Agent Operation (AO) is the unit of work in Airbyte Agents. AOs measure the processing intensity of a task, not just the number of requests.

- [Agent operations](/ai-agents/concepts/agent-operations.md): An Agent Operation (AO) is the unit of work in Airbyte Agents. AOs measure the processing intensity of a task, not just the number of requests.

#### architecture

Airbyte Agents is a cloud platform that gives AI agents authenticated, structured access to third-party SaaS data. It stores credentials, manages token refresh, and exposes a uniform execution interface so agents can read, search, and write data across dozens of services.

- [System architecture](/ai-agents/concepts/architecture.md): Airbyte Agents is a cloud platform that gives AI agents authenticated, structured access to third-party SaaS data. It stores credentials, manages token refresh, and exposes a uniform execution interface so agents can read, search, and write data across dozens of services.

##### connectors-and-credentials

A connector in Airbyte Agents is a stored credential and configuration for a third-party service. Connectors are what agents use to read, search, and write data in external systems. Each connector lives inside a workspace and is available to every interface that can access that workspace.

- [Connectors and credentials](/ai-agents/concepts/architecture/connectors-and-credentials.md): A connector in Airbyte Agents is a stored credential and configuration for a third-party service. Connectors are what agents use to read, search, and write data in external systems. Each connector lives inside a workspace and is available to every interface that can access that workspace.

##### organizations

An organization is the top-level container in Airbyte Agents. It represents a single account, maps to one billing subscription, and owns every workspace, connector, and credential beneath it.

- [Organizations](/ai-agents/concepts/architecture/organizations.md): An organization is the top-level container in Airbyte Agents. It represents a single account, maps to one billing subscription, and owns every workspace, connector, and credential beneath it.

##### workspaces

A workspace is an isolation boundary within an organization. Each workspace holds its own set of connectors and credentials. A token scoped to one workspace cannot access connectors in another.

- [Workspaces](/ai-agents/concepts/architecture/workspaces.md): A workspace is an isolation boundary within an organization. Each workspace holds its own set of connectors and credentials. A token scoped to one workspace cannot access connectors in another.

#### connect-ask-act

AI agents need real-time access to business data spread across dozens of systems, but building and maintaining those integrations is expensive and fragile. Airbyte Agents solves this with three layers: Connect your agents to any system, Ask questions across all your data in one searchable layer, and let agents Act on that data in real time.

- [Connect, Ask, Act](/ai-agents/concepts/connect-ask-act.md): AI agents need real-time access to business data spread across dozens of systems, but building and maintaining those integrations is expensive and fragile. Airbyte Agents solves this with three layers: Connect your agents to any system, Ask questions across all your data in one searchable layer, and let agents Act on that data in real time.

#### context-store

The Context Store is a managed, searchable replica of select entities from all your connected data sources. Airbyte populates it from the connectors in your workspace and gives agents a fast, consistent way to search your business data without hitting the underlying APIs for every request.

- [Context Store](/ai-agents/concepts/context-store.md): The Context Store is a managed, searchable replica of select entities from all your connected data sources. Airbyte populates it from the connectors in your workspace and gives agents a fast, consistent way to search your business data without hitting the underlying APIs for every request.

#### governance

Agent governance describes what an AI agent is allowed to do in your systems, on whose authority, and how you prove that. Governance is the umbrella term for the controls that mitigate that risk by narrowing access to only what each agent and person needs. Airbyte provides tools to control what data your agents, and the people who build them, can reach, and how they're allowed to interact with that data.

- [Governance](/ai-agents/concepts/governance.md): Agent governance describes what an AI agent is allowed to do in your systems, on whose authority, and how you prove that. Governance is the umbrella term for the controls that mitigate that risk by narrowing access to only what each agent and person needs. Airbyte provides tools to control what data your agents, and the people who build them, can reach, and how they're allowed to interact with that data.

#### semantic-search

Structured Context Store search matches records by exact or fuzzy field values. Semantic search is a different mode: a similarity search that finds relevant passages by meaning rather than by keyword. Instead of matching a filter, Airbyte embeds your natural-language prompt as a vector, compares it against the stored text, and returns the most similar passages, ranked by relevance.

- [Semantic search](/ai-agents/concepts/semantic-search.md): Structured Context Store search matches records by exact or fuzzy field values. Semantic search is a different mode: a similarity search that finds relevant passages by meaning rather than by keyword. Instead of matching a filter, Airbyte embeds your natural-language prompt as a vector, compares it against the stored text, and returns the most similar passages, ranked by relevance.

#### skills

Skills are Markdown instructions that tell your agents how to handle a particular kind of task. Airbyte serves them to every agent in your organization, regardless of the interface the agent uses: the web app, the MCP server, the API, the SDK, or the CLI. Airbyte provides skills for connectors and for common analytical tasks, and organization administrators can add custom skills that capture knowledge only your organization has.

- [Skills](/ai-agents/concepts/skills.md): Skills are Markdown instructions that tell your agents how to handle a particular kind of task. Airbyte serves them to every agent in your organization, regardless of the interface the agent uses: the web app, the MCP server, the API, the SDK, or the CLI. Airbyte provides skills for connectors and for common analytical tasks, and organization administrators can add custom skills that capture knowledge only your organization has.

#### time-zones

Airbyte Agents handles dates and times the same way across every interface: display in your local time, and store and transport in UTC. This page explains how that works and what to expect in each interface.

- [Time zones](/ai-agents/concepts/time-zones.md): Airbyte Agents handles dates and times the same way across every interface: display in your local time, and store and transport in UTC. This page explains how that works and what to expect in each interface.

### connectors

Airbyte's agent connectors equip AI agents to call third-party APIs through strongly typed, well-documented tools. All connectors ship inside the single airbyte-agent-sdk Python package and are ready to use in Airbyte Agents through any of the supported interfaces.

- [Agent connectors](/ai-agents/connectors.md): Airbyte's agent connectors equip AI agents to call third-party APIs through strongly typed, well-documented tools. All connectors ship inside the single airbyte-agent-sdk Python package and are ready to use in Airbyte Agents through any of the supported interfaces.

#### amazon-ads

The Amazon-Ads agent connector is a Python package that equips AI agents to interact with Amazon-Ads through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

- [Amazon-Ads](/ai-agents/connectors/amazon-ads.md): The Amazon-Ads agent connector is a Python package that equips AI agents to interact with Amazon-Ads through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

##### AUTH

This page documents the authentication and configuration options for the Amazon-Ads agent connector.

- [Amazon-Ads authentication](/ai-agents/connectors/amazon-ads/AUTH.md): This page documents the authentication and configuration options for the Amazon-Ads agent connector.

##### REFERENCE

This is the full reference documentation for the Amazon-Ads agent connector.

- [Amazon-Ads full reference](/ai-agents/connectors/amazon-ads/REFERENCE.md): This is the full reference documentation for the Amazon-Ads agent connector.

#### amazon-seller-partner

The Amazon-Seller-Partner agent connector is a Python package that equips AI agents to interact with Amazon-Seller-Partner through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

- [Amazon-Seller-Partner](/ai-agents/connectors/amazon-seller-partner.md): The Amazon-Seller-Partner agent connector is a Python package that equips AI agents to interact with Amazon-Seller-Partner through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

##### AUTH

This page documents the authentication and configuration options for the Amazon-Seller-Partner agent connector.

- [Amazon-Seller-Partner authentication](/ai-agents/connectors/amazon-seller-partner/AUTH.md): This page documents the authentication and configuration options for the Amazon-Seller-Partner agent connector.

##### REFERENCE

This is the full reference documentation for the Amazon-Seller-Partner agent connector.

- [Amazon-Seller-Partner full reference](/ai-agents/connectors/amazon-seller-partner/REFERENCE.md): This is the full reference documentation for the Amazon-Seller-Partner agent connector.

#### amplitude

The Amplitude agent connector is a Python package that equips AI agents to interact with Amplitude through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

- [Amplitude](/ai-agents/connectors/amplitude.md): The Amplitude agent connector is a Python package that equips AI agents to interact with Amplitude through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

##### AUTH

This page documents the authentication and configuration options for the Amplitude agent connector.

- [Amplitude authentication](/ai-agents/connectors/amplitude/AUTH.md): This page documents the authentication and configuration options for the Amplitude agent connector.

##### REFERENCE

This is the full reference documentation for the Amplitude agent connector.

- [Amplitude full reference](/ai-agents/connectors/amplitude/REFERENCE.md): This is the full reference documentation for the Amplitude agent connector.

#### asana

The Asana agent connector is a Python package that equips AI agents to interact with Asana through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

- [Asana](/ai-agents/connectors/asana.md): The Asana agent connector is a Python package that equips AI agents to interact with Asana through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

##### AUTH

This page documents the authentication and configuration options for the Asana agent connector.

- [Asana authentication](/ai-agents/connectors/asana/AUTH.md): This page documents the authentication and configuration options for the Asana agent connector.

##### REFERENCE

This is the full reference documentation for the Asana agent connector.

- [Asana full reference](/ai-agents/connectors/asana/REFERENCE.md): This is the full reference documentation for the Asana agent connector.

#### chargebee

The Chargebee agent connector is a Python package that equips AI agents to interact with Chargebee through strongly typed, well-documented tools. It's ready

Measured from the saved file

Linked document

UTF-8 text size: 529.7 KB542413 bytes
Token count: 116785o200k_base encoding
Markdown links: 16371637 distinct URLs
Fenced code blocks: 08261 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: docs.airbyte.com llms.txt

Opening blockquote

Airbyte is an open source platform designed for building and managing data pipelines, offering extensive connector options to facilitate data movement from various sources to destinations efficiently and effectively.

Context architecture

How the content is divided

7 H2 sections organize 1637 parsed Markdown links. 1637 links point to .md or .markdown paths; 0 link occurrences use another hostname. Linked destinations have not been recursively validated.

H2 section	Links	UTF-8 bytes	Tokens
searchLine 7	1	43	11
ai-agentsLine 11	541	152459	32838
communityLine 1630	16	4836	1049
developersLine 1729	58	10359	2409
integrationsLine 1822	785	281211	60634
platformLine 6967	189	82188	16427
release_notesLine 7986	46	10931	3344

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: 5 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

No H2 section named “Optional” was detected in this file.

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. 14 links have MCP-related labels or paths. 0 fenced lines match mcpServers, mcp_servers or an mcp add command.

Agent MCP ↗Source line 1224
MCP servers ↗Source line 1721
Airbyte knowledge MCP server ↗Source line 1727
Module airbyte.mcp ↗Source line 1794
Module airbyte.mcp.cloud ↗Source line 1795
Module airbyte.mcp.local ↗Source line 1796
Module airbyte.mcp.prompts ↗Source line 1797
Module airbyte.mcp.registry ↗Source line 1798
Module airbyte.mcp.server ↗Source line 1799
Airbyte MCP ↗Source line 7046
Advanced Airbyte MCP configuration ↗Source line 7052
Install the Airbyte MCP ↗Source line 7058

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

Index and bundle

llms-full.txt companion

Not established

HTML returned; not a verified text file

https://docs.airbyte.com/llms-full.txt ↗

Discovery: Conventional sibling path checked
HTTP response: 206 · text/html; charset=utf-8

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

173 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://docs.airbyte.com/llms.txt
Final endpoint: https://docs.airbyte.com/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: ced48feb769efbaed4dcc8d384cd53cbebadc5371a3453838f5efb66a506310e

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.