# Agent Interoperability Protocol Documentation

> AIP is a governed protocol for capability discovery, durable actions, approvals, transactions, delegation, compatibility profiles, and operational evidence.

Protocol version: 1.0
Machine-readable index: /docs/index.json
Search index: /docs/search-index.json

## Start Here

- [Documentation Home](/docs/raw/index.md): Use this documentation to choose an AIP integration path, implement it, operate a deployment, and judge the evidence behind the behavior you depend on. It is written primarily for application developers, with focused paths for connector dev
- [What AIP Is](/docs/raw/getting-started/what-is-aip.md): Agent Interoperability Protocol (AIP) gives agents, applications, tools, and workflows a shared contract for discovering capabilities and managing work that extends beyond a one-shot call. This page is for evaluators deciding whether AIP fi
- [Install AIP](/docs/raw/getting-started/installation.md): Install signed GetAIP 2.1.0 tools with npm, connect a supported agent, or build from reviewed source
- [Quickstart](/docs/raw/getting-started/quickstart.md): In this tutorial, you will start one product-neutral AIP daemon on loopback, discover its manifest, invoke the built-in health capability, and inspect the durable action state. You will also view the same capability surface through the MCP 
- [Connector Fleet Quickstart](/docs/raw/getting-started/connector-fleet-quickstart.md): In this tutorial, you will start the repository's deterministic connector-fleet topology and route one signed AIP action from getaip, through a product-neutral getaip-server gateway, to a standalone Cal.diy connector host. The provider is a
- [How AIP Works](/docs/raw/getting-started/how-aip-works.md): AIP turns a capability request into an identified, governed, and observable unit of work. This page is for developers who need to understand what happens between discovery and a terminal result before choosing an API, profile, connector, or

## Core Concepts

- [Capabilities and Contracts](/docs/raw/concepts/capabilities.md): An AIP capability describes a stable operation that a caller can discover and request. Its schemas describe data shape; its contract describes side effects, execution, retry, data, credential, approval, and recovery expectations. This page 
- [Actions and Sessions](/docs/raw/concepts/actions-and-sessions.md): Use this page to understand how an AIP client identifies, observes, cancels, and resumes work. It is for application and connector developers who need to track an invocation beyond one request or transport connection.
- [Identity and Trust](/docs/raw/concepts/identity-and-trust.md): Use this page to decide which identity values an AIP implementation may trust. It covers caller authentication, tenant and provider-account selection, credential resolution, operational reads, and connector dispatch. It is for gateway, runt
- [Approvals and Policy](/docs/raw/concepts/approvals-and-policy.md): Use this page to understand how an AIP capability can pause an action for approval and later resume the same queued action from durable, verified authorization. It is for application, runtime, and connector developers who need to preserve a
- [Transactions and Compensation](/docs/raw/concepts/transactions-and-compensation.md): Use this page to understand how AIP represents a consequential operation before, during, and after a connector attempt. It is for application, runtime, and connector developers who need to plan a mutation, own one commit attempt, recover an
- [Delegation](/docs/raw/concepts/delegation.md): Use this page to understand how one AIP participant asks another participant to execute a child action. It is for application, runtime, and gateway developers who need to preserve a parent-child graph, choose local or remote execution, and 
- [Profiles, Transports, and Connectors](/docs/raw/concepts/profiles-and-connectors.md): Use this page to choose the correct integration boundary without creating a second action model or bypassing AIP governance. It is for application, gateway, and connector developers who need to distinguish protocol semantics, wire delivery,

## Build with AIP

- [Use Native AIP](/docs/raw/guides/use-native-aip.md): Use this guide when a client needs the complete AIP action lifecycle without projecting it through MCP, A2A, or another compatibility protocol. You will discover a capability, invoke it with a stable action identity, follow durable state, a
- [Use the Rust SDK](/docs/raw/guides/use-rust-sdk.md): Use this guide when a Rust application needs AIP 1.0 types, validation, or an embedded protocol service. You will pin the reviewed SDK source, construct and validate one native action envelope, and then select optional modules by role. The 
- [Build a Trusted Local Module](/docs/raw/guides/trusted-local-modules.md): Use this guide to compile one reviewed, first-party capability into the same process as getaip-server. You will pair a context-aware action handler with an exact manifest contribution. You will then register its factory before startup and v
- [Use AIP through MCP](/docs/raw/guides/use-aip-through-mcp.md): Use this guide to connect an existing MCP client to getaip-server, inspect the projected surface, and call one native AIP capability through a stable facade tool. The same AIP gateway validates and dispatches the resulting action; the MCP l
- [Use AIP through A2A](/docs/raw/guides/use-aip-through-a2a.md): Use this guide to connect an A2A 1.0 client to getaip-server, discover the local Agent Card, invoke one native AIP capability, and read the projected task. The A2A profile translates JSON-RPC messages into native actions. AIP remains the ow
- [Examples and Reference Deployments](/docs/raw/guides/examples-and-reference-deployments.md): Use this page to choose the smallest source artifact that matches your goal. The repository contains short Rust examples, local smoke topologies, scenario harnesses, and qualification environments. These artifact classes answer different qu

## Connectors

- [Connector Index](/docs/raw/connectors.md): Use this catalog to choose the product boundary that matches your integration. Each connector translates a bounded provider API and event surface into AIP capabilities, contracts, lifecycle state, and errors. The connector does not change t
- [Cal.diy Connector](/docs/raw/connectors/cal-diy.md): Use the Cal.diy connector when an AIP application must read or change scheduling state through a typed, tenant-bound capability surface. The connector runs in its own aip-host-cal-diy process; product-neutral getaip-server routes admitted a
- [Quickstart](/docs/raw/connectors/cal-diy/getting-started/quickstart.md): In this tutorial, you will start the repository's deterministic Cal.diy fleet path and route one signed cap:caldiy:profile.get action from getaip, through product-neutral getaip-server, to the standalone aip-host-cal-diy process.
- [Authentication and Tenant Routing](/docs/raw/connectors/cal-diy/getting-started/authentication-and-tenant-routing.md): Use this procedure to give one aip-host-cal-diy instance access to one Cal.diy account and make that account reachable only through its admitted AIP tenant binding. The provider secret stays in the host deployment; an AIP action cannot sele
- [Configuration Reference](/docs/raw/connectors/cal-diy/reference/configuration.md): This reference describes the configuration accepted by aip-host-cal-diy at the reviewed source revision. A complete process combines eight Cal.diy-specific settings with the shared connector-host bootstrap contract.
- [Capabilities](/docs/raw/connectors/cal-diy/capabilities.md): Use this index to choose the Cal.diy operation family that owns your task and to interpret its AIP contract before invocation. The reviewed catalogue contains 81 unique capabilities. Each id begins with cap:caldiy: and maps to one fixed pro
- [Profile](/docs/raw/connectors/cal-diy/capabilities/profile.md): Use these capabilities to read or update the Cal.diy profile owned by the credential configured for the selected external account. Both operations use the provider route /v2/me, but they have different authorization, approval, idempotency, 
- [Event Types, Private Links, and Webhooks](/docs/raw/connectors/cal-diy/capabilities/event-types-and-private-links.md): Use this capability family to define bookable event types, issue restricted booking links, and attach webhooks to one event type. The reviewed connector exposes 15 fixed operations. Choose an AIP capability from this page instead of constru
- [Slots and Reservations](/docs/raw/connectors/cal-diy/capabilities/slots-and-reservations.md): Use these capabilities to query availability and to create, inspect, update, or release a short-lived slot reservation. The reviewed connector exposes five fixed operations, all pinned to Cal.diy API version 2024-09-04.
- [Bookings](/docs/raw/connectors/cal-diy/capabilities/bookings.md): Use these capabilities to find, create, reschedule, cancel, confirm, decline, mark, reassign, or relocate a Cal.diy booking. This page owns 12 core lifecycle operations. A separate capability page owns the 10 attendee, guest, calendar link,
- [Booking Attendees and Artifacts](/docs/raw/connectors/cal-diy/capabilities/booking-attendees-and-artifacts.md): Use these capabilities to manage the people attached to a booking or to read its generated calendar links, provider references, recording and transcript metadata, and conferencing sessions. This page owns the remaining 10 operations in the 
- [Calendars and Free/Busy](/docs/raw/connectors/cal-diy/capabilities/calendars-and-free-busy.md): Use these capabilities to inspect connected calendars and availability, work with events, choose the destination for new bookings, or change conflict detection. This page owns all 21 operations in the calendar family.
- [Conferencing](/docs/raw/connectors/cal-diy/capabilities/conferencing.md): Use these capabilities to list installed conferencing applications, read or change the default, and connect or disconnect an app supported by the published business-operation surface. This page owns all five conferencing operations.
- [Schedules](/docs/raw/connectors/cal-diy/capabilities/schedules.md): Use these capabilities to list or read schedules and to create, update, or delete weekly availability with dated overrides. This page owns all six schedule operations.
- [User Webhooks](/docs/raw/connectors/cal-diy/capabilities/webhooks.md): Use these capabilities to list, read, create, update, or delete user-level Cal.diy webhook subscriptions. This page owns five management operations and the exact trigger catalog they accept.
- [Create and Manage a Booking](/docs/raw/connectors/cal-diy/guides/create-and-manage-a-booking.md): Use this guide to create one booking through slot discovery, an immutable plan, trusted approval, a single commit attempt, and durable verification. It then shows how to choose a safe follow-up operation without replaying an uncertain provi
- [Availability and Calendar Conflicts](/docs/raw/connectors/cal-diy/guides/availability-and-calendar-conflicts.md): Use this guide when an expected Cal.diy slot is missing or when you need to change the schedule or calendars that can contribute conflicts. You will capture one repeatable baseline, change at most one input layer, and compare the same obser
- [API Versions and Provider Routes](/docs/raw/connectors/cal-diy/reference/api-versions-and-provider-routes.md): This reference maps every public Cal.diy capability to the provider method, cal-api-version header, and relative path selected by the reviewed connector. Use it to review connector changes, diagnose provider-contract drift, or trace a gover
- [Webhook Security and Replay](/docs/raw/connectors/cal-diy/reference/webhook-security-and-replay.md): This reference defines the reviewed Cal.diy inbound webhook boundary. It covers route and header validation, raw-body HMAC, version and time checks, deterministic event mapping, bounded replay state, idempotent duplicate handling, durable l
- [Operator-only and Excluded Routes](/docs/raw/connectors/cal-diy/reference/operator-only-and-excluded-routes.md): Use this reference when a Cal.diy provider route has no corresponding AIP capability. It distinguishes deliberate control-plane exclusions from the 81 published business operations and maps two deprecated aliases to their canonical capabili
- [Deploy the Cal.diy Connector](/docs/raw/connectors/cal-diy/operations/deployment.md): Use this guide to deploy one aip-host-cal-diy instance for one Cal.diy account, prove that its admitted identity and provider connection are ready, and introduce it to tenant traffic without creating a second account owner.
- [Recover the Cal.diy Connector](/docs/raw/connectors/cal-diy/operations/health-observability-and-recovery.md): Use this runbook when a Cal.diy host is unavailable, not ready, rejecting traffic, losing its lease, returning uncertain mutation outcomes, or retaining provider events for later delivery. It helps you identify the failing boundary and reco
- [Troubleshoot Cal.diy](/docs/raw/connectors/cal-diy/troubleshooting.md): Use this runbook to trace a Cal.diy failure to the signed host, pinned route, credential, provider request, durable mutation, or webhook boundary. The result should be one explicit decision: correct, wait, retry a read within policy, reconc
- [Qualify the Cal.diy Connector](/docs/raw/connectors/cal-diy/qualification.md): Use this page to decide whether one exact Cal.diy connector artifact has enough evidence for a bounded release claim. It separates source-defined tests, deterministic provider fixtures, standalone-host behavior, and actual Cal.diy observati
- [Cal.diy Connector Changelog](/docs/raw/connectors/cal-diy/changelog.md): Use this page to decide whether a Cal.diy connector change requires a new artifact, migration, or caller update. It records only changes that can be established from the reviewed source range.
- [Hermes Agent Connector](/docs/raw/connectors/hermes-agent.md): Use the Hermes Agent connector when an AIP application needs typed access to one or more Hermes API servers. The connector exposes health, discovery, chat, responses, structured runs, sessions, jobs, and governed operator work through endpo
- [Quickstart](/docs/raw/connectors/hermes-agent/getting-started/quickstart.md): In this tutorial, you will start the repository's deterministic Hermes product path and route one signed cap:hermesagent:qualification:models action from getaip, through product-neutral getaip-server, to the standalone aip-host-hermes-agent
- [Authentication and Endpoints](/docs/raw/connectors/hermes-agent/getting-started/authentication-and-endpoints.md): Use this guide to bind one or more owner-configured Hermes API servers to one AIP tenant. Each endpoint receives a stable ID, a fixed base URL, and its own owner-only APISERVERKEY file. The connector then publishes endpoint-qualified capabi
- [Configuration Reference](/docs/raw/connectors/hermes-agent/reference/configuration.md): This reference describes the configuration accepted by aip-host-hermes-agent at the reviewed source revision. A complete process combines two Hermes-specific file settings with the 45 shared connector-host settings.
- [Capabilities](/docs/raw/connectors/hermes-agent/capabilities.md): Use this index to select the Hermes operation that owns your task and read its AIP policy contract before invocation. Every configured endpoint publishes the same 35 operations under cap:hermesagent::.
- [Health, Models, and Discovery](/docs/raw/connectors/hermes-agent/capabilities/health-models-and-discovery.md): Use these six endpoint-qualified reads to inspect one configured Hermes API server. They report transport health, runtime detail, model aliases, the provider API surface, enabled skills, and resolved toolsets.
- [Chat and Responses](/docs/raw/connectors/hermes-agent/capabilities/chat-and-responses.md): Use these six endpoint-qualified capabilities to run a Hermes model turn, stream it, continue a Responses chain, read a stored response, or delete that stored response. They are for applications using a Hermes endpoint already admitted to t
- [Structured Runs and Approvals](/docs/raw/connectors/hermes-agent/capabilities/durable-runs-and-approvals.md): Use these five endpoint-qualified capabilities when an application needs a Hermes run ID, pollable status, structured lifecycle events, a provider approval continuation, or an explicit stop request.
- [Operator and Delegation](/docs/raw/connectors/hermes-agent/capabilities/operator-and-delegation.md): Use the endpoint-qualified operator capability when Hermes must plan a governed objective, call admitted AIP tools, pause for approval, resume the same provider run, or reconcile cancellation across a process restart.
- [Persistent Sessions](/docs/raw/connectors/hermes-agent/capabilities/sessions.md): Use these nine endpoint-qualified capabilities when an application needs an addressable Hermes transcript that can be listed, inspected, renamed, branched, continued, or deleted.
- [Scheduled Jobs](/docs/raw/connectors/hermes-agent/capabilities/scheduled-jobs.md): Use these eight endpoint-qualified capabilities to manage jobs in one Hermes profile's persistent scheduler.
- [Stream Chat and Responses](/docs/raw/connectors/hermes-agent/guides/chat-and-response-streaming.md): Use this guide to submit one streaming Hermes action, render incremental output, and reach a verified AIP result. It covers chatstream and responsesstream on an endpoint already admitted to the tenant catalogue.
- [Run an Operator Lifecycle](/docs/raw/connectors/hermes-agent/guides/operator-lifecycle.md): Use this guide to take one direct Hermes operator objective from its AIP approval to completion, a provider approval, a failure, or a controlled stop. It keeps every identity and approval decision attached to the work it governs.
- [Manage a Persisted Session](/docs/raw/connectors/hermes-agent/guides/session-lifecycle.md): Use this guide to create one addressable Hermes transcript, add a verified turn, branch it, and then close or delete it deliberately. The procedure keeps transcript identity, AIP execution identity, and optional memory scope separate throug
- [Streaming Cursors and Cancellation](/docs/raw/connectors/hermes-agent/reference/streaming-cursors-and-cancellation.md): Use this reference when a Hermes-backed Action must survive a delivery disconnect, deadline, or cancellation request. It defines the native AIP action-event position and the evidence available after interruption.
- [AIP MCP inside Hermes](/docs/raw/connectors/hermes-agent/reference/aip-mcp-inside-hermes.md): Use this reference when a Hermes role must discover or call admitted AIP capabilities through getaip-server Streamable HTTP MCP. The configuration gives Hermes a bounded model-facing tool surface. Native AIP identity, policy, approval, and 
- [Deploy the Hermes Agent Connector](/docs/raw/connectors/hermes-agent/operations/deployment.md): Use this guide to deploy one aip-host-hermes-agent instance for a fixed set of tenant-owned Hermes endpoints. The procedure proves host and endpoint readiness, introduces one protected read, and replaces or rolls back replicas without movin
- [Recover the Hermes Agent Connector](/docs/raw/connectors/hermes-agent/operations/health-observability-and-recovery.md): Use this runbook when a Hermes Agent host is unavailable, not ready, losing its lease, rejecting traffic, or retaining work whose provider outcome is uncertain. It helps you locate the failing boundary and recover the same AIP Action, opera
- [Troubleshoot the Hermes Agent Connector](/docs/raw/connectors/hermes-agent/troubleshooting.md): Use this runbook to trace a Hermes failure to the signed host, pinned route, endpoint credential, provider request, stream, operator binding, or delegated child. End with one explicit decision: correct, wait, retry a read within policy, rec
- [Qualify the Hermes Agent Connector](/docs/raw/connectors/hermes-agent/qualification.md): Use this page to decide whether one exact Hermes Agent connector artifact has enough evidence for a bounded release claim. It separates source-defined checks, deterministic provider fixtures, standalone-host behavior, path probes, and actua
- [Hermes Agent Connector Changelog](/docs/raw/connectors/hermes-agent/changelog.md): Use this page to decide whether a Hermes Agent connector change requires a new artifact, admission record, provider qualification, migration, or caller update. It records only changes established from the reviewed source range.
- [Chatwoot Connector](/docs/raw/connectors/chatwoot.md): Use the Chatwoot connector when an AIP application needs governed access to one Chatwoot account. The connector exposes 150 fixed account API operations, three conversation-oriented AIP composites, and optional authenticated webhook ingress
- [Run the Chatwoot Connector Quickstart](/docs/raw/connectors/chatwoot/getting-started/quickstart.md): In this tutorial, you will start the repository's deterministic Chatwoot path, read its tenant-bound account, and create one conversation after an explicit approval. Both Actions travel from getaip, through product-neutral getaip-server, to
- [Authenticate and Isolate a Chatwoot Account](/docs/raw/connectors/chatwoot/getting-started/authentication-and-account-isolation.md): Use this guide to bind one standalone Chatwoot host to one verified AIP tenant, one fixed provider origin, and one Chatwoot account. The API token stays in an owner-controlled file and is added to provider requests inside the connector.
- [Chatwoot Connector Configuration](/docs/raw/connectors/chatwoot/reference/configuration.md): This reference describes the configuration accepted by aip-host-chatwoot at the reviewed source revision. A complete process combines six Chatwoot-specific settings with the shared connector-host identity, storage, routing, lifecycle, and o
- [Chatwoot Capabilities](/docs/raw/connectors/chatwoot/capabilities.md): Use this index to choose the Chatwoot operation family that owns your task and to interpret its AIP contract before invocation. The reviewed surface contains 150 fixed account API operations and three connector-level composites.
- [Accounts, Agents, Teams, and Bots](/docs/raw/connectors/chatwoot/capabilities/accounts-agents-teams-and-bots.md): Use these 22 capabilities to read the configured account and manage its human agents, teams, team membership, and agent bots. The connector owns the account boundary; an Action cannot select another Chatwoot account through its input.
- [Inboxes, Integrations, and Webhooks](/docs/raw/connectors/chatwoot/capabilities/inboxes-integrations-and-webhooks.md): Use these 21 capabilities to inspect and configure Chatwoot inboxes, membership, integration hooks, and account webhook records. Select the family that owns the provider object; similarly named operations do not configure the same boundary.
- [Conversations and Messages](/docs/raw/connectors/chatwoot/capabilities/conversations-and-messages.md): Use these 28 provider-catalogue capabilities to find conversations, change their state, manage routing metadata, and work with messages. Start by choosing the exact capability ID; three similar connector composites use different punctuation
- [Contacts and Companies](/docs/raw/connectors/chatwoot/capabilities/contacts-and-companies.md): Use these 31 capabilities to find and maintain contacts, notes, companies, and their relationships. Treat every result as restricted customer data, including reads that need no human approval.
- [Automation, Macros, and Canned Responses](/docs/raw/connectors/chatwoot/capabilities/automation-macros-and-canned-responses.md): Use these 16 capabilities to manage reusable text and behavior in the configured Chatwoot account. Review both the immediate provider request and the later effects of the object being created, changed, cloned, or executed.
- [Labels, Attributes, and Filters](/docs/raw/connectors/chatwoot/capabilities/labels-attributes-and-filters.md): Use these 15 capabilities to manage account-level labels, custom-attribute definitions, and saved custom filters. Choose a definition operation only when the task is to maintain the reusable metadata object itself.
- [Reports, Audit, and Events](/docs/raw/connectors/chatwoot/capabilities/reports-audit-and-events.md): Use these 12 read-only capabilities to inspect Chatwoot audit records, reporting events, and metrics for the configured account. Keep each provider observation separate from AIP's own Action, approval, routing, Event, and outbox evidence.
- [Customer Portals](/docs/raw/connectors/chatwoot/capabilities/customer-portals.md): Use these five capabilities to list and create help-center portals, update a portal, and create categories or articles inside a portal. The reviewed surface is intentionally smaller than full portal CRUD.
- [AIP Composite Operations](/docs/raw/connectors/chatwoot/capabilities/aip-composite-operations.md): Use these three connector-owned capabilities when the strict message, status, or handoff contract matches the task. Their colon-named IDs, inputs, outputs, and provider behavior differ from the generic dot-named catalogue.
- [Manage a Chatwoot Conversation Lifecycle](/docs/raw/connectors/chatwoot/guides/conversation-and-message-lifecycle.md): Use this guide to read one existing Chatwoot conversation, send a public reply or private note, set its status, optionally hand it off, and verify final provider state. An optional branch covers creation when the pinned provider body is alr
- [Manage Chatwoot Contacts and Companies](/docs/raw/connectors/chatwoot/guides/contact-and-company-workflows.md): Use this guide to find authoritative customer records, resolve ambiguous matches, attach or detach one contact and company, or merge two contacts. Every mutation ends with a provider read instead of relying on the write response.
- [Chatwoot Webhook Security, Loop Prevention, and Replay](/docs/raw/connectors/chatwoot/reference/webhook-security-loop-prevention-and-replay.md): This reference defines the complete inbound boundary implemented by the reviewed Chatwoot host. It covers authentication, freshness, replay state, loop suppression, event identity, local persistence, and central publication.
- [Chatwoot Mutation Approval and Idempotency](/docs/raw/connectors/chatwoot/reference/mutation-approval-and-idempotency.md): This reference gives one governance model for all 153 Chatwoot capabilities at the reviewed revision. Use it to decide when approval is required, how to bind an idempotency key, and what to do after an uncertain result.
- [Deploy the Chatwoot Connector](/docs/raw/connectors/chatwoot/operations/deployment.md): Use this guide to deploy one immutable aip-host-chatwoot instance for one Chatwoot account. The result is an admitted, durable, observable route that can be replaced or rolled back without changing central getaip-server.
- [Operate and Recover the Chatwoot Connector](/docs/raw/connectors/chatwoot/operations/health-observability-and-recovery.md): Use this runbook to locate a Chatwoot connector failure before changing state. It starts with process, readiness, lease, storage, and provider evidence, then moves to Action, webhook, and central Event delivery.
- [Troubleshoot the Chatwoot Connector](/docs/raw/connectors/chatwoot/troubleshooting.md): Use this runbook to identify the boundary that failed and choose the narrowest safe next action. It covers process startup, connector-host readiness, signed Action handling, Chatwoot provider calls, webhook ingress, and durable Event delive
- [Qualify the Chatwoot Connector](/docs/raw/connectors/chatwoot/qualification.md): Use this page to decide whether one exact Chatwoot connector artifact supports a bounded release claim. It keeps source-defined tests, standalone-image checks, controlled provider fixtures, and actual Chatwoot observations as separate evide
- [Chatwoot Connector Changelog](/docs/raw/connectors/chatwoot/changelog.md): Use this page to decide whether a Chatwoot connector change requires a new artifact, admission record, migration, caller update, or qualification result. It records only changes established from the reviewed Git range.
- [Dify Connector](/docs/raw/connectors/dify.md): The Dify connector exposes configured Dify applications and workspace Knowledge APIs as governed AIP capabilities. Use it when AIP must invoke published apps, manage conversations or files, operate knowledge data, stream workflow events, or
- [Run the Dify Connector Quickstart](/docs/raw/connectors/dify/getting-started/quickstart.md): In this tutorial, you will start the repository's deterministic Dify path and route two signed reads from getaip, through product-neutral getaip-server, to the standalone aip-host-dify process.
- [Dify Authentication and Credential Scopes](/docs/raw/connectors/dify/getting-started/authentication-and-credential-scopes.md): The Dify connector crosses several identity boundaries in one Action. Keep each identity in its source-owned layer so a caller cannot select a provider credential, another workspace, or another end user's conversation context.
- [Dify Connector Configuration](/docs/raw/connectors/dify/reference/configuration.md): This reference describes the product settings accepted by aip-host-dify at the reviewed source revision. A complete process combines these settings with the shared connector-host identity, storage, routing, signing, lifecycle, and optional 
- [Dify Capabilities](/docs/raw/connectors/dify/capabilities.md): Use this index to choose the Dify family that owns a task and interpret its AIP contract before invocation. The reviewed frozen catalogue contains 79 provider operations. Each configured app also publishes one invocation alias.
- [Dify App Metadata and Files](/docs/raw/connectors/dify/capabilities/app-metadata-and-files.md): Use this family to inspect a published application's contract, read one Dify end user, upload one bounded user file, or retrieve one bounded file preview. All seven operations are available for every supported app mode.
- [Dify Chat Messages, Conversations, and Feedback](/docs/raw/connectors/dify/capabilities/chat-messages-conversations-and-feedback.md): Use this family to read chat history, suggestions, conversations, feedback, and conversation variables. It also supports feedback, rename, variable-update, and conversation-deletion mutations under AIP approval and replay controls.
- [Dify Workflows, Streaming, and Human Input](/docs/raw/connectors/dify/capabilities/workflows-streaming-and-human-input.md): Use this family to inspect workflow runs and logs, execute a published workflow version, stop active generation, resume workflow events, and answer a paused human-input form. The nine operations use one configured Dify app credential.
- [Dify Audio Transcription and Synthesis](/docs/raw/connectors/dify/capabilities/audio.md): Use this family to convert one authorized audio file to text or convert authorized text to speech. Both operations use the selected app credential and a stable Dify end-user identity.
- [Dify Annotations and Annotation Reply](/docs/raw/connectors/dify/capabilities/annotations.md): Use this family to list and maintain question-and-answer annotations for one Dify app. It also enables or disables annotation reply and exposes the provider background job that changes its search index.
- [Dify Knowledge Bases, Tags, and Retrieval](/docs/raw/connectors/dify/capabilities/knowledge-datasets-tags-and-retrieval.md): Use this family to discover, create, configure, organize, query, or remove Dify knowledge bases. The thirteen operations use one workspace-scoped knowledge credential rather than an app credential.
- [Dify Knowledge Documents and Indexing](/docs/raw/connectors/dify/capabilities/knowledge-documents-and-indexing.md): Use this family to create or replace knowledge documents, inspect their state, track indexing, control availability, download approved sources, or delete one document. All twelve operations use a workspace-scoped knowledge credential.
- [Dify Segments and Child Chunks](/docs/raw/connectors/dify/capabilities/knowledge-segments-and-child-chunks.md): Use this family to create, inspect, update, or delete document segments and their child chunks. The nine operations resolve every object through its workspace, dataset, document, and parent ownership chain.
- [Dify Models and Knowledge Metadata](/docs/raw/connectors/dify/capabilities/knowledge-metadata-and-models.md): Use this family to discover configured workspace models, define dataset metadata, enable built-in fields, or assign values to documents. All eight operations use one workspace-scoped knowledge credential.
- [Dify Knowledge Pipelines and Uploads](/docs/raw/connectors/dify/capabilities/knowledge-pipelines-and-uploads.md): Use this family to discover datasource nodes, upload a local source, run one datasource node, or run a full Dify knowledge pipeline. The four operations use one workspace-scoped knowledge credential.
- [Run Dify Chat and Workflows](/docs/raw/connectors/dify/guides/run-chat-and-workflows.md): Follow this guide to invoke one configured Dify app through its discovery-derived alias. The procedure applies to chat, completion, workflow, advanced-chat, agent-chat, and agent modes.
- [Manage Dify Knowledge](/docs/raw/connectors/dify/guides/manage-knowledge.md): Use this procedure to create or select one Dify knowledge base, ingest an approved source, wait for indexing, enrich its records, and verify retrieval. An optional final branch updates or removes provider state.
- [Dify Modes and Capability Projection](/docs/raw/connectors/dify/reference/modes-and-capability-projection.md): This reference defines how startup descriptors become Dify capabilities in tenant discovery. Use it to predict a manifest, diagnose a missing suffix, or detect a mode that exposes an unintended provider route.
- [Dify Streaming, Cancellation, and Binary Responses](/docs/raw/connectors/dify/reference/streaming-cancellation-and-binary-responses.md): This reference defines the Dify connector response paths after an action passes admission. Use it to select an execution mode and preserve uncertainty when a provider task cannot be proved stopped.
- [Deploy the Dify Connector](/docs/raw/connectors/dify/operations/deployment.md): Use this guide to deploy one immutable aip-host-dify instance. The result is an admitted, observable route with durable cancellation evidence and a defined replacement and rollback path.
- [Operate and Recover the Dify Connector](/docs/raw/connectors/dify/operations/health-observability-and-recovery.md): Use this runbook to locate a Dify connector failure before changing state. Inspect process, readiness, lease, storage, credential, Action, and provider evidence in that order.
- [Troubleshoot the Dify Connector](/docs/raw/connectors/dify/troubleshooting.md): Use this runbook to identify the failed boundary and choose the narrowest safe action. Pause related mutations whenever the provider outcome is uncertain.
- [Qualify the Dify Connector](/docs/raw/connectors/dify/qualification.md): Use this page to decide whether one exact Dify connector artifact supports a bounded release claim. Keep source tests, image checks, controlled fixtures, and actual Dify observations as separate evidence classes.
- [Dify Connector Changelog](/docs/raw/connectors/dify/changelog.md): Use this page to decide whether a Dify connector change requires a new artifact, admission record, migration, caller update, or qualification result. It records only changes established from the reviewed Git range.
- [CrewAI Connector](/docs/raw/connectors/crewai.md): The CrewAI connector exposes deployment-admitted crews as governed AIP capabilities. Use it when a tenant must run, observe, cancel, train, test, query, or reset an identified crew through a durable boundary.
- [Legacy CrewAI Connector Overview](/docs/raw/connectors/crewai/legacy-overview.md): The CrewAI integration has two deployment components:
- [Run the CrewAI Connector Quickstart](/docs/raw/connectors/crewai/getting-started/quickstart.md): In this tutorial, you will start the repository's deterministic CrewAI product path and route one governed run through the standalone Rust host and Python sidecar.
- [CrewAI Authentication and Admitted Crews](/docs/raw/connectors/crewai/getting-started/authentication-and-admitted-crews.md): The CrewAI integration has three distinct trust boundaries. Keeping them separate prevents an AIP caller from selecting Python code, changing the sidecar identity, or supplying deployment credentials through an Action.
- [CrewAI Connector Configuration](/docs/raw/connectors/crewai/reference/configuration.md): This reference describes the product settings accepted by aip-host-crewai and aip-crewai-sidecar at the reviewed source revision. A complete deployment combines them with the shared connector-host identity, PostgreSQL, routing, signing, lif
- [CrewAI Capabilities](/docs/raw/connectors/crewai/capabilities.md): Use this index to choose the CrewAI operation that owns a task and interpret its AIP contract before invocation. The reviewed connector has ten frozen operations, but each configured crew publishes only its admitted subset.
- [CrewAI Run Lifecycle and Replay](/docs/raw/connectors/crewai/capabilities/run-lifecycle-and-replay.md): These six operations start CrewAI work or inspect and control its durable sidecar job. Use one stable AIP Action ID and idempotency key for each start, then target that Action ID from separate status, event, or cancel Actions.
- [CrewAI Training and Testing](/docs/raw/connectors/crewai/capabilities/training-and-testing.md): The train and test operations run bounded CrewAI evaluation workflows through the durable sidecar job boundary. Both can invoke models, tools, and crew tasks repeatedly, so iteration count is part of approval and cost review.
- [CrewAI Knowledge and Memory](/docs/raw/connectors/crewai/capabilities/knowledge-and-memory.md): Use knowledgequery to retrieve bounded results from a configured crew knowledge store without running its task graph. Use memoryreset to clear one explicit CrewAI memory domain or all available domains.
- [Run a CrewAI Crew and Stream Events](/docs/raw/connectors/crewai/guides/run-a-crew-and-stream-events.md): Use this guide to execute one admitted CrewAI crew, render its ordered progress, and reach a verified terminal result. The same flow recovers after an application disconnect without starting a second crew run.
- [Train, Test, and Recover a CrewAI Crew](/docs/raw/connectors/crewai/guides/train-test-and-recover.md): Use this procedure to create one controlled CrewAI training artifact, evaluate the crew in a separate test Action, and retain enough evidence to accept or recover each outcome.
- [CrewAI Sidecar Contract and Lockfile](/docs/raw/connectors/crewai/reference/sidecar-contract-and-lockfile.md): This reference identifies the Python artifact behind the CrewAI connector and the private HTTP contract used by its Rust host. Use it to compare image provenance, request shapes, durable records, and failure responses.
- [CrewAI Idempotency, Cancellation, and Process Fencing](/docs/raw/connectors/crewai/reference/idempotency-cancellation-and-process-fencing.md): This reference explains how the CrewAI connector fences duplicate work, persists job identity, and reports cancellation. Use it to recover an Action without treating a transport result as proof of provider behavior.
- [Deploy the CrewAI Connector](/docs/raw/connectors/crewai/operations/deployment.md): Use this procedure to deploy one CrewAI connector instance as two separately identified processes. The Rust host owns AIP identity and lifecycle. The Python sidecar owns CrewAI execution and its durable job journal.
- [CrewAI Health, Observability, and Recovery](/docs/raw/connectors/crewai/operations/health-observability-and-recovery.md): Use this runbook when a CrewAI replica is unhealthy, not ready, losing its lease, or holding an Action without trustworthy terminal evidence. Preserve the host runtime and sidecar journal before changing the deployment.
- [Troubleshoot the CrewAI Connector](/docs/raw/connectors/crewai/troubleshooting.md): Use this index to move from one observed CrewAI connector symptom to a bounded check and safe response. Preserve the original Action, host runtime, and sidecar journal before restarting, retrying, archiving, or replacing state.
- [Qualify the CrewAI Connector](/docs/raw/connectors/crewai/qualification.md): Use this procedure to evaluate one exact CrewAI host and sidecar pair. The result applies only to the recorded artifacts, configuration, topology, CrewAI package set, crew code, and executed cases.
- [CrewAI Connector Changelog](/docs/raw/connectors/crewai/changelog.md): This changelog records source-confirmed CrewAI connector baselines and the operator action required by each change. Artifact digests and retained release evidence, not the version string alone, identify a deployable release.
- [Twenty Connector](/docs/raw/connectors/twenty.md): The Twenty connector exposes a fixed workspace-aware REST surface as governed AIP capabilities. Use it to manage standard or custom records, inspect or change selected workspace metadata, read generated OpenAPI documents, and ingest signed 
- [Legacy Twenty Connector Overview](/docs/raw/connectors/twenty/legacy-overview.md): Twenty Connector in the Agent Interoperability Protocol documentation.
- [Run the Twenty Connector Quickstart](/docs/raw/connectors/twenty/getting-started/quickstart.md): In this tutorial, you will read one Twenty collection and create one record through an already admitted connector. You will then read the original AIP Action until its governed mutation reaches a terminal state.
- [Authenticate and Isolate a Twenty Workspace](/docs/raw/connectors/twenty/getting-started/authentication-and-workspace-isolation.md): Use this guide to bind one standalone Twenty host to one verified AIP tenant, one fixed provider origin, and one Twenty workspace. The API token stays in an owner-controlled file and is added to provider requests inside the connector.
- [Twenty Connector Configuration](/docs/raw/connectors/twenty/reference/configuration.md): This reference describes the product settings accepted by aip-host-twenty at the reviewed source revision. A complete deployment combines them with the shared connector-host identity, PostgreSQL, routing, signing, lifecycle, and optional ca
- [Twenty Capabilities](/docs/raw/connectors/twenty/capabilities.md): Use this index to choose the Twenty operation family that owns a task and interpret its AIP contract before invocation. The reviewed connector contains 22 fixed operations across records, workspace metadata, and OpenAPI documents.
- [Twenty Record Query and Aggregation](/docs/raw/connectors/twenty/capabilities/record-query-and-aggregation.md): Use these four read operations to inspect records, fetch one stable ID, detect duplicate candidates, or request provider grouping. They accept standard and custom object names through fixed Twenty REST routes.
- [Twenty Record Create, Update, and Batch Upsert](/docs/raw/connectors/twenty/capabilities/record-create-update-and-batch-upsert.md): Use these four mutations to create or update standard and custom Twenty records. The connector binds each request to a fixed workspace, requires approval and idempotency, and adds deterministic create IDs when needed.
- [Twenty Record Delete, Restore, and Merge](/docs/raw/connectors/twenty/capabilities/record-delete-restore-and-merge.md): Use these seven mutations to change record lifecycle or consolidate duplicate records. Choose the narrowest operation, freeze its exact target, and preserve the original Action and idempotency identity through verification.
- [Twenty Workspace Metadata](/docs/raw/connectors/twenty/capabilities/workspace-metadata.md): Use these five operations to read or change one of twelve fixed Twenty metadata resource types. The connector selects the route from a closed resource enum and applies resource-specific list-query guards.
- [Twenty OpenAPI Documents](/docs/raw/connectors/twenty/capabilities/openapi-documents.md): Use these two Resource capabilities to read the configured workspace's current core or metadata OpenAPI document. The connector fixes both provider routes and adds its host-owned workspace credential.
- [Manage Standard and Custom Twenty Records](/docs/raw/connectors/twenty/guides/manage-standard-and-custom-records.md): Use this guide to manage one standard or custom object through a current schema snapshot, bounded record discovery, duplicate review, one governed mutation, and a provider read that verifies the intended result.
- [Manage Twenty Views, Layouts, and Webhook Metadata](/docs/raw/connectors/twenty/guides/manage-views-layouts-and-webhook-metadata.md): Use this guide to change one view, page-layout hierarchy, or provider webhook registration. Begin with the current workspace metadata schema and dependency graph, submit one critical mutation, then verify the provider result through a separ
- [Twenty Webhook Security, Replay, and Events](/docs/raw/connectors/twenty/reference/webhook-security-replay-and-events.md): This reference defines the reviewed Twenty inbound webhook boundary. It covers route activation, exact raw-body HMAC, clock and workspace checks, durable nonce replay fencing, deterministic event mapping, local acceptance, and central AIP p
- [Twenty Provider Boundaries and Exclusions](/docs/raw/connectors/twenty/reference/provider-boundaries-and-exclusions.md): Use this reference to decide whether an intended integration belongs inside the reviewed Twenty connector. The connector is a fixed, workspace-aware mapping; it is not a generic transport for the complete provider API.
- [Deploy the Twenty Connector](/docs/raw/connectors/twenty/operations/deployment.md): Use this procedure to deploy one standalone aip-host-twenty instance. The host owns AIP identity, fixed provider workspace access, durable Action state, optional webhook replay, local events, and central publication.
- [Twenty Health, Observability, and Recovery](/docs/raw/connectors/twenty/operations/health-observability-and-recovery.md): Use this runbook when a Twenty replica is unhealthy, not ready, losing its lease, holding an Action without trustworthy terminal evidence, or failing to publish a verified webhook event.
- [Troubleshoot the Twenty Connector](/docs/raw/connectors/twenty/troubleshooting.md): Use this runbook to identify the first safe check for a Twenty connector failure. Preserve original instance, replica, Action, key, provider, replay, event, and outbox identities before changing configuration or state.
- [Qualify the Twenty Connector](/docs/raw/connectors/twenty/qualification.md): Use this procedure to evaluate one exact Twenty connector deployment. The result applies only to the recorded artifact, manifest, provider revision, workspace, credentials, configuration, topology, and executed cases.
- [Twenty Connector Changelog](/docs/raw/connectors/twenty/changelog.md): Use this page to decide whether a Twenty connector change requires a new artifact, admission record, migration, caller update, or qualification result. It records only changes established from the reviewed Git history.
- [Build a Connector](/docs/raw/guides/build-a-connector.md): Use this guide to turn a bounded provider API into an AIP connector crate and a standalone connector-host binary. It is for Rust developers who own the provider mapping and can coordinate release, security, and deployment evidence.
- [Upgrade a Connector to the Frozen SDK](/docs/raw/guides/upgrade-connector-sdk.md): Use this guide to move an existing Rust connector from legacy connector or direct ActionHandler execution to the FrozenConnector boundary. It is for connector developers who need to preserve published capability behavior while adopting trus
- [Migrate from Bundled Connectors to the Fleet](/docs/raw/guides/migrate-bundled-connectors.md): Use this guide to move one product connector from a getaip-server process into an admitted standalone connector host. It is for operators and connector owners who can control the gateway, registry, host, provider credentials, ingress, and r

## Deploy and Operate

- [Production Deployment](/docs/raw/guides/production-deployment.md): Use this guide to design and stage a production AIP deployment. It is for operators who own the authenticated edge, product-neutral getaip-server replicas, durable state, connector registry, lifecycle control plane, standalone connector hos
- [Deploy the Connector Fleet](/docs/raw/guides/deploy-connector-fleet.md): Use this guide to deploy one connector type, provider boundary, tenant-owned instance, and standalone host replica behind the product-neutral getaip-server gateway. It is for operators who control registry migration, signed admission, gatew
- [Connector Host Configuration](/docs/raw/reference/connector-host-configuration.md): This reference lists the common process configuration shared by every standalone connector host. Use it to configure the host identity, admitted artifact, tenant boundary, durable storage, central services, delivery routes, and lifecycle li
- [Connector Credentials and Rotation](/docs/raw/guides/connector-credentials-and-rotation.md): Use this guide to replace one provider credential used by a standalone connector host. The procedure creates a new tenant-owned connector instance, starts replicas with the new material, changes new route assignments to the new credential r
- [Connector Host Lifecycle](/docs/raw/guides/connector-host-lifecycle.md): Use this guide to move one standalone connector-host replica through startup, readiness, normal service, graceful drain, offline retention, and recovery. It applies to the common host used by Cal.diy, Hermes Agent, Chatwoot, Dify, CrewAI, a
- [Connector Orchestration and Rollback](/docs/raw/guides/connector-orchestration.md): Use this guide to reconcile one connector instance through an external workload platform without giving that platform connector-registry administrator access. The procedure derives a deterministic plan from a verified admission package, an 
- [Legacy Bundled Deployment](/docs/raw/guides/legacy-bundled-deployment.md): Use this guide to keep an existing bundled Cal.diy or Hermes Agent deployment controlled while you prepare its migration to standalone connector hosts. The procedure freezes the compatibility baseline, preserves durable state, limits the ex
- [GetAIP Server Configuration](/docs/raw/reference/configuration.md): This reference lists the complete command-line configuration of the product-neutral getaip-server gateway. Use it to look up process identity, native authentication, durable state, callbacks, connector-fleet routing, NATS, delegation, and M
- [Environment Variables](/docs/raw/reference/environment-variables.md): This reference lists all 86 environment variables consumed by the product-neutral getaip-server gateway and separates them from connector control-plane, standalone host, product connector, and getaip settings. Use it to identify which proce
- [Observe and Recover](/docs/raw/guides/observe-and-recover.md): Use this guide when an AIP action is delayed, interrupted, or has an uncertain external outcome. You will establish whether the process can make progress, read the durable action and delivery records, choose a recovery action, and verify th
- [Operator Runbooks](/docs/raw/guides/operator-runbooks.md): Use these runbooks for the first response to a common getaip-server, connector-fleet, or connector-host incident. Select the symptom below, contain new effects, use the recorded checks to identify the failing owner, and stop when the eviden

## API Reference

- [Native HTTP API](/docs/raw/reference/http-api.md): This reference lists the native HTTP surface implemented by the product-neutral getaip-server gateway. Use it to select a route, identify its application authentication boundary, construct query parameters, and interpret the current respons
- [Connector Fleet HTTP API](/docs/raw/reference/connector-fleet-api.md): This reference describes the HTTP contracts used inside an AIP connector fleet. Use it to identify the correct process and route, construct a signed request, and interpret route, lease, replay, and capacity failures.
- [Transport Bindings](/docs/raw/reference/transport-bindings.md): Transport bindings move AIP messages without changing their protocol meaning. Use this reference to choose a binding, encode its frames, authenticate its peer, and recover after an interrupted delivery.
- [GetAIP CLI](/docs/raw/reference/cli.md): getaip is the developer and operator client for native AIP, NATS, Model Context Protocol (MCP), schemas, connector qualification, connector admission, orchestration plans, and receipt verification. This reference covers all 15 command famil
- [Errors and Retry Decisions](/docs/raw/reference/errors.md): AIP errors separate machine decisions from diagnostic text. Read the protocol code, category, retry hint, action or transaction state, and connector outcome evidence before deciding to correct, authorize, retry, reconcile, or stop.
- [Metrics Reference](/docs/raw/reference/metrics.md): AIP exposes process health, work capacity, fleet state, and bounded activity counters through three different metric surfaces. This reference is for operators who need to select the correct endpoint, recognize every current product-neutral 
- [Glossary](/docs/raw/reference/glossary.md): Use this glossary to resolve terms used across the AIP 1.0 protocol, the reviewed Rust implementation, and connector-fleet operations. It is for developers, operators, evaluators, and reviewers who need to distinguish similar concepts befor

## Architecture and Security

- [Architecture Index](/docs/raw/architecture.md): Use this index to choose the architecture page that answers a component, execution, connector-fleet, trust, or dependency question. It is for developers, connector developers, operators, security reviewers, and protocol implementers who nee
- [Architecture Overview](/docs/raw/architecture/overview.md): Use this page to understand which component owns each decision and state transition in the reviewed AIP implementation. It is for developers, connector developers, operators, and security reviewers who need a complete component map before c
- [Connector Fleet Architecture](/docs/raw/architecture/connector-fleet.md): Use this page to understand how the reviewed AIP implementation can add, isolate, and scale connector processes without adding product code or provider secrets to the central getaip-server binary. It is primarily for developers designing th
- [Connector Registry and Routing](/docs/raw/architecture/connector-registry-and-routing.md): Use this page to understand how the reviewed AIP connector registry turns an authenticated tenant and capability into one durable execution target. It explains which records participate, how discovery remains revision-consistent, how a new 
- [Connector Admission and Supply-Chain Trust](/docs/raw/architecture/connector-admission.md): Use this page to understand how the reviewed AIP implementation converts a signed connector release and independently trusted evidence into registry state that can receive traffic. It is primarily for security reviewers; connector developer
- [Gateway](/docs/raw/architecture/gateway.md): Use this page to understand how the reviewed gateway converts a native AIP envelope plus transport-established trust into the context used by the runtime. It explains where signatures, replay protection, identity and tenant resolution, quer
- [Runtime](/docs/raw/architecture/runtime.md): Use this page to understand what the reviewed AIP runtime owns between trusted gateway dispatch and connector execution. It explains how capability contracts, policy, idempotency, approvals, transactions, queues, callbacks, and recovery for
- [Dependency Graph](/docs/raw/architecture/dependency-graph.md): Use this page to decide where a change belongs in the reviewed Rust workspace and which lower-level contracts it may depend on. The graph groups crates by architectural responsibility so native AIP semantics do not drift into a transport, c
- [Security Model](/docs/raw/architecture/security-model.md): Use this page to evaluate the security boundaries in the reviewed AIP implementation. It identifies protected assets, attacker capabilities, deployment assumptions, fail-closed decisions, component-compromise scope, and risks that remain af

## Protocol Standard

- [Specification Suite](/docs/raw/spec.md): Use this index to choose the authoritative document for an AIP 1.0 behavior, wire shape, compatibility mapping, connector boundary, conformance claim, or implementation-status question. It is primarily for protocol implementers and reviewer
- [AIP 1.0 Normative Specification](/docs/raw/spec/aip-1.0.md): Agent Interoperability Protocol (AIP) defines a transport-independent semantic contract for discovering and invoking capabilities across agents, tools, workflow systems, customer channels, and enterprise applications.
- [JSON Schemas](/docs/raw/spec/schemas.md): Use this reference to select an AIP 1.0 schema, reproduce the committed 52-file catalog, validate data at the correct layer, and record enough provenance to identify the exact bytes you used. It is for protocol implementers and release revi
- [Compatibility Profiles](/docs/raw/spec/compatibility-profiles.md): Use this reference to select an implemented AIP profile, match its stable version and route surface, and identify which semantics remain native AIP responsibilities. It covers the three compatibility profiles and four native transport profi
- [Connector Contract](/docs/raw/spec/connector-contract.md): Use this reference to implement or review a product connector without recreating AIP governance at the provider edge. It identifies the callable surface, the trusted context a connector receives, the behavior it may advertise, and the evide
- [Conformance and Qualification](/docs/raw/reference/conformance.md): Use this reference to select the evidence layer that matches a claim, run the available check correctly, and report its result against an exact artifact. Conformance asks whether a named contract is satisfied. Qualification asks whether an 

## Qualification and Evidence

- [Implementation Status](/docs/raw/reference/implementation-status.md): Use this page to determine what the pinned AIP source implements and what its available evidence can establish. The key distinction is simple: source code, tests, and workflow definitions show that a behavior or check exists. They do not sh
- [Testing and Evidence Index](/docs/raw/testing.md): Use this index to choose the smallest procedure that can support your claim and to find the evidence you must retain. Start with the claim, not with whichever test command is easiest to run.
- [Connector Fleet Qualification and Failure Matrix](/docs/raw/testing/connector-fleet-qualification.md): Use this procedure to evaluate routing, lifecycle, trust, persistence, and failure recovery across a complete isolated connector fleet. The source-owned matrix exercises real AIP processes and PostgreSQL state against controlled provider be
- [Fleet Scale SLO](/docs/raw/testing/fleet-scale-slo.md): Use this procedure to check that connector catalog and route resolution remain within the source-owned aip-fleet-slo-v1 regression bounds. The gate uses a real PostgreSQL database and synthetic registry records. It measures registry behavio
- [Pinned Upstream Baselines](/docs/raw/connectors/upstream-baselines.md): Use this reference to identify the official product source revision against which each maintained connector mapping was reviewed. A baseline freezes the route and behavior audit. It does not vendor the product, add it to the AIP build, or i
- [Live Product E2E](/docs/raw/testing/live-product-e2e.md): Use this procedure to prove that one identified AIP build can interact safely with one identified deployment of Cal.diy, Hermes Agent, Chatwoot, Dify, CrewAI, or Twenty. A valid campaign exercises a real product boundary in an isolated envi
- [Cal.diy Isolated-Live Evidence](/docs/raw/testing/cal-diy-isolated-live.md): Historical result: PASS. The retained artifact records qualification at 2026-07-13T12:05:20.671410Z and restart verification at 2026-07-13T12:05:26.883165Z.
- [Hermes Isolated-Live Evidence](/docs/raw/testing/hermes-agent-isolated-live.md): Historical reported result: PASS. The retained narrative gives the execution date as 11 July 2026 but does not retain an exact UTC start or end.

## Project and Releases

- [Release Notes](/docs/raw/reference/release-notes.md): The current software release is GetAIP Core 2.0.0. It changes implementation names and software packaging while the Agent Interoperability Protocol remains version 1.0.
- [Release Artifacts and Verification](/docs/raw/reference/release-artifacts.md): Use this page to identify the exact GetAIP Core v2.0.0 source release and the bounded qualification evidence retained for it. Never infer a public image, signature, SBOM, attestation, native archive, or private package from the presence of 
- [License and Usage Terms](/docs/raw/reference/licensing.md): This page is a plain-language guide to the legal files retained with AIP Core 1.0.0. It is not legal advice and does not replace, modify, or grant rights beyond those files. If this page and a controlling license differ, the controlling lic
- [Contributing](/docs/raw/project/contributing.md): Use this guide to prepare a focused AIP code or documentation change whose implementation, tests, schemas, documentation, and evidence agree. A change is ready for review only when its claims are no broader than the exact artifacts and chec
- [Code of Conduct](/docs/raw/project/code-of-conduct.md): The Agent Interoperability Protocol project is committed to a professional, inclusive, and safe technical community. Everyone who participates in a project space must respect other people and the users affected by protocol, security, connec
- [Security Policy](/docs/raw/project/security.md): If you believe you found a vulnerability in AIP, email with the subject prefix [SECURITY]. Do not open a public issue or publish credentials, customer data, exploit payloads, or affected deployment identifiers before remediation has been co