@W-21461342 Configurable safety levels

Author: charithaT07

docs/cli/index.md

@@ -27,6 +27,32 @@ These flags are available on all commands that interact with B2C instances:
 | `--username`, `-u` | `SFCC_USERNAME`      | Username for Basic Auth            |
 | `--password`, `-p` | `SFCC_PASSWORD`      | Password/access key for Basic Auth |
 
+### Safety Mode
+
+Safety Mode provides protection against accidental or unwanted destructive operations. This is particularly important when using the CLI in automated environments, CI/CD pipelines, or as a tool for AI agents.
+
+| Environment Variable   | Values | Description |
+| ---------------------- | ------ | ----------- |
+| `SFCC_SAFETY_LEVEL` | `NONE` (default) | No restrictions |
+| | `NO_DELETE` | Block DELETE operations |
+| | `NO_UPDATE` | Block DELETE and destructive operations (reset, stop, restart) |
+| | `READ_ONLY` | Block all write operations (GET only) |
+
+**Example:**
+```bash
+# Prevent deletions in CI/CD
+export SFCC_SAFETY_LEVEL=NO_DELETE
+b2c sandbox create --realm test  # ✅ Allowed
+b2c sandbox delete test-id       # ❌ Blocked
+
+# Read-only mode for reporting
+export SFCC_SAFETY_LEVEL=READ_ONLY
+b2c sandbox list                 # ✅ Allowed
+b2c sandbox create --realm test  # ❌ Blocked
+```
+
+Safety Mode operates at the HTTP layer and cannot be bypassed by command-line flags. See the [Security Guide](/guide/security#operational-security-safety-mode) for detailed information.
+
 ### Other Environment Variables
 
 | Environment Variable         | Description                             |

docs/guide/ci-cd.md

@@ -438,3 +438,42 @@ Logs are always human-readable on stderr. The `--json` flag only controls the st
 All actions automatically configure:
 
 - **`NO_COLOR=1`** — clean log output without ANSI colors
+
+## Best Practices
+
+### Use Safety Mode
+
+Enable [Safety Mode](/guide/security#operational-security-safety-mode) in CI/CD to prevent accidental destructive operations:
+
+```yaml
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+    env:
+      # Prevent accidental deletions
+      SFCC_SAFETY_LEVEL:NO_DELETE
+    steps:
+      - uses: SalesforceCommerceCloud/b2c-developer-tooling/actions/setup@v1
+      - name: Deploy code
+        run: b2c code deploy
+```
+
+**Safety Levels for CI/CD:**
+
+- **`NO_DELETE`** (Recommended for most CI/CD) - Prevents deletions but allows deployments and updates
+- **`NO_UPDATE`** (Strict) - Only allows read and create operations, blocks updates and deletions
+- **`READ_ONLY`** (Monitoring/Reporting) - Only allows read operations
+
+**Example: Production deployment with safety:**
+```yaml
+deploy-production:
+  environment: production
+  env:
+    SFCC_SAFETY_LEVEL: NO_DELETE
+  steps:
+    - name: Deploy
+      run: |
+        b2c code deploy           # ✅ Allowed
+        b2c sandbox start prod    # ✅ Allowed
+        b2c sandbox delete test   # ❌ Blocked by safety mode
+```

docs/guide/configuration.md

@@ -82,6 +82,7 @@ You can configure the CLI using environment variables:
 | `MRT_PROJECT`                 | MRT project slug (`SFCC_MRT_PROJECT` also supported)           |
 | `MRT_ENVIRONMENT`             | MRT environment name (`SFCC_MRT_ENVIRONMENT`, `MRT_TARGET` also supported) |
 | `MRT_CLOUD_ORIGIN`            | MRT API origin URL override (`SFCC_MRT_CLOUD_ORIGIN` also supported) |
+| `SFCC_SAFETY_LEVEL`           | Safety mode: `NONE`, `NO_DELETE`, `NO_UPDATE`, `READ_ONLY` (see [Safety Mode](/guide/security#operational-security-safety-mode)) |
 
 ## .env File
 

docs/guide/security.md

@@ -64,6 +64,167 @@ When adding a new dependency that requires build scripts:
 
 This project uses [NPM trusted publishers](https://docs.npmjs.com/trusted-publishers) for package publication. Instead of storing long-lived npm tokens, packages are published via GitHub Actions using short-lived OIDC tokens that cannot be extracted or reused.
 
+## Operational Security: Safety Mode
+
+The CLI includes a **Safety Mode** feature that prevents accidental or unwanted destructive operations. This is particularly important when:
+
+- Using the CLI in automated environments (CI/CD pipelines)
+- Providing the CLI as a tool to AI agents/LLMs
+- Working in production environments
+- Training new team members
+- Running commands from untrusted scripts
+
+### How It Works
+
+Safety Mode uses a **hybrid protection approach**:
+
+1. **HTTP Middleware Layer** (Primary Protection)
+   - Intercepts ALL HTTP requests before they're sent
+   - Cannot be bypassed by command-line flags
+   - Works automatically for all commands
+   - LLM-proof: controlled via environment variable
+
+2. **Command-Level Checks** (Better UX)
+   - Provides early, user-friendly error messages
+   - Catches operations before HTTP requests
+
+### Safety Levels
+
+Configure via the `SFCC_SAFETY_LEVEL` environment variable:
+
+| Level | Description | Blocks |
+|-------|-------------|--------|
+| `NONE` | No restrictions (default) | Nothing |
+| `NO_DELETE` | Prevent deletions | DELETE operations |
+| `NO_UPDATE` | Prevent deletions and destructive updates | DELETE + reset/stop/restart |
+| `READ_ONLY` | Read-only mode | All writes (POST/PUT/PATCH/DELETE) |
+
+### Usage Examples
+
+#### Development (Allow Everything)
+```bash
+# No restrictions (default)
+unset SFCC_SAFETY_LEVEL
+# OR
+export SFCC_SAFETY_LEVEL=NONE
+
+b2c sandbox delete test-id  # ✅ Allowed
+```
+
+#### CI/CD (Prevent Deletions)
+```bash
+# Prevent accidental deletions in automated environments
+export SFCC_SAFETY_LEVEL=NO_DELETE
+
+b2c sandbox create --realm test  # ✅ Allowed
+b2c sandbox delete test-id       # ❌ Blocked
+```
+
+#### LLM/Agent Tools (Maximum Protection)
+```bash
+# Prevent AI agents from performing destructive operations
+export SFCC_SAFETY_LEVEL=NO_UPDATE
+
+b2c sandbox list                 # ✅ Allowed
+b2c sandbox create --realm test  # ✅ Allowed
+b2c sandbox delete test-id       # ❌ Blocked
+b2c sandbox reset test-id        # ❌ Blocked
+```
+
+#### Monitoring/Reporting (Read-Only)
+```bash
+# Absolute read-only mode
+export SFCC_SAFETY_LEVEL=READ_ONLY
+
+b2c sandbox list           # ✅ Allowed
+b2c sandbox get test-id    # ✅ Allowed
+b2c sandbox create test    # ❌ Blocked
+```
+
+### What Gets Blocked
+
+| Safety Level | DELETE | POST (create) | POST (reset) | PUT/PATCH | GET |
+|--------------|--------|---------------|--------------|-----------|-----|
+| **NONE** | ✅ | ✅ | ✅ | ✅ | ✅ |
+| **NO_DELETE** | ❌ | ✅ | ✅ | ✅ | ✅ |
+| **NO_UPDATE** | ❌ | ✅ | ❌ | ✅ | ✅ |
+| **READ_ONLY** | ❌ | ❌ | ❌ | ❌ | ✅ |
+
+### Protected Commands
+
+Safety Mode automatically protects ALL destructive commands across all topics:
+
+- **Sandbox**: `delete`, `reset`, `alias delete`
+- **Account Manager**: `users delete`, `users reset`, `clients delete`
+- **Code**: `delete`
+- **MRT**: `project delete`, `env delete`, `env var delete`, `env redirect delete`, `project notification delete`
+- **SLAS**: `client delete`
+- **eCDN**: All delete operations (certificates, zones, rules, etc.)
+
+### Why Environment Variable?
+
+Environment variables are more secure than command-line flags because:
+
+1. **LLMs Don't Control Them**: When an LLM uses the CLI as a tool, it controls commands and flags but NOT the environment
+2. **Session-Level**: Set once for the entire session/container
+3. **Audit Trail**: Can be logged and monitored in CI/CD
+4. **Cannot Be Bypassed**: Even `--force` flags don't override safety mode
+
+### Error Messages
+
+When an operation is blocked, you'll see clear error messages:
+
+```bash
+export SFCC_SAFETY_LEVEL=NO_DELETE
+b2c sandbox delete test-id
+
+# Error: Cannot delete sandbox: blocked by safety level NO_DELETE.
+#
+# Delete operations are blocked in NO_DELETE mode
+#
+# To allow this operation, unset or change the SFCC_SAFETY_LEVEL environment variable.
+```
+
+### Best Practices
+
+#### For CI/CD Pipelines
+```yaml
+# GitHub Actions example
+- name: Deploy to Production
+  env:
+    SFCC_SAFETY_LEVEL: NO_DELETE  # Prevent accidental deletions
+  run: |
+    b2c code deploy
+    b2c sandbox start production
+```
+
+#### For AI Agent Tools
+```bash
+# Provide CLI to LLMs with safety enabled
+export SFCC_SAFETY_LEVEL=NO_UPDATE
+# LLMs can now read and create, but cannot delete or reset
+```
+
+#### For Production Environments
+```bash
+# Set in shell profile for production access
+echo 'export SFCC_SAFETY_LEVEL=NO_UPDATE' >> ~/.bashrc
+```
+
+### Testing Safety Mode
+
+Verify safety mode is working:
+
+```bash
+export SFCC_SAFETY_LEVEL=NO_DELETE
+b2c sandbox delete fake-id
+
+# Expected: "blocked by safety level NO_DELETE"
+# NOT expected: Authentication error or API call
+```
+
+For comprehensive testing, see [GitHub Issue #67](https://github.com/SalesforceCommerceCloud/b2c-developer-tooling/issues/67).
+
 ## Best Practices
 
 ### For Contributors
@@ -78,3 +239,4 @@ This project uses [NPM trusted publishers](https://docs.npmjs.com/trusted-publis
 - Keep the CLI updated to receive security patches
 - Review the `pnpm-workspace.yaml` settings if you fork or modify this project
 - Consider using similar protections in your own projects
+- **Use Safety Mode** when running CLI in automated environments or providing it as a tool to AI agents

packages/b2c-cli/src/commands/am/clients/delete.ts

@@ -27,6 +27,9 @@ export default class ClientDelete extends AmCommand<typeof ClientDelete> {
   static examples = ['<%= config.bin %> <%= command.id %> <api-client-id>'];
 
   async run(): Promise<void> {
+    // Prevent deletion in safe mode
+    this.assertDestructiveOperationAllowed('delete API client');
+
     const apiClientId = this.args['api-client-id'];
 
     this.log(t('commands.client.delete.deleting', 'Deleting API client {{id}}...', {id: apiClientId}));

packages/b2c-cli/src/commands/am/users/delete.ts

@@ -35,6 +35,9 @@ export default class UserDelete extends AmCommand<typeof UserDelete> {
   };
 
   async run(): Promise<void> {
+    // Prevent deletion in safe mode
+    this.assertDestructiveOperationAllowed('delete user');
+
     const {login} = this.args;
     const {purge} = this.flags;
 

packages/b2c-cli/src/commands/am/users/reset.ts

@@ -28,6 +28,9 @@ export default class UserReset extends AmCommand<typeof UserReset> {
   ];
 
   async run(): Promise<void> {
+    // Prevent password reset in safe mode
+    this.assertDestructiveOperationAllowed('reset user password');
+
     const {login} = this.args;
 
     this.log(t('commands.user.reset.fetching', 'Fetching user {{login}}...', {login}));

packages/b2c-cli/src/commands/code/delete.ts

@@ -60,6 +60,9 @@ export default class CodeDelete extends InstanceCommand<typeof CodeDelete> {
   };
 
   async run(): Promise<void> {
+    // Prevent deletion in safe mode
+    this.assertDestructiveOperationAllowed('delete code version');
+
     this.requireOAuthCredentials();
 
     const codeVersion = this.args.codeVersion;

packages/b2c-cli/src/commands/ecdn/certificates/delete.ts

@@ -49,6 +49,9 @@ export default class EcdnCertificatesDelete extends EcdnZoneCommand<typeof EcdnC
   };
 
   async run(): Promise<DeleteOutput> {
+    // Prevent deletion in safe mode
+    this.assertDestructiveOperationAllowed('delete certificate');
+
     this.requireOAuthCredentials();
 
     const zoneId = await this.resolveZoneId();

packages/b2c-cli/src/commands/ecdn/logpush/jobs/delete.ts

@@ -37,6 +37,9 @@ export default class EcdnLogpushJobsDelete extends EcdnZoneCommand<typeof EcdnLo
   };
 
   async run(): Promise<DeleteOutput> {
+    // Prevent deletion in safe mode
+    this.assertDestructiveOperationAllowed('delete Logpush job');
+
     this.requireOAuthCredentials();
 
     const zoneId = await this.resolveZoneId();