AICHE +Gitlab Integration

The short answer: open GitLab, click into a merge request description or issue, press ⌃+⌥+R (Mac) or Ctrl+Alt+R (Windows/Linux), speak your complete context for 90-120 seconds, and AICHE inserts comprehensive documentation.

GitLab merge requests and CI/CD pipeline documentation require detailed change descriptions and configuration explanations that take 15+ minutes to type. Voice captures DevOps context faster than keyboard mechanics.

1. Open GitLab in your browser.
2. Navigate to create merge request or edit issue description.
3. Click into the description field.
4. Press your AICHE hotkey to start recording.
5. Speak your complete specification.
6. Press the hotkey again-AICHE transcribes and inserts the text.
7. Format with Markdown (headers, lists, code blocks) using GitLab's editor.

Merge Request Documentation

Comprehensive MR Descriptions

When creating merge requests, dictate complete context. Example: "Create merge request description for implementing caching layer. Summary: adds Redis caching to reduce database load for user profile and product catalog endpoints. Changes include: new CacheService class with get, set, and invalidate methods using Redis client, 5 minute TTL for user profiles, 15 minute TTL for product data. Updated UserController to check cache before database query, falls back to database on cache miss and populates cache for subsequent requests. Added cache invalidation on user profile updates and product modifications. Includes Docker compose configuration for local Redis instance on port 6379. Updated CI pipeline to start Redis service for integration tests. Testing: added unit tests for CacheService with mock Redis client, integration tests verify cache hit and miss scenarios, load tests show 60 percent reduction in database queries and 200 millisecond improvement in p95 latency. Breaking changes: none, feature is backward compatible with feature flag ENABLE_CACHE defaulting to false. Deployment notes: requires Redis deployment to production, environment variable REDIS_URL must be configured, run database migration to add cache_version column for invalidation tracking."

The complete MR description documents changes, testing, and deployment requirements comprehensively.

Code Review Responses

When responding to code review comments, dictate detailed explanations. Example: "Response to reviewer comment on error handling approach. Original suggestion to use global error handler acknowledged. However, global error handler doesn't work for this payment processing flow because we need transaction-specific rollback logic and payment provider reconciliation. Current implementation catches PaymentException at controller level, rolls back database transaction using @Transactional annotation, calls payment provider refund API, logs detailed error with correlation ID for support team debugging, and returns user-friendly error message without exposing internal details. Global handler would miss payment-specific cleanup and expose users to inconsistent state where payment succeeds but order creation fails. Alternative considered: custom exception hierarchy with PaymentException subclasses for different failure modes, but complexity outweighs benefits given single payment provider currently. Will revisit if we add multiple payment processors. Updated commit with additional comment explaining rationale per review feedback."

Detailed code review responses document technical decisions for future reference.

CI/CD Pipeline Documentation

Pipeline Configuration Explanations

For GitLab CI configuration, dictate pipeline logic. Example: "Document CI/CD pipeline stages and configuration. Pipeline has five stages: build, test, security, deploy staging, and deploy production. Build stage runs npm ci for clean dependency install, compiles TypeScript using tsc with strict mode, bundles application using Webpack with production optimization, and builds Docker image tagged with commit SHA. Test stage runs unit tests with Jest requiring 80% coverage, integration tests with test database, end-to-end tests using Cypress against preview environment, and performance tests checking page load under 2 seconds. Security stage scans dependencies with npm audit failing on high severity vulnerabilities, runs static code analysis with SonarQube checking code smells and security hotspots, and scans Docker image with Trivy for OS vulnerabilities. Deploy staging runs automatically on main branch merge using kubectl apply with rolling update strategy and runs smoke tests. Deploy production requires manual approval, uses blue-green deployment for zero downtime, runs database migrations with automatic rollback on failure, and sends Slack notification. Pipeline uses caching for node_modules improving build time from 8 minutes to 3 minutes. Environment variables for API keys stored in GitLab CI/CD settings with protected and masked flags."

Comprehensive pipeline documentation helps team understand CI/CD workflow and troubleshoot failures.

Troubleshooting Pipeline Failures

When documenting pipeline issues, dictate resolution steps. Example: "Document resolution for intermittent integration test failures in CI pipeline. Problem: tests fail randomly with database connection timeout errors approximately 20% of runs. Investigation: checked GitLab Runner logs showing database container not ready when tests start, added healthcheck but insufficient wait time, tests begin before migrations complete. Root cause: Docker Compose depends_on only waits for container start not application readiness. Solution: added wait-for-it script checking database accepts connections on port 5432 before running migrations, increased healthcheck interval to 2 seconds with 30 second timeout, added retry logic to database connection with exponential backoff up to 60 seconds. Results: test failure rate reduced from 20% to under 1%, pipeline reliability improved, and build times unchanged. Related issue: Postgres startup slow on CI runners with limited CPU, considering switching to reusable database between pipeline runs using cleanup script."

Troubleshooting documentation prevents repeating investigation for recurring issues.

Issue Management

Detailed Issue Descriptions

When creating issues, dictate comprehensive requirements. Example: "Create issue for implementing two-factor authentication. Description: add optional 2FA for user accounts improving security for sensitive operations. Requirements: support TOTP authenticator apps like Google Authenticator and Authy generating 6 digit codes, provide QR code enrollment showing secret key and account identifier, require user to verify setup by entering valid code before enabling, add recovery codes generating 10 single-use backup codes for account recovery, enforce 2FA for admin users with mandatory enrollment within 7 days, add remember device option storing encrypted cookie for 30 days on trusted devices. Implementation: use speakeasy library for TOTP generation and validation with 30 second time window, store encrypted secret keys in database using AES-256, track failed verification attempts limiting to 5 tries with exponential backoff. Testing: unit tests for TOTP generation and validation, integration tests for enrollment and authentication flows, security tests verifying encrypted storage and timing attack resistance. User experience: show warning banner after login prompting 2FA setup, provide clear instructions with screenshots in help documentation, support SMS fallback for users unable to use authenticator apps. Acceptance criteria: enrollment completion time under 2 minutes, successful authentication rate above 98%, recovery code usage documented in audit logs."

Detailed issue descriptions provide implementation clarity preventing scope creep.

Result: merge request descriptions with deployment notes, CI/CD pipeline documentation explaining stage dependencies, and issue specifications with acceptance criteria that took 22 minutes to type now take 5 minutes to dictate, and DevOps workflows become more documented because speaking captures context typing often abbreviates.

Do this now: open GitLab, create a new merge request or issue, click into the description, press your hotkey, and spend 2 minutes dictating complete context for your most recent code change including what changed, why, testing performed, and deployment considerations.