Add Sharepoint skill

📋 Shared Instructions: shared-instructions.md - Cross-cutting concerns.

References:

• sharepoint-reference.md - Column encoding, choice fields, lookups, API patterns (CRITICAL)
• api-authentication-reference.md - Graph API auth, token, site ID
• list-management-reference.md - Query, create, extend lists and columns

Add SharePoint

Two paths: existing lists (skip to Step 6) or new lists (full workflow).

Workflow

1. Check Memory Bank → 2. Plan → 3. Setup Graph API Auth → 4. Review Existing Lists → 5. Create Lists → 6. Get Connection ID → 7. Discover Sites → 8. Discover Tables → 9. Add Connector (tables and/or actions) → 10. Configure → 11. Build → 12. Update Memory Bank

---

Step 1: Check Memory Bank

Check for memory-bank.md per shared-instructions.md.

Step 2: Plan

Ask the user:

1. Which SharePoint list(s) do they need?
2. Do the lists already exist on their site, or do they need to create new ones?

If lists already exist: Skip to Step 6.

If creating new lists:

• Ask about the data they need and design an appropriate schema
• Reuse existing lists when possible (don't duplicate)
• Enter plan mode with EnterPlanMode, present the list designs with columns and types
• Get approval with ExitPlanMode

Step 3: Setup Graph API Auth (if creating lists)

See api-authentication-reference.md for full details.

az account show   # Verify Azure CLI logged in

$api = Initialize-SharePointGraphApi -SiteUrl "https://<tenant>.sharepoint.com/sites/<site-name>"
$headers = $api.Headers
$siteId = $api.SiteId

Requires Sites.Manage.All permission.

Step 4: Review Existing Lists (if creating lists)

Always query existing lists first before creating:

$existingLists = Invoke-RestMethod -Uri "https://graph.microsoft.com/v1.0/sites/$siteId/lists?`$select=id,displayName,description,list&`$filter=list/hidden eq false" -Headers $headers

See list-management-reference.md for Find-SimilarLists, Compare-ListSchemas, and Get-ListSchema functions.

Present findings to user with AskUserQuestion:

• Lists that can be reused (already exist with matching columns)
• Lists that need extension (exist but missing columns)
• Lists that must be created (no match found)

Step 5: Create Lists (if creating lists)

Get explicit confirmation before creating. Use safe functions from list-management-reference.md:

• New-SharePointListIfNotExists
• Add-SharePointColumnIfNotExists
• Add-SharePointLookupColumn (for cross-list references)

Step 6: Get Connection ID

Discovering sites and lists (Steps 7–8) needs a SharePoint connection ID on every CLI version: pa connection list-datasets and list-tables don't find or create connections the way pa app add data-source does on CLI 1.3.0+ (see connector-reference.md).

1. Run the /list-connections skill, filtered with --search sharepointonline.
2. Exactly one SharePoint connection: use its connection ID. Several: ask the user which one to use.
3. None: create one after asking the user.

   • CLI 1.3.0+: SharePoint supports single sign-on, so with the user's OK run:

     pa connection create --connector shared_sharepointonline --use-sso
     

     It creates the connection silently with the user's account and prints Connection ID: <id>.

   • Older CLIs, or if the user wants a different sign-in type: follow Create a connection with the user, then repeat step 1.

Step 7: Discover Sites

List available SharePoint sites the user has access to:

pa connection list-datasets --connector sharepointonline -c <connection-id>

Present the sites to the user and ask which one(s) they want to connect to. If the user already specified a site URL, confirm it appears in the list.

If pa connection list-datasets fails or returns no results:

• Auth error: Run pa auth login (or power-apps login on power-apps-only projects), then retry.
• Empty list: Confirm the connection ID is for a SharePoint Online connection and the user has access to at least one site. STOP if the list is empty after confirming.
• Any other non-zero exit: Report the exact error output. STOP.

Step 8: Discover Tables

For each selected site, list the available lists and document libraries:

pa connection list-tables --connector sharepointonline -c <connection-id> -d '<site-url>'

If pa connection list-tables fails or returns no results:

• Confirm the site URL from Step 7 is exact (copy from the output — do not retype).
• If still empty, the user may not have access to that site's lists. Ask them to verify permissions in SharePoint.
• Any other non-zero exit: Report the exact error output. STOP.

Present the tables to the user and ask which ones they want to add. Suggest tables that look relevant to their use case (based on memory bank context or the user's stated requirements). If lists were created in Step 5, they should appear here.

Step 9: Add Connector

SharePoint supports both tables and actions, so say which one you're adding. Pass the connection ID from Step 6, so the add doesn't need to look for a connection (on CLI 1.3.0+, also pass --non-interactive):

# Lists and libraries as tables: -c (connection ID), -d (site URL), and --table (list name)
pa app add data-source --connector sharepointonline -c <connection-id> -d '<site-url>' --table '<table-name>'

# Document library and file operations (CLI 1.2.0+): add the connector's actions
pa app add data-source --connector sharepointonline -c <connection-id> --as action

Run the table command for each list or library the user selected. The -d (dataset) is the SharePoint site URL from Step 7, --table is the list/library display name from Step 8. Add the actions only if the app needs file or folder operations; the CLI keeps them as a separate data source next to the tables.

Step 10: Configure

Read sharepoint-reference.md before writing any SharePoint code -- column encoding, choice fields, and lookups have critical gotchas.

What the CLI generates depends on how you added SharePoint in Step 9 (CLI 1.3.0):

Added with	Generated service	How you call it
--table '<list>'	One service per list, named after it (for example ProjectMilestonesService), with <List>Read / <List>Write models	getAll(options?), get(id), create(record), update(id, changedFields), delete(id)
--as action	SharePointService	The connector's operations, with positional parameters: GetItems(dataset, table, ...), PostItem(dataset, table, item), ListRootFolder(dataset), ListFolder(dataset, id), GetFileContent(dataset, id), and more

These examples were verified against CLI 1.3.0. Grep src/generated/services/ to confirm the service name and each method's exact signature before using it, especially in projects created with an older CLI.

Common operations:

// Lists added as tables: one typed service per list
const result = await ProjectMilestonesService.getAll({ top: 50, orderBy: ["Title asc"] });
const milestones = result.data ?? [];
await ProjectMilestonesService.create({ Title: "New milestone" });
await ProjectMilestonesService.update(id, { Title: "Renamed milestone" });

// Connector actions (added with --as action) take positional parameters, for example
// PostItem(dataset, table, item): dataset is the full site URL, table is the list's display name.
const dataset = "https://contoso.sharepoint.com/sites/your-site";
const items = await SharePointService.GetItems(dataset, "Your List Name");
await SharePointService.PostItem(dataset, "Your List Name", { Title: "New Item", Status: "Active" });

// Files in document libraries
const root = await SharePointService.ListRootFolder(dataset);
const files = await SharePointService.ListFolder(dataset, "<folder-id>");
const content = await SharePointService.GetFileContent(dataset, "<file-id>");

Key points:

• dataset (action calls) is always the full SharePoint site URL
• table (action calls) and --table (Step 9) take the list's display name
• List column names in the API may differ from display names (spaces become _x0020_, special chars encoded)
• Document library operations use folder/file IDs or server-relative URLs
• Choice columns use string values, not integer picklist codes (unlike Dataverse)

Use Grep to find specific methods in the generated services (generated files can be very large -- see connector-reference.md).

Step 11: Build

npm run build

Fix TypeScript errors before proceeding. Do NOT deploy yet.

Step 12: Update Memory Bank

Update memory-bank.md with: connector added, site URL, lists/libraries configured (or created), build status. Don't record the connection ID.