resend-build
Use when building, editing, or sending anything in Resend, including broadcasts, templates, segments, and contacts, or when checking what a past send did.
- License 许可
- Apache-2.0
- 所属仓库
- vercel/eve
- 最近更新
- 2026-11-15
- Stars
- ★ 5.9k
- Forks
- 656
01 前言 Front Matter
- description
- Use when building, editing, or sending anything in Resend, including broadcasts, templates, segments, and contacts, or when checking what a past send did.
03 SKILL.md 正文
Building in Resend
The tools are discoverable but the order matters, and several of them fail in ways that are quiet rather than loud. Read this before the first call, not after a call goes wrong.
Resend's tools are not preloaded. Find them with eve__search (the query resend__ lists only Resend's), then call one with eve__tool by its qualified name (resend__<tool>). Don't guess at a tool name or its arguments.
The build order
For a broadcast, which is one email to a segment:
list-domainsand pick a verified one for the from address.list-segmentsand pick the target. A broadcast requires a segment: there is no send-to-everyone option. If nothing covers the intended audience, say so and agree a segment with the user before creating one.create-broadcastwith the name, segment, subject, and the plain text version. This does not send.get-tiptap-json-contentwithinclude_schema: true, thencompose-broadcastto set the body.- Show the user what you built: subject, preview text, from, segment and its size, and the send time.
send-broadcast, which pauses for approval.
For a template, which is a reusable email with variables:
create-template. It is created as a draft.get-tiptap-json-contentwithinclude_schema: true, thencompose-template.publish-template. A template cannot be used for sending until it is published, and re-publishing is how an edit goes live.
The traps
Always read before you compose. Call get-tiptap-json-content before compose-broadcast or compose-template, even when you are certain the body is empty. Someone may have edited it in the dashboard, and composing without reading overwrites their work.
Two content modes, and switching loses work. Content set with compose-* is editable in the Resend visual editor; content set with update-broadcast's html and text fields is not. Moving between the two modes is lossy. Prefer compose-* so the user can refine it themselves, and if raw HTML is genuinely wanted, ask before switching an existing body over.
update-broadcast needs fields you have to fetch. The API requires from and segmentId to be set, and a broadcast created in the dashboard may have neither. Call get-broadcast first and include them in the update if they are missing, or the call fails on something unrelated to what you were changing.
Personalization needs fallbacks. Placeholders use triple braces: {{{FIRST_NAME}}}, {{{LAST_NAME}}}, {{{EMAIL}}}, and {{{RESEND_UNSUBSCRIBE_URL}}}. FIRST_NAME and LAST_NAME take a fallback with a pipe, as in {{{FIRST_NAME|there}}}. Use it every time. Without one, contacts with no first name get "Hi ,".
Include the unsubscribe URL. {{{RESEND_UNSUBSCRIBE_URL}}} is how a visible unsubscribe link gets into the body, and marketing mail needs one. See the deliverability skill for why this is a hard requirement rather than a courtesy.
Ids come in several forms. Most tools accept a UUID, and many also accept a template alias or a Resend dashboard URL. Pass whatever the user gave you rather than trying to convert it.
Sending and the approval gate
Sending a broadcast, an email, or a batch pauses for the user's approval, and so do deletes and changes to a contact's topic subscriptions. This is deliberate. Do not try to route around a gate by using a different tool to achieve the same effect.
- Say what the approval covers before you ask: the segment, the number of contacts, and the timing. That sentence is what the user is agreeing to.
- Never test by sending to a real segment. Send a one-off with
send-emailto an address the user names, or have them preview it in the dashboard. send-broadcasttakes an optionalscheduledAt. A scheduled send can be cancelled and an immediate one cannot, so when the user is hesitating, schedule it and tell them when the decision closes.cancel-emailcancels a scheduled one-off. There is a window, not an unlimited one.- If approval is denied, stop. Ask what to change rather than reissuing the call.
Reading what happened
get-email and list-emails carry delivery outcomes for one-off sends, get-broadcast carries a broadcast's state, and list-logs and get-log carry the API request and response when something failed at the call level rather than at the delivery level.
Report the numbers you read, over the volume you read them across. Do not convert a delivery count into a claim about inbox placement: deliverability covers what those numbers do and do not prove.
04 同仓库更多 Skills
writing-quality
The review rubric: writing-quality rules and reference lists for judging a content draft, covering AI-sounding phrasing, bloated wording, structure, and web-content specs. Load at the start of every review.
writing-quality
Writing-quality guardrails for any prose the agent drafts or edits: blog posts, page copy, Notion drafts, release notes, marketing text. Use this skill whenever writing or revising content meant for humans to read, to keep the prose natural, plain, and free of AI-sounding phrasing. Not needed for code, queries, or structural CMS work.
seo-aeo-best-practices
SEO and AEO best practices for metadata, Open Graph, sitemaps, robots.txt, hreflang, JSON-LD structured data, EEAT, and content optimized for search engines and AI answer surfaces. Use this skill when implementing page SEO, technical SEO, schema markup, international SEO, AI-overview readiness, or improving content for Google, ChatGPT, Perplexity, and similar assistants.
sanity-best-practices
Sanity development best practices for schema design, GROQ queries, TypeGen, Visual Editing, images, Portable Text, Studio structure, localization, migrations, Sanity Functions, Blueprints, and framework integrations such as Next.js, Nuxt, Astro, Remix, SvelteKit, Angular, Hydrogen, and the App SDK. Use this skill whenever working with Sanity schemas, defineType or defineField, GROQ or defineQuery, content modeling, Presentation or preview setups, Sanity-powered frontend integrations, Sanity Functions, documentEventHandler, defineDocumentFunction, defineMediaLibraryAssetFunction, @sanity/functions, @sanity/blueprints, sanity.blueprint.ts, event-driven content automation, or when reviewing and fixing a Sanity codebase.
portable-text-serialization
Render and serialize Portable Text to React, Svelte, Vue, Astro, HTML, Markdown, and plain text. Use when implementing Portable Text rendering in any frontend framework, building custom serializers for non-standard block types, converting Portable Text to HTML strings server-side, converting Portable Text to Markdown, extracting plain text from Portable Text, or troubleshooting rendering issues with marks, blocks, lists, or custom types.
portable-text-conversion
Convert HTML and Markdown content into Portable Text blocks for Sanity. Use when migrating content from legacy CMSs, importing HTML or Markdown into Sanity, building content pipelines that ingest external content, converting rich text between formats, or programmatically creating Portable Text documents. Covers @portabletext/markdown (markdownToPortableText), @portabletext/block-tools (htmlToBlocks), custom deserializers, and the Portable Text specification for manual block construction.