devops-deploy-activate

Activate deployment pipeline — secrets, platforms, supply chain hardening, verification.

Context Files

• $JAAN_CONTEXT_DIR/tech.md — Tech stack (determines deployment targets, package manager)
  • Uses sections: #current-stack, #frameworks, #constraints
• $JAAN_CONTEXT_DIR/config.md — Project configuration
• $JAAN_TEMPLATES_DIR/jaan-to:devops-deploy-activate.template.md — Activation report template
• $JAAN_LEARN_DIR/jaan-to:devops-deploy-activate.learn.md — Past lessons (loaded in Pre-Execution)
• ${CLAUDE_PLUGIN_ROOT}/docs/extending/language-protocol.md — Language resolution protocol

Input

Upstream Artifacts: $ARGUMENTS

• infra-scaffold — Path to infra-scaffold output (from /jaan-to:devops-infra-scaffold)
• Empty — Interactive: search $JAAN_OUTPUTS_DIR/devops/infra-scaffold/ for latest output

---

Pre-Execution Protocol

MANDATORY — Read and execute ALL steps in: ${CLAUDE_PLUGIN_ROOT}/docs/extending/pre-execution-protocol.md
Skill name: devops-deploy-activate
Execute: Step 0 (Init Guard) → A (Load Lessons) → B (Resolve Template) → C (Offer Template Seeding)

Also read context files if available:

• $JAAN_CONTEXT_DIR/tech.md — Know the tech stack for platform-specific activation
• $JAAN_CONTEXT_DIR/config.md — Project configuration

Language Settings

Read and apply language protocol: ${CLAUDE_PLUGIN_ROOT}/docs/extending/language-protocol.md
Override field for this skill: language_devops-deploy-activate

Language exception: Generated code output (YAML workflows, shell commands, deployment configs) is NOT affected by this setting and remains in English/code.

---

PHASE 1: Analysis (Read-Only)

Thinking Mode

ultrathink

Use extended reasoning for:

• Analyzing infra-scaffold output to determine required secrets and platforms
• Cross-referencing .env files with CI/CD workflow secret references
• Planning activation order (secrets before platform provisioning)
• Identifying supply chain hardening opportunities in GitHub Actions

Step 1: Parse Infra-Scaffold Output

Read infra-scaffold README and CI/CD workflow files:

1. Extract secret names referenced in workflows (${{ secrets.* }})
   1b. Extract variable names referenced in workflows (${{ vars.* }})
2. Identify deployment platform targets (Vercel, Railway, Fly.io)
3. Detect Docker registry configuration
4. List GitHub Actions used (for SHA pinning)

If no infra-scaffold found:

• Search $JAAN_OUTPUTS_DIR/devops/infra-scaffold/ for latest output
• If still none: ask user for manual input via AskUserQuestion

Present parsed summary:

INFRA-SCAFFOLD ANALYSIS
=======================
Secrets Referenced:    {list with which workflow uses each}
Deployment Platforms:  {list}
Docker Registry:       {registry or "none"}
GitHub Actions Used:   {list with current versions}

Step 2: Check CLI Tool Availability

Verify required tools are available:

• gh auth status (required — GitHub CLI)
• railway version (if Railway detected)
• vercel --version (if Vercel detected)
• fly version (if Fly.io detected)

Present status:

CLI TOOL STATUS
===============
gh:       ✓ Authenticated as {user} / ✗ Not found
railway:  ✓ v{version} / ✗ Not found / ⊘ Not needed
vercel:   ✓ v{version} / ✗ Not found / ⊘ Not needed
fly:      ✓ v{version} / ✗ Not found / ⊘ Not needed

If gh is missing or not authenticated: stop and ask user to install/authenticate.

Step 3: Detect Current State

Check what is already configured:

1. gh secret list — which secrets already exist
2. gh variable list — which repository variables already exist
3. Check for existing platform project links (.vercel/, fly.toml, railway.toml)
4. Check .github/workflows/ for existing CI/CD workflows

Present state:

CURRENT STATE
=============
Secrets Configured:    {count}/{total} ({list})
Secrets Missing:       {list}
Variables Configured:  {count}/{total} ({list})
Variables Missing:     {list}
Platform Links:        {found / none}
Existing Workflows:    {list or "none"}

Step 4: Build Activation Checklist

Create ordered activation plan:

ACTIVATION CHECKLIST
====================
                                          Status
1. GitHub Secrets                         {needed / configured / partial}
2. GitHub Repository Variables            {needed / configured / partial}
3. GitHub Actions SHA Pinning             {needed / done}
4. Backend Platform ({platform})          {needed / linked}
5. Frontend Platform ({platform})         {needed / linked}
6. Repository Variables (post-provision)  {needed / configured}
7. Remote Cache (Turborepo)               {needed / n/a}
8. Verification Pipeline                  {needed}

Items to activate: {count}
Already configured: {count}

Mark items as: needed / already configured / optional / n/a

---

HARD STOP — Review Activation Plan

Use AskUserQuestion:

• Question: "Proceed with activating {n} deployment components?"
• Header: "Activate"
• Options:
  • "Yes" — Proceed with activation
  • "No" — Cancel
  • "Edit" — Let me revise which components to activate

Do NOT proceed to Phase 2 without explicit approval.

---

PHASE 2: Execution

Step 5: Configure GitHub Secrets

For each required secret:

1. Show: name, description, which workflow uses it
2. Use AskUserQuestion to collect the value
3. Run gh secret set {NAME} with the provided value
4. Never log or display secret values after entry

Reference: See ${CLAUDE_PLUGIN_ROOT}/docs/extending/devops-deploy-activate-reference.md section "Secret Management Patterns" for per-platform secret requirements and naming conventions.

Confirm after each:

Secret {NAME} configured for repository.

Step 6: Harden GitHub Actions

Reference: See ${CLAUDE_PLUGIN_ROOT}/docs/extending/devops-deploy-activate-reference.md section "GitHub Actions SHA Pinning" for the resolution process using gh api.

For each GitHub Action reference in workflow files:

1. Parse action reference (e.g., actions/checkout@v4)
2. Resolve to full SHA digest via gh api repos/{owner}/{repo}/git/ref/tags/{tag}
3. Show proposed change: actions/checkout@v4 → actions/checkout@{sha} # v4
4. Apply via Edit(.github/**)

Present pinning summary:

SHA PINNING RESULTS
===================
Pinned:    {count} actions
Skipped:   {count} (already pinned)
Failed:    {count} (show reasons)

Step 7: Provision Backend Platform

Based on detected deployment target:

Reference: See ${CLAUDE_PLUGIN_ROOT}/docs/extending/devops-deploy-activate-reference.md section "Platform CLI Provisioning" for Railway, Fly.io, and other platform setup commands.

Railway:

1. railway init — Create project
2. railway link — Link to repository
3. Configure environment variables via railway variables set
4. Verify: railway status

Fly.io:

1. fly launch — Create app (interactive)
2. fly secrets set — Configure secrets
3. Verify: fly status

Capture service URL from platform CLI output for repository variables (Step 8b).

Confirm:

Backend platform ({platform}) provisioned and linked.

Step 8: Connect Frontend Platform

Based on detected deployment target:

Vercel:

1. vercel link — Link to repository
2. Configure environment variables via Vercel dashboard or CLI
3. Set up preview deployments for PRs
4. Verify: vercel inspect

Capture production URL from platform CLI output for repository variables (Step 8b).

Confirm:

Frontend platform ({platform}) connected with preview deployments.

Step 8b: Configure Repository Variables

Set non-sensitive configuration values as GitHub repository variables:

1. gh variable list — show currently configured variables
2. For each required variable (from Step 1b ${{ vars.* }} extraction):
   • Detect value from platform provisioning output if possible (service URLs)
   • Confirm value via AskUserQuestion before setting
   • Run gh variable set {NAME} --body "{value}"
   • For environment-scoped: gh variable set {NAME} --env production --body "{value}"
3. Present configured variables:

REPOSITORY VARIABLES
====================
{NAME}:           {value}
{NAME}:           {value}
Environment-scoped: {list}

Reference: See ${CLAUDE_PLUGIN_ROOT}/docs/extending/devops-deploy-activate-reference.md section "Repository Variable Patterns" for common variables, URL capture commands, and secrets-vs-variables guidance.

Step 9: Configure Remote Cache (if monorepo)

Only if turbo.json exists in the project:

1. Guide user to create Turborepo remote cache token
2. Set TURBO_TOKEN and TURBO_TEAM as GitHub secrets
3. Verify with turbo run build --dry

Reference: See ${CLAUDE_PLUGIN_ROOT}/docs/extending/devops-deploy-activate-reference.md section "Caching Hierarchy" for the full caching strategy (pnpm store → Next.js → Docker → Turborepo).

If turbo.json not found: skip with note.

Step 10: Trigger Verification Pipeline

1. Suggest: push a test commit or use gh workflow run ci.yml
2. Monitor pipeline status via gh run list --limit 1
3. Wait for completion and report results

PIPELINE VERIFICATION
=====================
Workflow:   {workflow_name}
Run ID:     {run_id}
Status:     ✓ Pass / ✗ Failed at stage {stage}
Duration:   {time}
URL:        {run_url}

If failed: show failure details and suggest fixes.

Step 11: Quality Check

Secrets:

• All required secrets configured in GitHub
• No secrets logged or displayed in output
• Secrets match workflow references

Supply Chain:

• All GitHub Actions pinned to SHA digests
• No mutable tags (v4, latest) remaining in workflows

Platforms:

• Backend platform connected and responding
• Frontend platform connected with preview deployments
• Remote cache configured (if applicable)

Repository Variables:

• All required ${{ vars.* }} references have corresponding variables set
• Non-sensitive values used as variables (not secrets)
• Environment-scoped variables set where needed

Pipeline:

• CI pipeline triggered and completed
• All workflow stages passed

If any check fails, fix before proceeding.

Step 12: Generate ID and Folder Structure

source "${CLAUDE_PLUGIN_ROOT}/scripts/lib/id-generator.sh"
SUBDOMAIN_DIR="$JAAN_OUTPUTS_DIR/devops/deploy-activate"
mkdir -p "$SUBDOMAIN_DIR"
NEXT_ID=$(generate_next_id "$SUBDOMAIN_DIR")
slug="{activation-slug}"
OUTPUT_FOLDER="${SUBDOMAIN_DIR}/${NEXT_ID}-${slug}"

Preview:

Output Configuration

• ID: {NEXT_ID}
• Folder: $JAAN_OUTPUTS_DIR/devops/deploy-activate/{NEXT_ID}-{slug}/
• Main file: {NEXT_ID}-{slug}.md

Step 13: Write Activation Report

Use template from: $JAAN_TEMPLATES_DIR/jaan-to:devops-deploy-activate.template.md

Write {NEXT_ID}-{slug}.md with:

• Executive Summary
• Secrets configured (names only, never values)
• Actions hardened (SHA pinning results)
• Platforms provisioned (with connection status)
• Repository variables configured (names and values — non-secret)
• Pipeline verification results
• Remote cache status
• Next steps for ongoing maintenance

Update index:

source "${CLAUDE_PLUGIN_ROOT}/scripts/lib/index-updater.sh"
add_to_index \
  "$SUBDOMAIN_DIR/README.md" \
  "$NEXT_ID" \
  "${NEXT_ID}-${slug}" \
  "{Activation Title}" \
  "{Executive summary — 1-2 sentences}"

Confirm:

Activation report written to: $JAAN_OUTPUTS_DIR/devops/deploy-activate/{NEXT_ID}-{slug}/{NEXT_ID}-{slug}.md
Index updated: $JAAN_OUTPUTS_DIR/devops/deploy-activate/README.md

Step 14: Suggest Next Actions

Deployment pipeline activated!

Immediate Steps:

• Monitor the first real deployment
• Verify preview deployments work on PR creation
• Check secret rotation schedule

Recommended Follow-ups:

• Set up monitoring and alerting for deployed services
• Configure branch protection rules (gh api repos/{owner}/{repo}/branches/{branch}/protection)
• Run /jaan-to:release-iterate-changelog for your first release
• Schedule secret rotation (recommended: 90 days)

Step 15: Capture Feedback

Use AskUserQuestion:

• Question: "How did the deployment activation turn out?"
• Header: "Feedback"
• Options:
  • "Perfect!" — Done
  • "Needs fixes" — What should I improve?
  • "Learn from this" — Capture a lesson for future runs

If "Learn from this": Run /jaan-to:learn-add devops-deploy-activate "{feedback}"

---

Scope Boundaries

• Does NOT generate infrastructure files (that's /jaan-to:devops-infra-scaffold)
• Does NOT copy files into the project (that's /jaan-to:dev-output-integrate)
• Requires user to have CLI tools installed and authenticated
• Secrets are set by the USER through guided prompts (skill never handles secret values directly)

---

DAG Position

devops-infra-scaffold + dev-output-integrate
  |
  v
devops-deploy-activate

---

Skill Alignment

• Two-phase workflow with HARD STOP for human approval
• Multi-platform support (GitHub Actions, GitLab CI, etc.)
• Template-driven output structure
• Output to standardized $JAAN_OUTPUTS_DIR path

Definition of Done

• All required GitHub secrets configured
• All required GitHub repository variables configured
• GitHub Actions pinned to SHA digests
• Backend platform provisioned and connected
• Frontend platform connected with preview deployments
• Remote cache configured (if monorepo)
• CI pipeline triggered and passed
• Activation report written to output directory
• Index updated with executive summary
• User approved final result