Plugin check: Run node "${PLUGIN_ROOT}/scripts/check-version.js" — if it outputs a message, show it to the user before proceeding.
Route to the correct deployment workflow before running SPA build, attachment, or activation
logic.
Initial request: $ARGUMENTS
Classify explicit request signals first:
powerpages.config.json → SPA site.portalconfig,or website.yml → Non-SPA site
When the request is not explicit, search the current workspace for:
powerpages.config.json;.portalconfig/ and website.yml, including a nested.powerpages-site/.
Exclude dependency, build-output, temporary, generated-output, and version-control directories.
Use request paths and the current directory to resolve one project. Never choose the first result
arbitrarily.
🚦 Gate (plan · deploy-site:0.project-type): Select the exact project when SPA and
non-SPA sites, or multiple valid roots, remain possible.
Trigger: Project type or root is ambiguous after discovery.
Why: The two site types require incompatible PAC upload commands.
Cancel leaves: Nothing; no build or upload has started.
If ambiguous, use AskUserQuestion to select the exact project root and detected type.
${PLUGIN_ROOT}/skills/deploy-site/workflows/non-spa-site.md. Do not continue into the
SPA phases below.
The authoritative markers and prompts live in workflows/non-spa-site.md.
Gate deploy-site:non-spa-4.permissions-audit handles permission-sensitive changes before upload.
Gate deploy-site:non-spa-5.upload approves the exact root, environment, website identity, model, and local change summary immediately before upload.
The following phases retain the existing SPA behavior.
Goal: Ensure PAC CLI is installed and available on the system PATH
Actions:
pac help to check if the PAC CLI is installed and available on the system PATH.```bash
pac help
```
https://aka.ms/PowerPlatformCLI using the following approach:```bash
dotnet tool install --global Microsoft.PowerApps.CLI.Tool
```
dotnet is also not available, direct the user to pac help again.Output: PAC CLI installed and verified
Goal: Ensure the user is authenticated with PAC CLI and has a valid session
Actions:
pac auth who to check the current authentication status.```bash
pac auth who
```
Environment ID:Cloud: (e.g., Public, UsGov, UsGovHigh, UsGovDod, China)Proceed to Phase 3.
AskUserQuestion to ask for the environment URL:| Question | Header | Options |
|---|---|---|
You are not authenticated with PAC CLI. Please provide your Power Pages environment URL (e.g., https://org12345.crm.dynamics.com) so I can authenticate you. | Auth | (free text input via "Other") |
Provide two placeholder options to guide the user:
```bash
pac auth create --environment "
```
This will open a browser window for the user to sign in.
pac auth who again.Output: Authenticated PAC CLI session with environment name and URL extracted
Goal: Ensure the user is deploying to the correct target environment
Actions:
🚦 Gate (consent · deploy-site:3.confirm-env): Echo the current environment and require explicit confirmation before any upload. Covers the follow-up "pick a different env" sub-prompt in the same section — wrong-env deploys are the #1 destructive shared-state failure for this skill, so this gate must fire even when PAC CLI shows a recognizable env.
Trigger: Phase 3 entry; environment resolved from PAC CLI.
Why we ask: Site uploaded to wrong tenant / wrong env — committed to a Dataverse instance the user did not intend; cleanup requires manual deletion or another deploy from a different env.
Cancel leaves: Nothing — no upload fired.
Use AskUserQuestion with the following structure:
| Question | Header | Options |
|---|---|---|
| You are currently connected to environment: | Environment | Yes, use this environment, No, let me choose a different one |
pac org list to retrieve all available environments:```bash
pac org list
```
AskUserQuestion to present the available environments as options (pick up to 4 most relevant, or let user specify).```bash
pac org select --environment "
```
pac auth who again.Output: Confirmed target environment for deployment
Goal: Locate the project, build it, and upload to Power Pages
Actions:
Determine the project root directory. The project root is the directory containing powerpages.config.json. Use Glob to search for it:
**/powerpages.config.json
🚦 Gate (plan · deploy-site:4.1.multi-project): More than onepowerpages.config.jsoncandidate found — pick the right project to deploy.
Trigger: Phase 4.1 glob returned multiple matches.
Why we ask: Wrong project uploaded — pollutes the target env with files from a different site.
Cancel leaves: Nothing — no upload fired.
If found in the current working directory or a subdirectory, use that directory as PROJECT_ROOT. If multiple are found, ask the user which one to deploy using AskUserQuestion.
If not found, ask the user to provide the path to the project root.
If .powerpages-site already exists (i.e., this is not the first deployment), table permissions and site settings may have drifted from the code since the last deployment. Offer to audit before deploying.
🚦 Gate (plan · deploy-site:4.2.audit-permissions): Re-deployment detected — offer to run/audit-permissionsto catch drift between code and table permissions before push.
Trigger:.powerpages-siteexists (not first deployment).
Why we ask: Stale permission YAML deploys, causing 403s for users until next audit run.
Cancel leaves: Nothing — audit is read-only; declining just proceeds to build + upload.
Use AskUserQuestion:
| Question | Header | Options |
|---|---|---|
| This site has been deployed before. Would you like to run a permissions audit to verify table permissions match your current code before deploying? | Audit | Yes, audit permissions (Recommended), Skip — permissions are up to date |
If "Yes": Invoke /audit-permissions to run the audit. After the audit completes, resume with Step 4.3.
If "Skip": Proceed to Step 4.3.
If .powerpages-site does not exist (first deployment), skip this step — there are no existing permissions to audit.
Before uploading, ensure the site is built:
cd "<PROJECT_ROOT>"
npm run build
If the build fails, stop and help the user fix the build errors before retrying.
Run the upload command:
pac pages upload-code-site --rootPath "<PROJECT_ROOT>"
If the upload succeeds: Proceed to Phase 5 to verify the deployment.
If the upload fails: Check the error message:
.js) attachments → proceed to Phase 6.html type attachments are currently blocked → this is a misleading error. See Troubleshooting: HTML Blocked Attachment Error belowOutput: Site built and uploaded to Power Pages
Goal: Confirm the deployment was successful and handle post-deployment steps
Actions:
.powerpages-site FolderConfirm .powerpages-site exists and list its contents (web-roles/, site-settings/, table-permissions/).
Reference: ${PLUGIN_ROOT}/references/skill-tracking-reference.md
Follow the skill tracking instructions in the reference to record this skill's usage. Use --skillName "DeploySite".
Review the output from the pac pages upload-code-site command in Phase 4 (or Phase 6 retry). Verify it reported a successful upload with no errors.
Stage and commit deployment artifacts:
git add -A
git commit -m "Deploy site to Power Pages"
Run the activation status check:
node "${PLUGIN_ROOT}/scripts/check-activation-status.js" --projectRoot "<PROJECT_ROOT>"
Evaluate the JSON result:
activated is true: Inform the user their site is already activated (show websiteUrl if present). Proceed to step 5.6, then skip to Suggest Next Steps. Do NOT ask about activation.activated is false: Proceed to step 5.5.1.error is present: Fall back to step 5.5.1. Do not block the deployment flow.🚦 Gate (plan · deploy-site:5.5.1.activate): Site deployed but not yet activated — offer to invoke/activate-siteto provision the subdomain.
Trigger: Phase 5.5 detectedactivated:falseon a fresh deploy.
Why we ask: Auto-activating writes the wrong subdomain (permanent for the site); auto-skipping leaves the site without a live URL.
Cancel leaves: Nothing — no activation API call fired.
Ask the user if they want to activate the site using AskUserQuestion:
| Question | Header | Options |
|---|---|---|
| Site deployed successfully! Would you like to activate (provision) the site now so it gets a live URL? | Activate | Activate now (Recommended) — Provision the site with a subdomain and make it live, Skip for now — I'll activate later |
If "Activate now": Invoke the /activate-site skill. After activation completes, proceed to step 5.6 to clear the site cache.
If "Skip for now": Suggest next steps (see Suggest Next Steps).
After confirming the site is activated (either it was already activated in step 5.5, or the user just activated it in step 5.5.1), offer to clear the runtime cache so the deployed changes are immediately visible.
Prerequisites: The site must be activated and the project root must be known (from Phase 4.1).
🚦 Gate (plan · deploy-site:5.6.restart-cache): Restart the deployed site to flush its runtime cache. Brief production downtime (a few seconds) — explicit user consent required even though the action is recoverable.
Trigger: Site confirmed activated (either pre-existing or just activated in 5.5.1).
Why we ask: Production users see stale content for several minutes until cache TTL expires.
Cancel leaves: Nothing — the deploy itself already succeeded; cache will refresh on its own.
Use AskUserQuestion to confirm before proceeding:
| Question | Header | Options |
|---|---|---|
| Would you like to restart the site so your latest changes are immediately visible? This may cause a brief downtime (a few seconds). | Restart | Yes, restart site (Recommended), Skip — I'll restart later |
If "Skip": Skip to Suggest Next Steps.
If "Yes": Run the cache-clearing script, passing the project root:
node "${PLUGIN_ROOT}/scripts/clear-site-cache.js" --projectRoot "<PROJECT_ROOT>"
The script reads siteName from powerpages.config.json, looks up the website via the Power Platform API, and restarts it to flush the runtime cache.
Evaluate the result:
success is true: Inform the user: "Site cache cleared — your latest changes should now be visible at success is false: Warn the user that cache clearing failed and show the error, but do not block the deployment flow. The deployment itself succeeded; cache will eventually refresh on its own. Suggest the user can manually clear cache from the Power Pages admin center if needed.Output: Deployment verified, changes committed, activation offered, cache cleared
Goal: Resolve blocked JavaScript attachment errors and retry deployment
Actions:
Tell the user:
"The upload failed because JavaScript (.js) file attachments are blocked in your Power Pages environment. This is a security setting that prevents uploading .js files. To deploy a SPA site, this restriction needs to be relaxed for .js files."
🚦 Gate (consent · deploy-site:6.2.unblock-js): Reactiveblockedattachmentsmodification — destructive shared-state change (tenant-wide env setting). Same shape asdeploy-pipeline:7.6.2.blocked-attachments.
Trigger: Upload failed with the blocked-.jserror.
Why we ask: Auto-unblocking modifies a tenant security setting without explicit consent — visible across the entire environment, not just this site.
Cancel leaves:attachment-block-modifiedis only possible if the user approved and then a downstream step partial-completed. Pure Cancel here leaves nothing — the original blockedattachments value is untouched.
Use AskUserQuestion:
| Question | Header | Options |
|---|---|---|
| Would you like to remove the JavaScript (.js) block from the environment's blocked attachments list? This is required to deploy SPA sites. | Unblock JS | Yes, remove the .js block (Recommended), No, do not change environment settings |
If "No": Stop and inform the user that the deployment cannot proceed without unblocking .js attachments.
If "Yes": Proceed to 6.3.
pac env list-settings to retrieve the current environment settings:```bash
pac env list-settings
```
blockedattachments property in the output. It will contain a semicolon-separated list of file extensions (e.g., ade;adp;app;asa;ashx;asmx;asp;bas;bat;cdx;cer;chm;class;cmd;com;config;cnt;cpl;crt;csh;der;dll;exe;fxp;hlp;hta;htr;htw;ida;idc;idq;inf;ins;isp;its;js;jse;ksh;lnk;mad;maf;mag;mam;maq;mar;mas;mat;mau;mav;maw;mda;mdb;mde;mdt;mdw;mdz;msc;msh;msh1;msh1xml;msh2;msh2xml;mshxml;msi;msp;mst;ops;pcd;pif;prf;prg;printer;pst;reg;rem;scf;scr;sct;shb;shs;shtm;shtml;soap;stm;tmp;url;vb;vbe;vbs;vsmacros;vss;vst;vsw;ws;wsc;wsf;wsh).js from the list. Parse the semicolon-separated values, filter out js, and rejoin with semicolons.```bash
pac env update-settings --name blockedattachments --value "
```
Run the upload command again:
pac pages upload-code-site --rootPath "<PROJECT_ROOT>"
If it succeeds: Proceed to Phase 5 to verify the deployment.
If it fails again with a different error, present the error to the user and help troubleshoot.
Output: JavaScript unblocked and site deployed successfully
If the upload fails with an error like:
Error: Unable to upload webfile name 'index.html' with record Id <GUID> as '.html' type attachments are currently blocked on this environment.
This error is misleading. The .html extension is not actually blocked — the real cause is a stale environment manifest file in the .powerpages-site folder. This manifest maps local files to Dataverse record IDs from a previous upload. When the mapping becomes outdated (e.g., after environment changes or record deletions), the upload fails with this confusing error.
.powerpages-site folder. It follows the naming pattern <environment-host>-manifest.yml (e.g., demo1.crm.dynamics.com-manifest.yml). Use your file-search tool (e.g., Glob with pattern <PROJECT_ROOT>/.powerpages-site/*-manifest.yml) to find it.<PROJECT_ROOT>/.powerpages-site/<environment-host>-manifest.yml.```bash
pac pages upload-code-site --rootPath "
```
If the retry succeeds, proceed to Phase 5. If it fails with a different error, present the error to the user and help them troubleshoot.
Important: Do NOT attempt to unblock .html in the environment's blocked attachments list — the error is not caused by the attachment block setting.
If the user skips activation (or after activation completes), suggest:
/activate-site — Provision the site with a subdomain and make it live (if not already activated)/test-site — Test the deployed site in the browser (verify pages load, check API calls)/audit-permissions — Audit table permissions against current code (recommended after redeployments)/setup-datamodel — Create Dataverse tables for dynamic content/add-seo — Add meta tags, robots.txt, sitemap.xml, faviconpac pages uploadAlways use pac pages upload-code-site — never use pac pages upload. The pac pages upload command is designed for portal-studio-style sites and will corrupt SPA site metadata if used on a SPA site project. This applies to every upload step in this skill (Phase 4.4, Phase 6.4, and troubleshooting retries).
powerpages.config.json found, ask which project to deployBefore starting Phase 1, create a task list with all phases using TaskCreate:
| Task subject | activeForm | Description |
|---|---|---|
| Verify PAC CLI installation | Verifying PAC CLI | Check if PAC CLI is installed, install if missing |
| Verify authentication | Verifying authentication | Check current auth status, authenticate if needed |
| Confirm target environment | Confirming environment | Show current environment, let user confirm or switch |
| Deploy the SPA site | Deploying site | Locate project root, build, and upload via pac pages upload-code-site |
| Verify deployment | Verifying deployment | Confirm .powerpages-site folder exists, review upload output, commit changes, offer activation |
| Handle blocked JavaScript | Resolving JS block | If upload fails due to blocked JS, offer to unblock and retry |
Mark each task in_progress when starting it and completed when done via TaskUpdate. Phase 6 may be marked completed immediately if no JavaScript blocking issue is encountered. This gives the user visibility into progress and keeps the workflow deterministic.
Begin with Phase 1: Verify PAC CLI