ac-framework 2.3.0 → 2.4.0

package/framework/mobile_development/.windsurf/skills/vibe-security/references/payments.md

@@ -0,0 +1,63 @@
+# Payment Security (Stripe)
+
+## Never Trust Client-Submitted Prices
+
+The #1 payment vulnerability in vibe-coded apps: the price comes from the client. An attacker can set any amount, including $0.
+
+```typescript
+// BAD: price comes from the request body
+const session = await stripe.checkout.sessions.create({
+  line_items: [{
+    price_data: {
+      currency: 'usd',
+      unit_amount: req.body.price, // attacker controls this
+      product_data: { name: req.body.name },
+    },
+    quantity: 1,
+  }],
+});
+
+// GOOD: look up the price server-side
+const product = await db.products.findUnique({ where: { id: req.body.productId } });
+if (!product) return new Response('Not found', { status: 404 });
+
+const session = await stripe.checkout.sessions.create({
+  line_items: [{ price: product.stripePriceId, quantity: 1 }],
+});
+```
+
+Use Stripe Price IDs (created via the Stripe dashboard or API) rather than constructing prices from your database. This way, prices are defined in Stripe and can't be manipulated.
+
+## Webhook Signature Verification
+
+Stripe webhooks must have their signatures verified. This requires the **raw request body** — parsing the body as JSON first destroys the signature.
+
+```typescript
+// Express: webhook route MUST use express.raw() BEFORE express.json()
+app.post('/webhook', express.raw({ type: 'application/json' }), (req, res) => {
+  const sig = req.headers['stripe-signature'];
+  const event = stripe.webhooks.constructEvent(req.body, sig, webhookSecret);
+  // ... handle event
+});
+
+// Next.js App Router: use request.text(), NOT request.json()
+export async function POST(request: Request) {
+  const body = await request.text();
+  const sig = request.headers.get('stripe-signature')!;
+  const event = stripe.webhooks.constructEvent(body, sig, webhookSecret);
+  // ... handle event
+}
+```
+
+## Subscription Status Validation
+
+Check subscription status **server-side on every protected request** using your database (kept in sync via webhooks). Do not rely on:
+- A cached session value from login time
+- A client-side flag
+- A JWT claim that was set at token creation and never refreshed
+
+Subscriptions can be cancelled, expire, or change tier at any time. Your database (updated via webhooks) is the source of truth.
+
+## Checkout Session Metadata
+
+Validate that checkout session metadata (user ID, plan, etc.) was set **server-side** when creating the session, not passed from the client. If metadata comes from the client, an attacker can claim to be a different user or select a different plan.

package/framework/mobile_development/.windsurf/skills/vibe-security/references/rate-limiting.md

@@ -0,0 +1,55 @@
+# Rate Limiting & Abuse Prevention
+
+## Where Rate Limiting Is Required
+
+Every one of these endpoints needs rate limiting. AI assistants almost never add it:
+
+- **Auth endpoints** — login, register, password reset, OTP verification, magic link. Without limits, attackers can brute-force passwords or enumerate accounts.
+- **AI API calls** — Any endpoint that calls OpenAI, Anthropic, or similar. A single user can drain your entire monthly budget in minutes.
+- **Email / SMS sending** — Attackers can use your app as a spam relay.
+- **File processing** — Upload, resize, convert. CPU-intensive operations without limits enable denial-of-service.
+- **Webhook-like endpoints** — Anything accepting external input at scale.
+
+## Don't Store Rate Limits in Public Tables
+
+If rate limit counters live in a Supabase public table, users can reset their own counters via the REST API. Use:
+
+- **Upstash Redis** — Serverless Redis with built-in rate limiting primitives
+- **Private schema table** — Not exposed via PostgREST
+- **Middleware-level limiting** — At the edge or API gateway
+- **In-memory stores** — For single-server deployments (Redis for multi-server)
+
+## Combine Per-IP and Per-User Limiting
+
+- IP-only limits are defeated by rotating IPs (trivial with VPNs or botnets)
+- User-only limits are defeated by creating new accounts
+- Use both together for effective protection
+
+## Billing Protection
+
+- Set billing alerts on every cloud provider (AWS, GCP, Vercel, etc.)
+- Set **hard spending caps** on AI API providers (OpenAI, Anthropic)
+- Use per-user usage quotas with hard limits, not just soft warnings
+- Monitor for anomalous usage patterns (sudden spikes, requests at odd hours)
+
+## Implementation Pattern
+
+```typescript
+// Example: rate limiting with Upstash Redis
+import { Ratelimit } from '@upstash/ratelimit';
+import { Redis } from '@upstash/redis';
+
+const ratelimit = new Ratelimit({
+  redis: Redis.fromEnv(),
+  limiter: Ratelimit.slidingWindow(10, '1 m'), // 10 requests per minute
+});
+
+export async function POST(request: Request) {
+  const ip = request.headers.get('x-forwarded-for') ?? '127.0.0.1';
+  const { success } = await ratelimit.limit(ip);
+  if (!success) {
+    return new Response('Too many requests', { status: 429 });
+  }
+  // ... handle request
+}
+```

package/framework/mobile_development/.windsurf/skills/vibe-security/references/secrets-and-env.md

@@ -0,0 +1,47 @@
+# Secrets & Environment Variables
+
+## Hardcoded Credentials
+
+Never hardcode API keys, tokens, passwords, or credentials in source code. This includes:
+- Strings that look like API keys in source files
+- Connection strings with embedded passwords
+- Private keys or certificates in the repo
+
+If a secret was ever committed to Git history, consider it compromised — deleting the file doesn't remove it from history. The key must be rotated immediately. Run `gitleaks detect` to scan for leaked secrets.
+
+## Client-Side Environment Variable Prefixes
+
+These prefixes cause env vars to be inlined into the client bundle at build time. Everything in the bundle is visible to anyone:
+
+| Framework | Client Prefix | Danger |
+|-----------|--------------|--------|
+| Next.js | `NEXT_PUBLIC_` | Inlined into browser JS at build time |
+| Vite | `VITE_` | Inlined into browser JS at build time |
+| Expo / React Native | `EXPO_PUBLIC_` | Baked into the app bundle |
+| Create React App | `REACT_APP_` | Inlined into browser JS at build time |
+
+**What belongs client-side:**
+- Stripe publishable key (`pk_live_*`, `pk_test_*`)
+- Supabase anon key
+- Firebase client config (apiKey, authDomain, projectId)
+- Public analytics IDs
+
+**What must NEVER be client-side:**
+- Supabase `service_role` key (bypasses all RLS)
+- Stripe secret key (`sk_live_*`, `sk_test_*`)
+- Any database connection string
+- Any third-party API secret key
+- JWT signing secrets
+- OAuth client secrets
+
+## .gitignore
+
+Ensure `.env`, `.env.local`, `.env.*.local`, and any file containing secrets is in `.gitignore` **before the first commit**. Check that `.env.example` or `.env.sample` files contain only placeholder values, not real keys.
+
+## Detection Tips
+
+When auditing, search for:
+- Files named `.env` that are tracked by git (`git ls-files | grep .env`)
+- Strings matching common key patterns: `sk_live_`, `sk_test_`, `AKIA`, `ghp_`, `glpat-`, `xoxb-`, `Bearer `
+- `process.env.NEXT_PUBLIC_` or `import.meta.env.VITE_` referencing anything with "secret", "private", "service", or "key" in the name
+- Hardcoded URLs containing credentials (e.g., `postgresql://user:password@host`)

package/framework/mobile_development/.windsurf/workflows/ac-lite.md

@@ -43,6 +43,57 @@ acfm memory stats
 
 ---
 
+
+## SynapseGrid Collaboration Protocol (Optional but Recommended for Complex Tasks)
+
+Use SynapseGrid when a task benefits from role-based collaboration and explicit review loops.
+
+**Delegate to SynapseGrid when:**
+- Scope spans planning + implementation + review
+- Risk is medium/high (security, migrations, API contract changes)
+- You need auditable outputs (transcript, meeting summary, artifacts)
+- A single-pass implementation is likely to miss edge cases
+
+**Role delegation map:**
+- `planner` -> plan/constraints/acceptance criteria
+- `critic` -> risk analysis and challenge assumptions
+- `coder` -> implementation and concrete edits
+- `reviewer` -> final verification and readiness checks
+
+**Preferred MCP flow:**
+1. `collab_start_session`
+2. `collab_invoke_team`
+3. `collab_wait_run`
+4. `collab_get_result`
+5. Optional: `collab_get_transcript`, `collab_get_meeting_log`, `collab_status`
+
+**CLI fallback:**
+```bash
+acfm agents setup
+acfm agents runtime set auto
+acfm agents doctor --verbose
+acfm agents start --task "..." --mux auto
+acfm agents transcript --role all --limit 80
+acfm agents summary
+acfm agents artifacts --watch --interval 1200
+```
+
+**Model and runtime controls:**
+```bash
+acfm agents model list
+acfm agents model choose
+acfm agents runtime get
+acfm agents runtime install-zellij
+```
+
+**Artifacts to inspect:**
+- `~/.acfm/synapsegrid/<sessionId>/meeting-log.md`
+- `~/.acfm/synapsegrid/<sessionId>/meeting-summary.md`
+- `~/.acfm/synapsegrid/<sessionId>/turns/raw/*.ndjson`
+
+**Lite rule:** delegate only when collaboration adds value; otherwise continue with normal `ac-lite` flow.
+
+
 ## Conditional Skill Gates
 
 ### Platform/Stack Gate
@@ -57,6 +108,9 @@ Load based on selected implementation stack:
 - `mobile-ios-design` [IF iOS UI changes]
 - `mobile-android-design` [IF Android UI changes]
 
+### Security Gate
+- `vibe-security` [MANDATORY final audit before closure]
+
 ### Clarity/Scale Gate
 - `spec-clarification` [IF requirements ambiguous]
 - `spec-analysis` [IF cross-artifact consistency needs verification]
@@ -87,7 +141,8 @@ Before `openspec-apply-change`, all must be true:
 ### Gate B: Ready to Close
 Before `openspec-archive-change`, all must be true:
 - `openspec-verify-change` completed
-- Critical issues resolved (or explicitly accepted with warning)
+- `vibe-security` executed as final security validation
+- No unresolved CRITICAL/HIGH findings remain (or explicitly accepted with warning)
 - Task list reflects final state
 - Reusable context saved to memory
 - `documentation` / `sync-index` / changelog steps executed as required
@@ -105,8 +160,9 @@ Before `openspec-archive-change`, all must be true:
 6. Pass Gate A.
 7. Implement (`openspec-apply-change`).
 8. Verify (`openspec-verify-change`).
-9. Pass Gate B.
-10. Archive/sync and generate closure docs.
+9. Run `vibe-security` for final security audit and remediate findings.
+10. Pass Gate B.
+11. Archive/sync and generate closure docs.
 
 ### Existing Mobile Change
 1. Confirm initialization and target change.
@@ -116,8 +172,9 @@ Before `openspec-archive-change`, all must be true:
 5. Pass Gate A.
 6. Implement and debug as needed.
 7. Verify.
-8. Pass Gate B.
-9. Archive/sync.
+8. Run `vibe-security` for final security audit and remediate findings.
+9. Pass Gate B.
+10. Archive/sync.
 
 ---
 

package/framework/mobile_development/.windsurf/workflows/ac.md

@@ -67,6 +67,12 @@ If you attempt to proceed without completing a required step, you MUST STOP and
 | `sync-index` | Keep project index and generated guidance synchronized with implementation. | Documentation sync | `openspec-apply-change` |
 | `changelog-generator` | Create user-facing release notes from implementation changes. | Release communication | `openspec-archive-change` |
 
+### Security Validation Skills
+
+| Skill | Description | Primary Use | Required Before |
+|-------|-------------|-------------|-----------------|
+| `vibe-security` | Final security audit for AI-generated/vibe-coded risks (exposed keys, broken access control, auth/payment/data validation gaps). | Final security validation before closure | `openspec-archive-change` |
+
 ---
 
 ## Persistent Memory Protocol (Mandatory)
@@ -97,81 +103,74 @@ The AC Framework includes a persistent memory system that agents must use proact
 
 If SynapseGrid is enabled in `acfm init`, AC Framework installs the collaborative MCP server automatically for detected assistants.
 
+**When to delegate to SynapseGrid:**
+- Multi-step features with architecture + implementation + review work
+- Tasks requiring critical analysis before coding (security, API contracts, refactors)
+- Long-running work where transcript, artifacts, and resumability are important
+- Situations where one assistant would benefit from role-based challenge/review
+
+**Role delegation model (what each agent does best):**
+- `planner`: breaks down scope, constraints, approach, acceptance criteria
+- `critic`: challenges assumptions, finds risks/blind spots, proposes safer alternatives
+- `coder`: implements code and applies concrete changes
+- `reviewer`: validates quality, consistency with spec/tasks, and production readiness
+
 **Session-start requirement when collaboration is enabled:**
-1. Prefer the available SynapseGrid MCP tools for collaborative session control before falling back to direct CLI.
-2. Use shared session state and transcript as the source of truth for role-by-role collaboration.
-3. If collaborative MCP is unavailable, use CLI fallback commands and keep behavior equivalent.
-
-**How to use SynapseGrid collaboration:**
-```text
-Preferred: use SynapseGrid MCP tools (ac-framework-collab) for start/status/step/resume/stop.
-Fallback: use AC Framework agents CLI commands directly.
-```
+1. Prefer SynapseGrid MCP tools for start/wait/result/cancel/status flows.
+2. Use transcript + meeting summary + artifacts as source of truth.
+3. If collaborative MCP is unavailable, use equivalent CLI commands.
+
+**Preferred MCP flow (recommended):**
+1. `collab_start_session`
+2. `collab_invoke_team`
+3. `collab_wait_run`
+4. `collab_get_result`
+5. Optional diagnostics: `collab_get_transcript`, `collab_get_meeting_log`, `collab_status`
+
+**CLI fallback and operations:**
 ```bash
-# Install/update collaborative MCP server integrations
-acfm agents install-mcps
-
-# Start collaborative runtime with visible 4-pane tmux war-room
-acfm agents start --task "design and implement feature X" --attach --model-coder provider/model
+# Setup/runtime
+acfm agents setup
+acfm agents runtime install-zellij
+acfm agents runtime set auto
+acfm agents doctor --verbose
 
-# If already running in background, attach to panes and inspect logs
+# Start/run
+acfm agents start --task "design and implement feature X" --mux auto
 acfm agents live
-acfm agents logs --role all --lines 120
+acfm agents status
 
-# Resume/list/export operations
-acfm agents resume
-acfm agents list
+# Visibility and artifacts
+acfm agents transcript --role all --limit 80
+acfm agents summary
+acfm agents artifacts --watch --interval 1200
 acfm agents export --format md --out synapse-session.md
 
-# Fast diagnostics for OpenCode resolution errors
-acfm agents setup
-```
-
-**Troubleshooting notes:**
-- If transcript shows `Agent failed: spawn opencode ENOENT`, run `acfm agents setup` and restart/resume session.
-- If session exists but panes are not visible, use `acfm agents live` (or `acfm agents attach`) or `tmux attach -t <session-name>`.
-- For MCP-first flows, ensure session start enables worker spawning so tmux panes are created.
-- Per-role models can be set at start (`--model-planner`, `--model-critic`, `--model-coder`, `--model-reviewer`) or persisted via `acfm agents model set`.
+# Model management
+acfm agents model list
+acfm agents model choose
+acfm agents model set --role coder provider/model
 
-## SynapseGrid Collaborative MCP Protocol (Optional)
-
-If SynapseGrid is enabled in `acfm init`, AC Framework installs the collaborative MCP server automatically for detected assistants.
-
-**Session-start requirement when collaboration is enabled:**
-1. Prefer the available SynapseGrid MCP tools for collaborative session control before falling back to direct CLI.
-2. Use shared session state and transcript as the source of truth for role-by-role collaboration.
-3. If collaborative MCP is unavailable, use CLI fallback commands and keep behavior equivalent.
-
-**How to use SynapseGrid collaboration:**
-```text
-Preferred: use SynapseGrid MCP tools (ac-framework-collab) for start/status/step/resume/stop.
-Fallback: use AC Framework agents CLI commands directly.
+# Lifecycle
+acfm agents resume
+acfm agents stop
 ```
-```bash
-# Install/update collaborative MCP server integrations
-acfm agents install-mcps
 
-# Start collaborative runtime with visible 4-pane tmux war-room
-acfm agents start --task "design and implement feature X" --attach --model-coder provider/model
+**Collaboration artifacts (deterministic, not prompt-dependent):**
+- `~/.acfm/synapsegrid/<sessionId>/transcript.jsonl`
+- `~/.acfm/synapsegrid/<sessionId>/meeting-log.md`
+- `~/.acfm/synapsegrid/<sessionId>/meeting-summary.md`
+- `~/.acfm/synapsegrid/<sessionId>/turns/*.json`
+- `~/.acfm/synapsegrid/<sessionId>/turns/raw/*.ndjson`
+- `~/.acfm/synapsegrid/<sessionId>/turns/raw/*.stderr.log`
+- `~/.acfm/synapsegrid/<sessionId>/diagnostics.json`
 
-# If already running in background, attach to panes and inspect logs
-acfm agents live
-acfm agents logs --role all --lines 120
-
-# Resume/list/export operations
-acfm agents resume
-acfm agents list
-acfm agents export --format md --out synapse-session.md
-
-# Fast diagnostics for OpenCode resolution errors
-acfm agents setup
-```
+**Troubleshooting and robust behavior:**
+- Use `acfm agents doctor --verbose` for capability probe diagnostics.
+- In `--mux auto`, runtime prefers zellij and falls back to tmux when zellij startup fails.
+- If model/provider errors appear, validate with `opencode auth list` and `opencode models`.
+- If workers are running but unclear, inspect `logs`, `transcript`, and `artifacts` before restarting.
 
-**Troubleshooting notes:**
-- If transcript shows `Agent failed: spawn opencode ENOENT`, run `acfm agents setup` and restart/resume session.
-- If session exists but panes are not visible, use `acfm agents live` (or `acfm agents attach`) or `tmux attach -t <session-name>`.
-- For MCP-first flows, ensure session start enables worker spawning so tmux panes are created.
-- Per-role models can be set at start (`--model-planner`, `--model-critic`, `--model-coder`, `--model-reviewer`) or persisted via `acfm agents model set`.
 
 ---
 
@@ -205,7 +204,7 @@ Before proceeding to next phase, verify outputs exist:
 | Phase 2 | exploration notes + platform/stack decision recorded | [ ] |
 | Phase 3 | `proposal.md`, `specs/`, `design.md`, `tasks.md` + `spec-analysis` review | [ ] |
 | Phase 4 | implementation done + tasks updated + issues debugged | [ ] |
-| Phase 5 | verification passed + docs/index synced + archive/sync complete | [ ] |
+| Phase 5 | verification passed + `vibe-security` audit complete + docs/index synced + archive/sync complete | [ ] |
 
 **Rule 4: Pre-Implementation Safety Check**
 Before `openspec-apply-change`, ALL must be TRUE:
@@ -285,11 +284,12 @@ When starting a mobile project **from scratch**, follow this **MANDATORY** workf
     v
   PHASE 5 - VALIDATION AND CLOSURE
     19) openspec-verify-change
-    20) documentation
-    21) sync-index
-    22) openspec-sync-specs [IF syncing specs without archive]
-    23) openspec-archive-change OR openspec-bulk-archive-change
-    24) changelog-generator
+    20) vibe-security (final security audit)
+    21) documentation
+    22) sync-index
+    23) openspec-sync-specs [IF syncing specs without archive]
+    24) openspec-archive-change OR openspec-bulk-archive-change
+    25) changelog-generator
 ```
 
 **Conditional Skill Notes:**
@@ -338,11 +338,12 @@ When working on an **existing mobile codebase** (feature, bugfix, refactor):
 
   PHASE 5 - VALIDATION AND CLOSURE
     16) openspec-verify-change
-    17) documentation
-    18) sync-index
-    19) openspec-sync-specs [if needed]
-    20) openspec-archive-change / openspec-bulk-archive-change
-    21) changelog-generator
+    17) vibe-security (final security audit)
+    18) documentation
+    19) sync-index
+    20) openspec-sync-specs [if needed]
+    21) openspec-archive-change / openspec-bulk-archive-change
+    22) changelog-generator
 ```
 
 ---

package/framework/mobile_development/AGENTS.md

@@ -67,6 +67,12 @@ If you attempt to proceed without completing a required step, you MUST STOP and
 | `sync-index` | Keep project index and generated guidance synchronized with implementation. | Documentation sync | `openspec-apply-change` |
 | `changelog-generator` | Create user-facing release notes from implementation changes. | Release communication | `openspec-archive-change` |
 
+### Security Validation Skills
+
+| Skill | Description | Primary Use | Required Before |
+|-------|-------------|-------------|-----------------|
+| `vibe-security` | Final security audit for AI-generated/vibe-coded risks (exposed keys, broken access control, auth/payment/data validation gaps). | Final security validation before closure | `openspec-archive-change` |
+
 ---
 
 ## Persistent Memory Protocol (Mandatory)
@@ -93,6 +99,80 @@ The AC Framework includes a persistent memory system that agents must use proact
 
 **User communication:** `Memory saved: [brief description]`
 
+
+### SynapseGrid Collaborative MCP Protocol (Optional)
+
+If SynapseGrid is enabled in `acfm init`, AC Framework installs the collaborative MCP server automatically for detected assistants.
+
+**When to delegate to SynapseGrid:**
+- Multi-step features with architecture + implementation + review work
+- Tasks requiring critical analysis before coding (security, API contracts, refactors)
+- Long-running work where transcript, artifacts, and resumability are important
+- Situations where one assistant would benefit from role-based challenge/review
+
+**Role delegation model (what each agent does best):**
+- `planner`: breaks down scope, constraints, approach, acceptance criteria
+- `critic`: challenges assumptions, finds risks/blind spots, proposes safer alternatives
+- `coder`: implements code and applies concrete changes
+- `reviewer`: validates quality, consistency with spec/tasks, and production readiness
+
+**Session-start requirement when collaboration is enabled:**
+1. Prefer SynapseGrid MCP tools for start/wait/result/cancel/status flows.
+2. Use transcript + meeting summary + artifacts as source of truth.
+3. If collaborative MCP is unavailable, use equivalent CLI commands.
+
+**Preferred MCP flow (recommended):**
+1. `collab_start_session`
+2. `collab_invoke_team`
+3. `collab_wait_run`
+4. `collab_get_result`
+5. Optional diagnostics: `collab_get_transcript`, `collab_get_meeting_log`, `collab_status`
+
+**CLI fallback and operations:**
+```bash
+# Setup/runtime
+acfm agents setup
+acfm agents runtime install-zellij
+acfm agents runtime set auto
+acfm agents doctor --verbose
+
+# Start/run
+acfm agents start --task "design and implement feature X" --mux auto
+acfm agents live
+acfm agents status
+
+# Visibility and artifacts
+acfm agents transcript --role all --limit 80
+acfm agents summary
+acfm agents artifacts --watch --interval 1200
+acfm agents export --format md --out synapse-session.md
+
+# Model management
+acfm agents model list
+acfm agents model choose
+acfm agents model set --role coder provider/model
+
+# Lifecycle
+acfm agents resume
+acfm agents stop
+```
+
+**Collaboration artifacts (deterministic, not prompt-dependent):**
+- `~/.acfm/synapsegrid/<sessionId>/transcript.jsonl`
+- `~/.acfm/synapsegrid/<sessionId>/meeting-log.md`
+- `~/.acfm/synapsegrid/<sessionId>/meeting-summary.md`
+- `~/.acfm/synapsegrid/<sessionId>/turns/*.json`
+- `~/.acfm/synapsegrid/<sessionId>/turns/raw/*.ndjson`
+- `~/.acfm/synapsegrid/<sessionId>/turns/raw/*.stderr.log`
+- `~/.acfm/synapsegrid/<sessionId>/diagnostics.json`
+
+**Troubleshooting and robust behavior:**
+- Use `acfm agents doctor --verbose` for capability probe diagnostics.
+- In `--mux auto`, runtime prefers zellij and falls back to tmux when zellij startup fails.
+- If model/provider errors appear, validate with `opencode auth list` and `opencode models`.
+- If workers are running but unclear, inspect `logs`, `transcript`, and `artifacts` before restarting.
+
+
 ---
 
 ## CRITICAL: How to Use Skills - ZERO SKIP POLICY
@@ -125,7 +205,7 @@ Before proceeding to next phase, verify outputs exist:
 | Phase 2 | exploration notes + platform/stack decision recorded | [ ] |
 | Phase 3 | `proposal.md`, `specs/`, `design.md`, `tasks.md` + `spec-analysis` review | [ ] |
 | Phase 4 | implementation done + tasks updated + issues debugged | [ ] |
-| Phase 5 | verification passed + docs/index synced + archive/sync complete | [ ] |
+| Phase 5 | verification passed + `vibe-security` audit complete + docs/index synced + archive/sync complete | [ ] |
 
 **Rule 4: Pre-Implementation Safety Check**
 Before `openspec-apply-change`, ALL must be TRUE:
@@ -205,11 +285,12 @@ When starting a mobile project **from scratch**, follow this **MANDATORY** workf
     v
   PHASE 5 - VALIDATION AND CLOSURE
     19) openspec-verify-change
-    20) documentation
-    21) sync-index
-    22) openspec-sync-specs [IF syncing specs without archive]
-    23) openspec-archive-change OR openspec-bulk-archive-change
-    24) changelog-generator
+    20) vibe-security (final security audit)
+    21) documentation
+    22) sync-index
+    23) openspec-sync-specs [IF syncing specs without archive]
+    24) openspec-archive-change OR openspec-bulk-archive-change
+    25) changelog-generator
 ```
 
 **Conditional Skill Notes:**
@@ -258,11 +339,12 @@ When working on an **existing mobile codebase** (feature, bugfix, refactor):
 
   PHASE 5 - VALIDATION AND CLOSURE
     16) openspec-verify-change
-    17) documentation
-    18) sync-index
-    19) openspec-sync-specs [if needed]
-    20) openspec-archive-change / openspec-bulk-archive-change
-    21) changelog-generator
+    17) vibe-security (final security audit)
+    18) documentation
+    19) sync-index
+    20) openspec-sync-specs [if needed]
+    21) openspec-archive-change / openspec-bulk-archive-change
+    22) changelog-generator
 ```
 
 ---