# Duckie | Docs

## Docs

- [Welcome to Duckie](https://docs.duckie.ai/introduction.md): AI-powered customer support that works alongside your team
- [Quickstart](https://docs.duckie.ai/quickstart.md): Get up and running with Duckie in 10 minutes
- [Launchpad](https://docs.duckie.ai/launchpad.md): Train Duckie from past support conversations and review generated content
- [Core Concepts](https://docs.duckie.ai/core-concepts.md): Understand the key building blocks of Duckie
- [How Duckie Works](https://docs.duckie.ai/how-duckie-works.md): Understanding the flow from message to response
- [Agents Overview](https://docs.duckie.ai/agents/overview.md): Learn about Duckie agents and what they can do
- [Creating Agents](https://docs.duckie.ai/agents/creating-agents.md): Step-by-step guide to creating your first agent
- [Agent Configuration](https://docs.duckie.ai/agents/configuration.md): Deep dive into all agent settings
- [Autonomous Agents](https://docs.duckie.ai/agents/autonomous-agents.md): Configure agents that decide which resources and tools to use at run time
- [Duckie Assistant Agents](https://docs.duckie.ai/agents/duckie-assistant-agents.md): Configure Duckie copilot agents that work with Duckie objects
- [Sub-Agents](https://docs.duckie.ai/agents/sub-agents.md): Delegate work from an autonomous agent or workflow to another agent
- [Workflows Overview](https://docs.duckie.ai/workflows/overview.md): Visual, graph-based automations for precise control
- [Creating Workflows](https://docs.duckie.ai/workflows/creating-workflows.md): Build visual workflows for your agents
- [Nodes and Conditions](https://docs.duckie.ai/workflows/nodes-and-conditions.md): Reference for workflow node types and conditions
- [Runbooks Overview](https://docs.duckie.ai/runbooks/overview.md): Flexible, AI-driven procedures for your agents
- [Creating Runbooks](https://docs.duckie.ai/runbooks/creating-runbooks.md): Build effective runbooks for your agents
- [Snippets](https://docs.duckie.ai/runbooks/snippets.md): Reusable instruction fragments for runbooks
- [Tools Overview](https://docs.duckie.ai/tools/overview.md): Actions your agents can perform
- [Duckie Tools](https://docs.duckie.ai/tools/duckie-tools.md): Built-in tools available in Duckie
- [App Tools](https://docs.duckie.ai/tools/app-tools.md): Tools from connected apps and Duckie-provided app platforms
- [Custom Tools](https://docs.duckie.ai/tools/custom-tools.md): Define HTTP API tools for your agents
- [MCP Servers](https://docs.duckie.ai/tools/mcp-servers.md): Discover tools from Model Context Protocol servers
- [Knowledge Base Overview](https://docs.duckie.ai/knowledge/overview.md): Give your agent the information it needs
- [Connecting Knowledge Sources](https://docs.duckie.ai/knowledge/connecting-sources.md): Sync external documentation to your knowledge base
- [Custom Knowledge](https://docs.duckie.ai/knowledge/custom-knowledge.md): Create knowledge articles directly in Duckie
- [Knowledge Gaps](https://docs.duckie.ai/knowledge/knowledge-gaps.md): Fill gaps in your agent's knowledge
- [Knowledge Tags](https://docs.duckie.ai/knowledge/tags.md): Organize and filter knowledge for agents
- [Guidelines Overview](https://docs.duckie.ai/guidelines/overview.md): Shape how your agent communicates
- [Creating Guidelines](https://docs.duckie.ai/guidelines/creating-guidelines.md): Write effective guidelines for your agents
- [Guardrails Overview](https://docs.duckie.ai/guardrails/overview.md): Keep your agent safe and compliant
- [Escalation Rules](https://docs.duckie.ai/guardrails/escalation-rules.md): Define when your agent should hand off to humans
- [Restrictions](https://docs.duckie.ai/guardrails/restrictions.md): Hard limits on what your agent cannot do
- [Tagging & Classification Overview](https://docs.duckie.ai/tagging/overview.md): Automatically organize and analyze conversations
- [Categories](https://docs.duckie.ai/tagging/categories.md): High-level conversation classification
- [Attributes](https://docs.duckie.ai/tagging/attributes.md): Custom fields for granular tagging
- [Resolution Rules](https://docs.duckie.ai/tagging/resolution-rules.md): Define what "resolved" means for your team
- [Deployments Overview](https://docs.duckie.ai/deployments/overview.md): Connect agents to real customer channels
- [Creating Deployments](https://docs.duckie.ai/deployments/creating-deployments.md): Deploy your agent to a channel
- [Scheduled Deployments](https://docs.duckie.ai/deployments/scheduled-deployments.md): Run an agent on a recurring schedule
- [Custom Webhooks](https://docs.duckie.ai/deployments/custom-webhooks.md): Trigger an agent from a custom JSON webhook
- [Deployment Modes](https://docs.duckie.ai/deployments/deployment-modes.md): Control how your agent interacts with customers
- [Triggers and Events](https://docs.duckie.ai/deployments/triggers-and-events.md): Configure when your agent responds
- [Self-Hosting](https://docs.duckie.ai/self-hosting.md): Run Duckie in your infrastructure
- [Testing Overview](https://docs.duckie.ai/testing/overview.md): Validate your agent before going live
- [Replay Testing](https://docs.duckie.ai/testing/replay-testing.md): Test agents against real historical conversations
- [Playground](https://docs.duckie.ai/testing/playground.md): Interactive testing for real-time validation
- [Batch Testing](https://docs.duckie.ai/testing/batch-testing.md): Automated regression testing for agents
- [Analytics Overview](https://docs.duckie.ai/analytics/overview.md): Understand and improve agent performance
- [Performance Metrics](https://docs.duckie.ai/analytics/performance-metrics.md): Track agent volume, resolution, deflection, escalation, and timing
- [Breakdown Analysis](https://docs.duckie.ai/analytics/breakdown.md): Visualize ticket distribution
- [Run History](https://docs.duckie.ai/analytics/runs.md): Complete execution history
- [Alerts](https://docs.duckie.ai/analytics/alerts.md): Configure guardrail, tool failure, and agent-triggered alerts
- [Security Overview](https://docs.duckie.ai/security/overview.md): Configure Duckie with scoped access, safe tools, guardrails, testing, and review workflows
- [Access Control](https://docs.duckie.ai/security/access-control.md): Apply least privilege across teammates, API keys, MCP clients, integrations, agents, and approvals
- [Tool & Integration Security](https://docs.duckie.ai/security/tool-and-integration-security.md): Configure tools, credentials, approvals, and external actions safely
- [AI Safety & Prompt Injection](https://docs.duckie.ai/security/ai-safety.md): Reduce prompt-injection and misuse risk with scoped tools, workflows, guardrails, and testing
- [Account-Safe Actions](https://docs.duckie.ai/security/account-safe-actions.md): Configure support agents to act on the right customer, account, and record
- [Managing Connections](https://docs.duckie.ai/settings/connections.md): Connect and manage your integrations
- [Organization Settings](https://docs.duckie.ai/settings/organization.md): Manage your organization's configuration
- [Workspaces](https://docs.duckie.ai/settings/workspaces.md): Create and manage isolated Duckie workspaces under your organization
- [Team Members & Roles](https://docs.duckie.ai/settings/members.md): Manage who has access to your Duckie workspace
- [Overview](https://docs.duckie.ai/concepts/overview.md): Decision guides for designing effective Duckie agent systems
- [Choosing an Agent Architecture](https://docs.duckie.ai/concepts/choosing-an-agent-architecture.md): Decide how to combine agents, workflows, runbooks, tools, and deployments
- [Autonomous Agents vs Workflows vs Runbooks](https://docs.duckie.ai/concepts/autonomous-agents-vs-workflows-vs-runbooks.md): Choose the right Duckie building block for flexible support, repeatable processes, and procedural guidance
- [Single Agents vs Sub-Agents](https://docs.duckie.ai/concepts/single-agents-vs-sub-agents.md): Decide when one agent should own a conversation and when to delegate focused work to specialist agents
- [Routing and Orchestration Patterns](https://docs.duckie.ai/concepts/routing-and-orchestration-patterns.md): Choose how Duckie sends work to the right agent, workflow, or human review path
- [Context, Knowledge, and Memory](https://docs.duckie.ai/concepts/context-knowledge-and-memory.md): Understand what Duckie knows now, what it can look up, and what carries forward in a ticket
- [Agent Tables](https://docs.duckie.ai/concepts/agent-tables.md): Use scoped structured tables as working memory for autonomous agents
- [Tools, Permissions, and Side Effects](https://docs.duckie.ai/concepts/tools-permissions-and-side-effects.md): Understand what agents can do, how access is controlled, and how to test actions safely
- [Guardrails, Escalation, and Human Review](https://docs.duckie.ai/concepts/guardrails-escalation-and-human-review.md): Know when Duckie should continue, stop, ask for approval, or hand off to a teammate
- [Workspaces and System Boundaries](https://docs.duckie.ai/concepts/workspaces-and-system-boundaries.md): Decide how to separate teams, products, regions, and support motions in Duckie
- [Testing, Observability, and Iteration](https://docs.duckie.ai/concepts/testing-observability-and-iteration.md): Use Duckie's testing and analytics surfaces to launch agents safely, monitor quality, and improve over time
- [Integrations Overview](https://docs.duckie.ai/integrations/overview.md): Connect Duckie to your existing tools
- [Supported Integrations](https://docs.duckie.ai/integrations/supported-integrations.md): Complete list of integrations and their capabilities
- [Zendesk](https://docs.duckie.ai/integrations/ticketing/zendesk.md): Connect Zendesk to manage support tickets with Duckie
- [Freshdesk](https://docs.duckie.ai/integrations/ticketing/freshdesk.md): Connect Freshdesk to manage support tickets with Duckie
- [Intercom](https://docs.duckie.ai/integrations/ticketing/intercom.md): Connect Intercom to manage customer conversations with Duckie
- [Plain](https://docs.duckie.ai/integrations/ticketing/plain.md): Connect Plain to manage customer support threads with Duckie
- [Pylon](https://docs.duckie.ai/integrations/ticketing/pylon.md): Connect Pylon to manage customer issues with Duckie
- [HubSpot](https://docs.duckie.ai/integrations/ticketing/hubspot.md): Connect HubSpot to manage CRM and support with Duckie
- [Slack](https://docs.duckie.ai/integrations/messaging/slack.md): Connect Slack for real-time messaging with Duckie
- [Discord](https://docs.duckie.ai/integrations/messaging/discord.md): Add the Duckie bot to your Discord server
- [Microsoft Teams](https://docs.duckie.ai/integrations/messaging/teams.md): Connect Microsoft Teams for enterprise messaging with Duckie
- [Gmail](https://docs.duckie.ai/integrations/messaging/gmail.md): Connect Gmail to send and receive emails with Duckie
- [Website Widget](https://docs.duckie.ai/integrations/messaging/widget.md): Embed a Duckie chat widget on your website
- [Notion](https://docs.duckie.ai/integrations/knowledge/notion.md): Sync Notion pages and databases to your Duckie knowledge base
- [Confluence](https://docs.duckie.ai/integrations/knowledge/confluence.md): Sync Confluence spaces and pages to your Duckie knowledge base
- [Google Drive](https://docs.duckie.ai/integrations/knowledge/google-drive.md): Sync Google Drive documents to your Duckie knowledge base
- [Sentry](https://docs.duckie.ai/integrations/knowledge/sentry.md): Connect Sentry to monitor errors and provide context to Duckie
- [Fireflies](https://docs.duckie.ai/integrations/knowledge/fireflies.md): Connect Fireflies to capture meeting transcripts for Duckie
- [Guru](https://docs.duckie.ai/integrations/knowledge/guru.md): Connect Guru to sync knowledge cards with Duckie
- [Skilljar](https://docs.duckie.ai/integrations/knowledge/skilljar.md): Connect Skilljar to sync courses and training content with Duckie
- [Jira](https://docs.duckie.ai/integrations/issues/jira.md): Connect Jira to manage issues and track work with Duckie
- [Linear](https://docs.duckie.ai/integrations/issues/linear.md): Connect Linear to manage issues and projects with Duckie
- [GitHub](https://docs.duckie.ai/integrations/issues/github.md): Connect GitHub to access repositories and issues with Duckie
- [Duckie API Overview](https://docs.duckie.ai/api/overview.md): Export Duckie data and connect MCP clients
- [Authentication](https://docs.duckie.ai/api/authentication.md): Authenticate Duckie API requests with bearer API keys
- [MCP Endpoint](https://docs.duckie.ai/api/mcp.md): Connect MCP clients to Duckie runs and core objects
- [Pagination and Filtering](https://docs.duckie.ai/api/pagination-and-filtering.md): Page through API results and filter run exports
- [Rate Limits](https://docs.duckie.ai/api/rate-limits.md): Understand API request limits and retry headers
- [Error Responses](https://docs.duckie.ai/api/errors.md): Handle API errors consistently
- [API-Key Management](https://docs.duckie.ai/api/api-key-management.md): Create, copy, rename, expire, and revoke Duckie API keys
- [List runs](https://docs.duckie.ai/api-reference/runs/list-runs.md): Returns a cursor-paginated list of run summaries. The list response does not include conversation messages or execution steps.
- [Retrieve a run](https://docs.duckie.ai/api-reference/runs/retrieve-a-run.md): Returns one run with conversation messages, execution steps, evaluation attributes, and redacted tool payloads.
- [List run tool filter options](https://docs.duckie.ai/api-reference/runs/list-run-tool-filter-options.md): Returns historical tool-call names that can be used with the `tool_name` run filter.
- [List tools](https://docs.duckie.ai/api-reference/tools/list-tools.md): Returns the organization's tool catalog, including Duckie tools, app tools, custom tools, and MCP tools.
- [Retrieve a tool](https://docs.duckie.ai/api-reference/tools/retrieve-a-tool.md)
- [List agents](https://docs.duckie.ai/api-reference/agents/list-agents.md)
- [Retrieve an agent](https://docs.duckie.ai/api-reference/agents/retrieve-an-agent.md)
- [List guidelines](https://docs.duckie.ai/api-reference/guidelines/list-guidelines.md): Returns published guidelines.
- [Retrieve a guideline](https://docs.duckie.ai/api-reference/guidelines/retrieve-a-guideline.md)
- [List guardrails](https://docs.duckie.ai/api-reference/guardrails/list-guardrails.md)
- [Retrieve a guardrail](https://docs.duckie.ai/api-reference/guardrails/retrieve-a-guardrail.md)
- [List runbooks](https://docs.duckie.ai/api-reference/runbooks/list-runbooks.md): Returns published, non-deleted runbooks.
- [Retrieve a runbook](https://docs.duckie.ai/api-reference/runbooks/retrieve-a-runbook.md)
- [Overview](https://docs.duckie.ai/examples/overview.md): Example Duckie patterns and architectures for real customer operations
- [Designing Agent Systems](https://docs.duckie.ai/examples/designing-agent-systems.md): Combine deterministic workflows with autonomous agents for reliable operations
- [Duckie as an External Subagent](https://docs.duckie.ai/examples/duckie-as-external-subagent.md): Call a focused Duckie agent from a customer-owned orchestration agent
- [Overview](https://docs.duckie.ai/examples/ticket-triage/overview.md): A blueprint for support ticket triage with Duckie
- [MVP](https://docs.duckie.ai/examples/ticket-triage/mvp.md): The smallest useful autonomous ticket triage system before broader escalation, reporting, and feedback loops
- [System Map](https://docs.duckie.ai/examples/ticket-triage/system-map.md): The multi-channel, multi-agent architecture for support ticket triage
- [Intake and Customer Context](https://docs.duckie.ai/examples/ticket-triage/intake-and-customer-context.md): Normalize new tickets and attach the context needed for reliable triage
- [Intent, Tagging, and Routing](https://docs.duckie.ai/examples/ticket-triage/intent-tagging-and-routing.md): Classify support tickets and turn the classification into deterministic routing
- [Escalation and Handoff](https://docs.duckie.ai/examples/ticket-triage/escalation-and-handoff.md): Route unclear or high-risk tickets to human reviewers and specialist teams
- [Reporting and Feedback](https://docs.duckie.ai/examples/ticket-triage/reporting-and-feedback.md): Use scheduled reports and run review to improve ticket triage
- [Rollout Plan](https://docs.duckie.ai/examples/ticket-triage/rollout-plan.md): Move ticket triage from shadow mode to live tagging and routing
- [Overview](https://docs.duckie.ai/examples/bug-intake-to-engineering/overview.md): A blueprint for turning customer bug reports into engineering-ready issues
- [MVP](https://docs.duckie.ai/examples/bug-intake-to-engineering/mvp.md): The smallest useful bug intake system before full two-way support and engineering sync
- [System Map](https://docs.duckie.ai/examples/bug-intake-to-engineering/system-map.md): The two-way support-to-engineering architecture for bug intake
- [Intake and Repro Gathering](https://docs.duckie.ai/examples/bug-intake-to-engineering/intake-and-repro-gathering.md): Turn messy customer bug reports into structured repro-ready cases
- [Engineering Issue Workflow](https://docs.duckie.ai/examples/bug-intake-to-engineering/engineering-issue-workflow.md): Create and sync Linear or Jira issues from support bug reports
- [Engineering Collaboration](https://docs.duckie.ai/examples/bug-intake-to-engineering/engineering-collaboration.md): Route engineering questions through Slack and keep support in the loop
- [Customer Follow-Up Loop](https://docs.duckie.ai/examples/bug-intake-to-engineering/customer-follow-up-loop.md): Send engineering questions and resolutions back through the original customer conversation
- [Reporting and Feedback](https://docs.duckie.ai/examples/bug-intake-to-engineering/reporting-and-feedback.md): Report on bug intake health and improve the support-to-engineering loop
- [Rollout Plan](https://docs.duckie.ai/examples/bug-intake-to-engineering/rollout-plan.md): Move bug intake from assisted drafting to live two-way support and engineering sync
- [Overview](https://docs.duckie.ai/examples/wismo-order-status/overview.md): A blueprint for automating WISMO order-status operations with Duckie
- [MVP](https://docs.duckie.ai/examples/wismo-order-status/mvp.md): The smallest useful WISMO system before full exception handling, approvals, and proactive automation
- [System Map](https://docs.duckie.ai/examples/wismo-order-status/system-map.md): The workflow backbone, specialist agents, and proactive loops for WISMO operations
- [Intake and Order Context](https://docs.duckie.ai/examples/wismo-order-status/intake-and-order-context.md): Normalize customer WISMO requests and backend events into a consistent case context
- [Condition Routing and Agents](https://docs.duckie.ai/examples/wismo-order-status/condition-routing-and-agents.md): Route WISMO cases to the right specialist agent or controlled workflow
- [Proactive Follow-Up](https://docs.duckie.ai/examples/wismo-order-status/proactive-follow-up.md): Send WISMO updates when backend order, supplier, warehouse, or carrier status changes
- [Escalation and Resolution](https://docs.duckie.ai/examples/wismo-order-status/escalation-and-resolution.md): Route WISMO exceptions, approvals, and high-impact actions through a structured human loop
- [Rollout Plan](https://docs.duckie.ai/examples/wismo-order-status/rollout-plan.md): Move the WISMO blueprint from testing to live customer-facing operations
- [Overview](https://docs.duckie.ai/examples/disputes-and-chargebacks/overview.md): A blueprint for automating payment dispute operations with Duckie
- [MVP](https://docs.duckie.ai/examples/disputes-and-chargebacks/mvp.md): The smallest useful autonomous dispute operations system before full submission, escalation, reporting, and feedback loops
- [System Map](https://docs.duckie.ai/examples/disputes-and-chargebacks/system-map.md): The multi-channel, multi-agent architecture for dispute operations
- [Intake and Triage](https://docs.duckie.ai/examples/disputes-and-chargebacks/intake-and-triage.md): Deploy the entry point for new dispute and chargeback cases
- [Dispute Handling Workflow](https://docs.duckie.ai/examples/disputes-and-chargebacks/dispute-handling-workflow.md): Build the deterministic backbone for dispute case handling
- [Account Communication Agent](https://docs.duckie.ai/examples/disputes-and-chargebacks/account-communication-agent.md): Gather missing information from the account or merchant
- [Evidence and Approval](https://docs.duckie.ai/examples/disputes-and-chargebacks/evidence-and-approval.md): Draft evidence packets and route them for human review
- [Escalation Agent](https://docs.duckie.ai/examples/disputes-and-chargebacks/escalation-agent.md): Route human-in-the-loop decisions through an internal Slack channel
- [Daily Reporting Agent](https://docs.duckie.ai/examples/disputes-and-chargebacks/daily-reporting-agent.md): Post a scheduled stakeholder digest for dispute operations
- [Feedback and Analytics](https://docs.duckie.ai/examples/disputes-and-chargebacks/feedback-and-analytics.md): Use outcomes, run review, and stakeholder feedback to improve the system
- [Rollout Plan](https://docs.duckie.ai/examples/disputes-and-chargebacks/rollout-plan.md): Move the disputes blueprint from testing to live operations

## OpenAPI Specs

- [openapi](https://docs.duckie.ai/api-reference/openapi.json)

## Optional

- [Website](https://duckie.ai)
- [Blog](https://duckie.ai/blog)