agent-skills

Design, build, validate, and test a Docyrus automation (workflow) end-to-end using the `docyrus automation` CLI commands. Use when the user wants to automate a process — fire actions when records are created/modified/deleted, on a schedule, from a webhook/email/webform, or on a button click — and chain action nodes (send email/notification, create/update records, request approval/input, HTTP request, query a data source, generate a document, run an AI prompt/agent, execute a script, wait). Covers picking the right trigger, composing the action-node graph (conditions, field mappings, parent/child sequencing), and proving it runs. Triggers on "create an automation", "when a record is created/updated do X", "send an email/notification automatically", "schedule a recurring job", "webhook/email trigger", "approval workflow", "build a workflow", "automation trigger", "action node", `docyrus automation create`, `docyrus automation create-trigger`, `docyrus automation create-node`, or any Docyrus automation/workflow

Build Docyrus React TypeScript web applications end-to-end, combining authentication, API access, @docyrus/app-utils runtime helpers, @docyrus/devtools in-app debugging, generated collections, TanStack Query/Form patterns, and production-grade UI implementation with preferred component libraries. Use when creating or modifying Docyrus-backed apps that use @docyrus/api-client, @docyrus/signin, @docyrus/app-utils, @docyrus/devtools, Docyrus collections, queries, dashboards, forms, tables, layouts, or Docyrus UI components.

Use the Docyrus CLI (`docyrus`) to interact with the Docyrus platform from the terminal. Use when the user asks to authenticate, list apps, query or manage data records (`ds`), manage dev app data source schema objects (`studio`) including data sources, fields, enums, data views, forms, webforms, HTML/PDF export templates, and email templates, manage automations and their triggers/action nodes (`automation`), build or edit custom AI agents and their sub-resources (`agent`), set an app's AI agent context or manage app-scoped AI tools (`apps`), list tenant email accounts and send emails (`messaging`), manage tenant account settings — brands, the current user's profile, and tenant regional/formatting preferences (`account`), discover and call connectors (`connect`), send API requests, switch tenants/accounts, discover tenant OpenAPI specs, drive browser automation (`browser`), chat with platform agents (`docy`), launch the pi cowork/coding agents (`opsy`/`cody`/`coder`), run the agent bridge server (`server`), m

Create a new Docyrus app from the terminal with `docyrus apps create` and bootstrap a local dev workspace. Use when the user wants to scaffold a new app (web app / `external-app`, React techstack), create an app with an empty git repo, remix an existing template app, list remixable template apps, clone an existing app's repo locally, or generate the `.env` for local development against Docyrus. Triggers on "create an app", "new web app", "scaffold a Docyrus app", "start a new Docyrus frontend/project", "empty app repo", "remix a template app", "clone my app locally", "set up local development for an app", `docyrus apps create`, `docyrus apps clone`, `docyrus apps templates`, or any Docyrus app-creation + local-dev-bootstrap task via the CLI. For the full CLI command index see docyrus-cli-app; for building the app's code after cloning see docyrus-app-dev-react.

Architecture and building-blocks reference for the Docyrus platform — what the platform is made of (apps, data sources, fields, enumerations, custom queries, automations) and how the pieces fit together. Use when you need conceptual context about how Docyrus works, what its building blocks and capabilities are, or to decide which dedicated skill to use for a task. This is the conceptual map and router; hands-on how-to lives in the dedicated design / CLI / API skills it points to (data sources, automations, agents, AI tools, templates, webforms, ACL, querying, the CLI, and the REST API).

Create, read, update, delete, run, and consume saved Docyrus custom query records (tenant_custom_query) — reusable, parameterized SQL query templates whose body is DSQL — using the `docyrus dsql *-custom-query` CLI commands, and call them from a frontend via `@docyrus/api-client`'s `RestApiClient.runCustomQuery`. Use when the user wants to save a DSQL query as a reusable named report/template, parameterize it with runtime filters (Handlebars `{{filter}}` binding), test/run a saved query (`--debug` compiled SQL, `--simulate` EXPLAIN ANALYZE), manage the lifecycle of custom query records, or run a saved custom query from a React/frontend app. Triggers on "custom query", "saved query", "query template", "reusable report", "parameterized query", "run a saved query", "runCustomQuery", `docyrus dsql create-custom-query`, `docyrus dsql run-custom-query`, `list/get/update/delete-custom-query`, or wiring a saved query into an app. For authoring the DSQL body itself and ad-hoc one-off queries, see docyrus-dsql-query-de

Design, validate, and test a Docyrus email template end-to-end using the `docyrus studio` email-template CLI commands. Use when the user wants a reusable, data-bound email — a templated subject + HTML/text body with record placeholders ({{field}} Handlebars) that an automation send-email node (or a manual send) fills from a record. Covers writing the subject/body with the record context, optionally binding the template to a data source, and proving the template renders/sends. Triggers on "create an email template", "design a templated email", "email with record fields merged in", "notification email body", "studio create-email-template", or any email-template authoring + validation task in Docyrus. For HTML/PDF print/report templates use docyrus-print-pdf-template-design; for the automation that sends the email use docyrus-automation-design.

Design, validate, and test a Docyrus HTML/PDF print/export (report) template end-to-end using the `docyrus studio` html-template CLI commands. Use when the user wants a printable/exportable document rendered from a record — an invoice, quote, receipt, report, certificate, packing slip — authored as HTML + CSS with {{field}} Handlebars placeholders, with page setup (orientation, format, margins, header/footer, filename) and a data-source binding, then rendered to HTML/PDF for a record. Covers writing the body/header/footer/styles, the page options, marking a default template, and rendering it against a real record to prove it works. Triggers on "create a PDF/print template", "design an invoice/quote/report template", "printable document from a record", "export to PDF", "studio create-html-template", or any print/PDF report-template authoring + validation task. For emails use docyrus-email-template-design; for the automation that generates a document use docyrus-automation-design.

Design, configure, validate, and test Docyrus data views (saved grid/gallery views) with the `docyrus studio` data-view CLI commands. A data view is a named saved configuration of how one data source's records are presented — visible columns, order/pinning/grouping, row height, table/gallery display, sorting, filters (column filters + react-querybuilder query), quick-filter fields, row/cell color rules, paging, inline editing, and a bound form. Use when creating or editing a saved view / tab / list view, hiding or reordering columns, setting a default filtered/sorted view, grouping rows, adding color rules, configuring paging or inline editing, or binding a form to a view. Triggers on "create a data view", "saved view", "add a grid tab", "default view", "filter/sort a view", "hide columns", "group by", "color rules", "gallery view", `docyrus studio create-data-view`, `update-data-view`, `list-data-views`. For the schema use docyrus-data-source-design; for the React grid UI use docyrus-data-grid-page-design.

Plan, structure, and track implementation of a Docyrus-backed application using the repo-local project plan system. Use when starting a new Docyrus project, designing implementation phases, breaking features into tasks, drafting the data model and feature map as Mermaid diagrams, or managing progress across data sources, UI pages, dashboards, templates, ACL, automations, and AI tools. Triggers on "create a project plan", "design phases", "plan a Docyrus app", "set up the plan", "add a feature to the plan", "track tasks", "update task status", "draw the data model", "feature mindmap", or any request to structure work for a Docyrus-backed application.

End-to-end self-test a Docyrus-backed web app in a real browser — sign in headlessly, drive the UI, then read runtime problems back from @docyrus/devtools. Use after building or changing a Docyrus app feature to prove it works in the browser — authenticate without typing credentials via `docyrus auth sso-session`, navigate/click/fill with the docyrus browser tools, and fetch collected API errors, usage issues, and console errors via `docyrus browser devtools`. Triggers on "test the app in the browser", "e2e test this flow", "sign me into the app", "self-test the feature I built", "verify it works end to end", "check for runtime errors", "what's broken on the page", "is @docyrus/devtools installed".

Discover and use Docyrus integration connectors from the terminal with the `docyrus connect` CLI commands — call third-party/external APIs (Microsoft Graph/Outlook, Twilio, Meta, etc.) through a connector's managed provider auth. Use when the user wants to find available connectors, inspect a connector's actions and input/output schemas, list a tenant's connections for a provider, run a connector action by provider slug + action key (e.g. send an SMS, send an email, fetch from an external API), or make a raw authenticated HTTP request through a connector's provider auth. Triggers on "list connectors", "what integrations are available", "run a connector action", "call the X API through Docyrus", "send SMS/email via a connector", "connect curl", "provider auth", `docyrus connect`, `docyrus connect run-action`, `docyrus connect curl`, or any connector/integration usage task. NOTE — connections (credentials) are created via the Docyrus UI / OAuth flow, not this CLI — this skill discovers and *uses* connectors. Fo

Use Docyrus CLI browser commands for browser automation (local Chrome or remote Cloudflare Browser Rendering). Use when you need to start a browser session, navigate pages, inspect page state, interact with elements, capture screenshots, evaluate JavaScript, read console output, inspect network requests, or run CDP scripts.

Read and update a Docyrus user's own profile and the active tenant's (workspace's) regional & formatting preferences using the `docyrus account user-profile` and `docyrus account tenant-preferences` CLI commands. Use when the user wants to view or change their personal profile (first/last name, mobile, date of birth, gender, time zone, language) or the workspace-wide regional settings (default language, locale, time zone, date/time format strings, first day of week, fiscal year, business/working hours, base currency, currency position, decimal/thousand separators, decimal precision). Triggers on "update my profile", "change my name/phone/timezone/language", "set my date of birth", "view my account info", "change the workspace language/locale/timezone", "set the date or time format", "set business/working hours", "change base currency", "decimal/thousand separator", "first day of the week", "fiscal year", "regional settings", "tenant preferences", `docyrus account user-profile`, `docyrus account tenant-prefere

Create, read, update, and delete a tenant's brand identity records (`tenant_brand`) using the `docyrus account brands` CLI commands. Use when the user wants to manage a Docyrus brand — set or change the brand's colors, typography, spacing, logos/favicons, voice & messaging guidelines, slide/presentation defaults, chart styling, illustration rules, or content constraints; mark a default brand; archive a brand; or import branding from a website. Triggers on "create a brand", "set brand colors", "update the brand voice/typography", "add a logo to the brand", "make this the default brand", "archive a brand", "import brand from website", "brand guidelines", `docyrus account brands`, `account brands create`, `account brands update`, `tenant_brand`, or any tenant brand identity / brand-kit management task in Docyrus. For the full CLI command index see docyrus-cli-app; for platform concepts see docyrus-platform.

Design, configure, and validate a Docyrus tenant ACL (Access Control Layer) end-to-end using the `docyrus acl` CLI commands. Use when the user wants to define permission roles, assign operation-level permissions to roles for data sources / apps / AI tools / settings / devtools, set up row-level record filters (role queries), build an organizational hierarchy (hierarchy units), or look up available system operations. Triggers on "create a role", "set permissions for a role", "grant access to a data source", "role query", "row-level filter", "hierarchy unit", "org-chart access", "acl rules", "what operations", `docyrus acl roles`, `docyrus acl rules`, `docyrus acl role-queries`, `docyrus acl hierarchy-units`, `docyrus acl operations`, or any ACL role/permission design + configuration task in Docyrus.

Design, build, validate, and test a custom Docyrus AI agent end-to-end using the `docyrus agent` CLI commands and its sub-resources. Use when the user wants to create or configure a custom AI agent (assistant/skill/workflow agent) in a dev app — set its instructions, pick its models, toggle capabilities (tools, data sources, files, knowledge base, web search), attach data sources, attach tools, add knowledge docs, wire MCP servers, connect agents (forward/subagent), add dynamic contexts, build workflow steps, or create deployments. Covers choosing the agent shape, the required vs optional config, the sub-resource attach order, and proving the agent is well-formed. Triggers on "create an AI agent", "build an assistant", "configure an agent", "give the agent a data source / tool / knowledge", "agent workflow steps", "agent deployment", "connect agents", `docyrus agent create`, `docyrus agent tools`, `docyrus agent data-sources`, `docyrus agent workflow-steps`, or any custom-agent design + validation task. For a

Develop applications using the Docyrus API with @docyrus/api-client and @docyrus/signin libraries. Use when building apps that authenticate with Docyrus OAuth2 (PKCE, iframe, client credentials, device code), make REST API calls to Docyrus data source endpoints, construct query payloads with filters, aggregations, formulas, pivots, and child queries, or integrate with external connectors (discover connectors, send requests through provider auth, run actions). Triggers on tasks involving Docyrus API integration, @docyrus/api-client usage, @docyrus/signin authentication, data source query building, Docyrus REST endpoint consumption, connector discovery, or external provider requests.

Design, validate, and test a Docyrus data source (schema/table/collection) end-to-end using the `docyrus studio` CLI commands. Use when the user wants to model a new entity or table, create or change a data source and its fields, choose the right field types, set up relations between data sources, add select/status/enum option lists, or apply data-access rules — and then confirm the schema is correct and prove it works with sample records. Triggers on "design a data source", "create a table/collection/entity", "add fields", "model a schema", "set up a relation", "add dropdown/status options", "studio create-data-source", "studio create-field", "studio create-enums", or any data-source schema modelling + validation + testing task in Docyrus.

Write, discover, and run DSQL (Docyrus Structured Query Language) queries against logical Docyrus data sources. Use when the user wants to query, report on, or aggregate data from Docyrus — list records, count tasks by status, join contacts with users, build time-series breakdowns, answer "show me all projects with more than 5 tasks", or any read-only data question. Also covers using the AI-powered `docyrus dsql ask` command (natural language → DSQL → run → results). Triggers on "query data", "show me all X", "count by Y", "join Z with W", "DSQL query", "docyrus dsql", "run a report", "SQL over data sources", "list records from", "aggregate", "how many", "docyrus dsql ask", "docyrus dsql query", "docyrus dsql schema", or any data retrieval / reporting task against Docyrus logical data sources.

Design, validate, and test a Docyrus webform — a public, unauthenticated form that collects external submissions into records of a data source — using the `docyrus studio` webform CLI commands. Use when the user wants a public intake/lead/contact/registration/survey form whose submissions create records (e.g. a website contact form that lands in a CRM data source), needs the form's public URL or embed snippet, wants to bind a form to a data source, or wants a submission to kick off an automation. Covers designing the form schema, binding it to a data source, activating it, getting the public/submit/embed URLs, and proving a submission creates a record. Triggers on "create a webform", "public form", "contact/lead/intake form", "form that creates records", "embed a form on a website", "form submission webhook", "studio create-webform", or any webform authoring + validation task. For the data source the form writes into, use docyrus-data-source-design; for an automation that fires on submission, use docyrus-auto

Add and manage app-scoped AI tools for the Docyrus system base agent ("Docy") using the `docyrus apps ai-tools` CLI commands, and set app-level agent guidance with `docyrus apps set-agent-context`. App-scoped tools are owned by the app and, once the app is installed in a tenant, are automatically available to Docy (the built-in system base assistant) — this is how you give Docy new tools; there is no per-agent wiring and they are not for custom agents. Use when building custom tools the base agent can call — covering all four executable tool types — `data_source_query` (read one data source with author-fixed shaping and parameter-bound filters), `custom_query` (hand-written read-only SELECT SQL with Handlebars templating), `secure_exec` (sandboxed JavaScript that calls the Docyrus REST API), and `client_side` (tool executed in the user's browser/app). Triggers on tasks like "add a tool to Docy / the base agent", "create an AI tool for an app", "add a data source query tool", "build a custom query tool", "writ

Run after making Docyrus API changes to catch bugs, performance issues, and code quality problems. Use when implementing or modifying code that uses Docyrus collection hooks (.list, .get, .create, .update, .delete), direct RestApiClient calls, query payloads with filters/calculations/formulas/childQueries/pivots, or TanStack Query integration with Docyrus data sources. Triggers on tasks involving Docyrus API logic, data fetching, mutations, or query payload construction.