Sharepoint
Skill ComeOnOliver/skillshub/skills/TerminalSkills/skills/sharepoint
π§ The right skill, one API call. AI agent skills registry with token-efficient skill resolution. 5,000+ skills from 500+ top repos.From the repository description
npx -y skills add ComeOnOliver/skillshub --skill sharepointAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
7.5 KB, ~1.9k tokens by cl100k_base, as published. Nobody here has run it
SharePoint
Overview
This skill helps AI agents integrate with SharePoint Online for document management, list operations, site automation, and custom development. It covers Microsoft Graph API for sites/lists/files, SharePoint REST API for advanced operations, search, permissions management, and SharePoint Framework (SPFx) for custom web parts.
Instructions
Step 1: Authentication
import { ClientSecretCredential } from '@azure/identity';
import { Client } from '@microsoft/microsoft-graph-client';
import { TokenCredentialAuthenticationProvider } from '@microsoft/microsoft-graph-client/authProviders/azureTokenCredentials';
// Permissions: Sites.ReadWrite.All, Files.ReadWrite.All, Lists.ReadWrite.All
const credential = new ClientSecretCredential(
process.env.AZURE_TENANT_ID, process.env.AZURE_CLIENT_ID, process.env.AZURE_CLIENT_SECRET
);
const authProvider = new TokenCredentialAuthenticationProvider(credential, {
scopes: ['https://graph.microsoft.com/.default'],
});
const graphClient = Client.initWithMiddleware({ authProvider });
Step 2: Sites & Document Libraries
// Get site by URL
const site = await graphClient.api('/sites/contoso.sharepoint.com:/sites/engineering').get();
// Get document libraries (drives)
const drives = await graphClient.api(`/sites/${site.id}/drives`).get();
const driveId = drives.value[0].id;
// List folder contents
const items = await graphClient.api(`/drives/${driveId}/root/children`)
.select('id,name,size,lastModifiedDateTime,webUrl').get();
// Create folder
await graphClient.api(`/drives/${driveId}/root/children`).post({
name: 'New Project', folder: {}, '@microsoft.graph.conflictBehavior': 'rename',
});
Step 3: File Operations
// Upload small file (< 4MB)
await graphClient.api(`/drives/${driveId}/root:/Reports/monthly-report.pdf:/content`).putStream(fileContent);
// Upload large file (> 4MB) β resumable upload session
const session = await graphClient
.api(`/drives/${driveId}/root:/LargeFiles/dataset.csv:/createUploadSession`)
.post({ item: { '@microsoft.graph.conflictBehavior': 'replace', name: 'dataset.csv' } });
// Upload in 10MB chunks with Content-Range headers to session.uploadUrl
// Download, copy, move, delete
await graphClient.api(`/drives/${driveId}/items/${itemId}/content`).getStream();
await graphClient.api(`/drives/${driveId}/items/${itemId}/copy`).post({ parentReference: { driveId, path: '/root:/Archive' }, name: 'backup.pdf' });
await graphClient.api(`/drives/${driveId}/items/${itemId}`).patch({ parentReference: { id: targetFolderId } });
await graphClient.api(`/drives/${driveId}/items/${itemId}`).delete();
// Create sharing link
await graphClient.api(`/drives/${driveId}/items/${itemId}/createLink`).post({
type: 'view', scope: 'organization',
});
Step 4: SharePoint Lists
// Create a list
await graphClient.api(`/sites/${siteId}/lists`).post({
displayName: 'Project Tracker',
columns: [
{ name: 'ProjectName', text: {} },
{ name: 'Status', choice: { choices: ['Not Started', 'In Progress', 'Complete'] } },
{ name: 'DueDate', dateTime: {} },
{ name: 'Budget', currency: {} },
],
list: { template: 'genericList' },
});
// Get items with filter
const items = await graphClient.api(`/sites/${siteId}/lists/${listId}/items`)
.expand('fields($select=ProjectName,Status,DueDate)')
.filter("fields/Status eq 'In Progress'")
.orderby('fields/DueDate').get();
// Create / Update / Delete items
await graphClient.api(`/sites/${siteId}/lists/${listId}/items`).post({
fields: { ProjectName: 'Website Redesign', Status: 'In Progress', DueDate: '2026-04-01', Budget: 25000 },
});
await graphClient.api(`/sites/${siteId}/lists/${listId}/items/${itemId}/fields`).patch({ Status: 'Complete' });
await graphClient.api(`/sites/${siteId}/lists/${listId}/items/${itemId}`).delete();
// Bulk create (batch request)
const batchBody = {
requests: items.map((item, i) => ({
id: `${i}`, method: 'POST', url: `/sites/${siteId}/lists/${listId}/items`,
body: { fields: item }, headers: { 'Content-Type': 'application/json' },
})),
};
await graphClient.api('/$batch').post(batchBody);
Step 5: Search
// Search across all SharePoint content
const results = await graphClient.api('/search/query').post({
requests: [{
entityTypes: ['driveItem', 'listItem', 'site'],
query: { queryString: 'quarterly report 2026' },
from: 0, size: 25,
}],
});
// Search with filters (file type, author, site)
await graphClient.api('/search/query').post({
requests: [{ entityTypes: ['driveItem'], query: { queryString: 'filetype:pdf AND author:"John Smith"' } }],
});
Step 6: Webhooks
// Subscribe to document library changes
const subscription = await graphClient.api('/subscriptions').post({
changeType: 'created,updated,deleted',
notificationUrl: 'https://your-app.com/api/sharepoint-webhook',
resource: `/drives/${driveId}/root`,
expirationDateTime: new Date(Date.now() + 3 * 24 * 60 * 60 * 1000).toISOString(),
clientState: 'your-secret-token',
});
// Renew before expiry (max 3 days for SharePoint)
await graphClient.api(`/subscriptions/${subscription.id}`).patch({
expirationDateTime: new Date(Date.now() + 3 * 24 * 60 * 60 * 1000).toISOString(),
});
Examples
Example 1: Automate monthly report uploads to a SharePoint document library
User prompt: "Write a script that uploads a generated PDF report to our SharePoint Engineering site's Reports folder each month, creating the folder if it doesn't exist."
The agent will:
- Authenticate with Microsoft Graph using the Azure AD app credentials
- Resolve the site ID via
/sites/contoso.sharepoint.com:/sites/engineering - Get the default drive ID from
/sites/{siteId}/drives - Check if the monthly folder exists under Reports; if not, create it with
POST /drives/{driveId}/root/children - Upload the PDF using the small-file API (
PUT /drives/{driveId}/root:/Reports/2026-02/monthly-report.pdf:/content) or a resumable upload session if the file exceeds 4MB
Example 2: Build a project tracker with SharePoint lists
User prompt: "Create a SharePoint list called 'Project Tracker' with columns for project name, status, due date, and budget. Then add three sample projects and filter for in-progress items."
The agent will:
- Create the list via
POST /sites/{siteId}/listswith text, choice, dateTime, and currency columns - Add three items using
POST /sites/{siteId}/lists/{listId}/itemswith realistic data (e.g., "Website Redesign" at $25,000 In Progress) - Query filtered results with
.filter("fields/Status eq 'In Progress'")and.orderby('fields/DueDate') - Return the filtered items showing only in-progress projects sorted by due date
Guidelines
- Use Graph API over SharePoint REST API when possible β cleaner, better documented, same auth as Teams
- Batch requests for bulk operations (
/$batch) β max 20 requests per batch - Large file uploads (> 4MB) must use upload sessions β direct PUT has size limits
- SharePoint webhooks expire after 3 days max β set up a renewal job
- Use
$selectand$expandto reduce payload size β don't fetch everything - Delta queries (
/drives/{id}/root/delta) for syncing β tracks changes since last check - Document libraries are just "drives" in Graph API β same endpoints as OneDrive
- Column names in list APIs are internal names (no spaces) β use
fieldsobject for values - Rate limits: 10,000 API calls per minute per tenant β use batching for high-volume operations
- Search API indexes content inside documents (Word, PDF, etc.) β not just filenames
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.