sandbox-notion
Call the Notion API from a Code node or CodeAct action, with NodeTool's request builder and rich-text/block readers running on the host
- License 许可
- AGPL-3.0
- 最近更新
- 2026-11-15
- Stars
- ★ 635
- Forks
- 91
01 前言 Front Matter
- name
- sandbox-notion
- description
- Call the Notion API from a Code node or CodeAct action, with NodeTool's request builder and rich-text/block readers running on the host
02 资源文件 1
- 参考文档package.json871 B
03 SKILL.md 正文
Notion in the sandbox
Specifier: @nodetool-ai/sandbox-notion. Import it at the top of the body.
Notion is an ordinary HTTPS API and fetch reaches it. What costs a script its
first three attempts is everything around the call — the Notion-Version
header, the rich-text arrays every piece of text is wrapped in, and the block
tree a page's content comes back as. Those three are this pack.
Nothing here sends a request. request builds one; the guest's own fetch
sends it, under the run's fetch cap and SSRF guard.
request — an authenticated Notion request
import { request } from "@nodetool-ai/sandbox-notion";
const token = await nodetool.secrets.get("NOTION_API_KEY");
const req = await request({
token,
path: "search",
method: "POST",
body: { query: inputs.query, page_size: 10 }
});
const res = await fetch(req.url, req);
if (!res.ok) throw new Error(`Notion search failed: ${res.status}`);
return { results: res.json.results };
Options: token (required), path (required — relative to /v1 unless it
starts with /), method (default GET), body (JSON-encoded when present),
query (query-string bag), version (default 2022-06-28).
Returns {url, method, headers, body?}, which fetch(req.url, req) takes
whole.
plainText — a rich-text array as a string
import { plainText } from "@nodetool-ai/sandbox-notion";
const title = await plainText(page.properties.Name.title);
Takes the array Notion puts under every title, rich_text and heading.
Anything else gives "".
toMarkdown — a page's blocks as markdown
import { request, toMarkdown } from "@nodetool-ai/sandbox-notion";
const req = await request({ token, path: `blocks/${inputs.page_id}/children`, query: { page_size: 100 } });
const blocks = (await fetch(req.url, req)).json.results;
return { markdown: await toMarkdown(blocks) };
Handles paragraphs, the three heading levels, bulleted and numbered items, to-dos, quotes, code fences and dividers. A block markdown has no form for — a database, an embed, a synced block — contributes nothing rather than a placeholder, so the result is text a model or a file can take as-is.
Common flows
Create a page in a database
const req = await request({
token, path: "pages", method: "POST",
body: {
parent: { database_id: inputs.database_id },
properties: { Name: { title: [{ text: { content: inputs.title } }] } }
}
});
const page = (await fetch(req.url, req)).json;
Query a database
const req = await request({
token, path: `databases/${inputs.database_id}/query`, method: "POST",
body: { filter: { property: "Status", select: { equals: "Open" } }, page_size: 50 }
});
Gotchas
- Every export is async. A host call is a round trip.
- Nested blocks are not fetched.
toMarkdownis pure over the blocks you hand it. A block withhas_childrenneeds its ownblocks/{id}/childrencall. - Notion paginates at 100. Follow
next_cursorwithquery: { start_cursor }and watch the run's fetch cap. - The integration needs access. A page shared with nobody returns 404 even with a valid token — share it with the integration in Notion first.
04 同仓库更多 Skills
humanizer
Rewrite AI-sounding text so it reads like the writer without changing what it says. Use when editing or reviewing prose for AI tells: not-X-but-Y contrasts, one-line closers, staged openers, forced triads, dashes everywhere, inflated claims, sales language, stock AI words, bold labels, or filler.
nodetool
Use the NodeTool MCP server to build, validate, run, and debug visual AI workflows, and to work with assets, collections, and jobs. Use when the user mentions NodeTool or asks for a workflow graph, a node, or a media generation run on their NodeTool server.
wan-2-6-prompting
Prompt Alibaba's Wan 2.6 video — text-to-video with a global style line plus timing-bracketed shots, and image-to-video that describes only what changes over time. Covers the shot-mode switch per provider, the prompt budget, negative prompts, motion intensity wording, audio input and prompt expansion. Use whenever the model id names a Wan 2.6 video route (alibaba/wan-2.6/text-to-video, alibaba/wan-2.6/image-to-video, wan-video/wan-2.6-t2v, wan-video/wan-2.6-i2v, wan-video/wan2.6-i2v-flash) on generate_video, animate_image or a TextToVideo node. Not for Wan 2.5 or 2.7, or the Wan 2.6 image routes.
video-workflow
Build a reusable NodeTool video workflow for batches or repeated inputs. For a single finished piece, use a storyboard skill.
video-clone
Analyze a reference video and recreate its shot structure with a new product or cast in NodeTool.
video-audio-continuity
Keep sound continuous across a multi-scene piece cut from generated video — why one clip per scene hard-cuts the audio at every boundary, when to write all the scenes into a single generation instead, the sound-design vocabulary that survives a provider's copyright filter, and how to check the length, the aspect ratio and the mix of what came back. Use when a piece has more than one scene and the video model writes its own audio (Seedance 2, Veo 3, MiniMax H3, Kling 2.6 and later), or when a cut's sound drops out at the shot changes.