gsd-trae 1.0.0 → 1.0.2

package/refs/vbenchmark/tasks/glue-code/file-processing/pdf-merger/PROMPT.md

@@ -1,50 +0,0 @@
-# Task: PDF Merger and Manipulator
-
-## Objective
-Build a PDF manipulation service that merges, splits, and transforms PDF files.
-
-## Requirements
-
-1. **Merge Operations**
-   - Merge multiple PDFs into one
-   - Specify page order
-   - Add blank pages between docs
-   - Set output metadata
-
-2. **Split Operations**
-   - Split by page ranges
-   - Extract specific pages
-   - Split into single pages
-   - Split by size limit
-
-3. **Transform Operations**
-   - Rotate pages
-   - Add watermark (text or image)
-   - Add page numbers
-   - Compress/optimize
-
-4. **API Endpoints**
-   - `POST /pdf/merge` - Merge PDFs
-   - `POST /pdf/split` - Split PDF
-   - `POST /pdf/rotate` - Rotate pages
-   - `POST /pdf/watermark` - Add watermark
-
-## Technical Stack
-- Python 3.11+
-- FastAPI
-- pypdf or PyMuPDF
-- Pillow for image watermarks
-
-## Files to Create
-- `app/main.py` - FastAPI app
-- `app/services/merger.py` - Merge logic
-- `app/services/splitter.py` - Split logic
-- `app/services/transformer.py` - Transform ops
-- `app/routers/pdf.py` - API endpoints
-
-## Success Criteria
-- [ ] Merge creates valid PDF
-- [ ] Split extracts correct pages
-- [ ] Rotation works correctly
-- [ ] Watermark appears on all pages
-- [ ] Output is valid PDF

package/refs/vbenchmark/tasks/glue-code/file-processing/pdf-merger/task.yaml

@@ -1,24 +0,0 @@
-name: PDF Merger and Manipulator
-category: glue-code
-subcategory: file-processing
-description: Merge and manipulate PDF files with Python
-difficulty: medium
-stack: python
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/test_pdf.py
-
-tags:
-  - python
-  - pdf
-  - merge
-  - manipulation
-  - pypdf
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/file-processing/video-transcoder/PROMPT.md

@@ -1,50 +0,0 @@
-# Task: Video Transcoder Service
-
-## Objective
-Build a video transcoding service using FFmpeg that converts videos to multiple formats and resolutions.
-
-## Requirements
-
-1. **Transcoding**
-   - Multiple output formats (MP4, WebM, HLS)
-   - Multiple resolutions (1080p, 720p, 480p)
-   - Codec selection (H.264, H.265, VP9)
-   - Audio transcoding (AAC, Opus)
-
-2. **HLS Output**
-   - Generate HLS playlist
-   - Segment videos for streaming
-   - Multiple quality variants
-   - Master playlist
-
-3. **Job Management**
-   - Queue transcoding jobs
-   - Progress tracking
-   - Cancel running jobs
-   - Webhook on completion
-
-4. **API Endpoints**
-   - `POST /transcode` - Start job
-   - `GET /jobs/{id}` - Get status
-   - `DELETE /jobs/{id}` - Cancel job
-   - `GET /jobs/{id}/progress` - Stream progress
-
-## Technical Stack
-- Go 1.21+
-- FFmpeg (exec wrapper)
-- Redis for job queue
-- S3 for input/output storage
-
-## Files to Create
-- `cmd/transcoder/main.go` - Entry point
-- `internal/ffmpeg/transcoder.go` - FFmpeg wrapper
-- `internal/ffmpeg/hls.go` - HLS generation
-- `internal/jobs/queue.go` - Job queue
-- `internal/handlers/transcode.go` - API handlers
-
-## Success Criteria
-- [ ] Videos transcode to MP4 correctly
-- [ ] HLS output plays in browser
-- [ ] Progress updates in real-time
-- [ ] Jobs can be cancelled
-- [ ] Multiple resolutions generated

package/refs/vbenchmark/tasks/glue-code/file-processing/video-transcoder/task.yaml

@@ -1,27 +0,0 @@
-name: Video Transcoder Service
-category: glue-code
-subcategory: file-processing
-description: FFmpeg-based video transcoding in Go
-difficulty: hard
-stack: go
-timeout: 900
-tokenLimit: 200000
-
-tests:
-  functional: tests/transcode_test.go
-
-docker:
-  compose: docker-compose.yaml
-
-tags:
-  - go
-  - ffmpeg
-  - video
-  - transcoding
-  - streaming
-
-evaluation:
-  functional: 0.5
-  quality: 0.25
-  cost: 0.1
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/migration/data-backfill/PROMPT.md

@@ -1,50 +0,0 @@
-# Task: Data Backfill Pipeline
-
-## Objective
-Build a data backfill system that migrates/transforms large datasets with progress tracking and resumability.
-
-## Requirements
-
-1. **Batch Processing**
-   - Process records in batches
-   - Configurable batch size
-   - Parallel batch processing
-   - Memory-efficient iteration
-
-2. **Progress Tracking**
-   - Track processed record count
-   - Estimate time remaining
-   - Save checkpoint periodically
-   - Resume from checkpoint
-
-3. **Error Handling**
-   - Log failed records
-   - Continue on individual failures
-   - Retry failed batches
-   - Generate error report
-
-4. **CLI Interface**
-   - `backfill run <job>` - Start backfill
-   - `backfill status <job>` - Check progress
-   - `backfill resume <job>` - Resume from checkpoint
-   - `backfill retry-failed <job>` - Retry failures
-
-## Technical Stack
-- Python 3.11+
-- SQLAlchemy for DB access
-- Click for CLI
-- Redis for progress state
-
-## Files to Create
-- `backfill/cli.py` - CLI commands
-- `backfill/runner.py` - Backfill executor
-- `backfill/checkpoint.py` - Checkpoint management
-- `backfill/jobs/user_migration.py` - Example job
-- `backfill/reporter.py` - Progress reporting
-
-## Success Criteria
-- [ ] Processes all records correctly
-- [ ] Progress persists across restarts
-- [ ] Resume continues from checkpoint
-- [ ] Failed records logged
-- [ ] Retry-failed processes failures only

package/refs/vbenchmark/tasks/glue-code/migration/data-backfill/task.yaml

@@ -1,24 +0,0 @@
-name: Data Backfill Pipeline
-category: glue-code
-subcategory: migration
-description: Implement data backfill with progress tracking in Python
-difficulty: medium
-stack: python
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/test_backfill.py
-
-tags:
-  - python
-  - backfill
-  - migration
-  - data
-  - progress
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/migration/database-versioning/PROMPT.md

@@ -1,50 +0,0 @@
-# Task: Database Migration System
-
-## Objective
-Build a database schema migration system that tracks, applies, and rolls back migrations.
-
-## Requirements
-
-1. **Migration Files**
-   - Timestamped migration files
-   - Up and down functions
-   - SQL or programmatic migrations
-   - Migration template generation
-
-2. **Operations**
-   - `migrate up` - Apply pending
-   - `migrate down` - Rollback last
-   - `migrate status` - Show status
-   - `migrate create <name>` - New migration
-
-3. **Tracking**
-   - Store applied migrations in DB
-   - Lock to prevent concurrent runs
-   - Checksum validation
-   - Execution timing
-
-4. **Safety**
-   - Transaction per migration
-   - Dry run mode
-   - Backup before destructive ops
-   - Validate before apply
-
-## Technical Stack
-- TypeScript/Node.js
-- PostgreSQL
-- Commander for CLI
-
-## Files to Create
-- `src/cli.ts` - CLI commands
-- `src/migrator.ts` - Migration runner
-- `src/tracker.ts` - Migration tracking
-- `src/generator.ts` - Migration generator
-- `src/lock.ts` - Concurrent lock
-- `migrations/` - Migration files
-
-## Success Criteria
-- [ ] Migrations apply in order
-- [ ] Rollback works correctly
-- [ ] Status shows correct state
-- [ ] Concurrent runs prevented
-- [ ] Transactions rollback on error

package/refs/vbenchmark/tasks/glue-code/migration/database-versioning/task.yaml

@@ -1,24 +0,0 @@
-name: Database Migration System
-category: glue-code
-subcategory: migration
-description: Build database schema migration system in TypeScript
-difficulty: medium
-stack: nodejs
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/migration.test.ts
-
-tags:
-  - typescript
-  - database
-  - migrations
-  - schema
-  - versioning
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/queue/kafka-producer/PROMPT.md

@@ -1,49 +0,0 @@
-# Task: High-Throughput Kafka Producer
-
-## Objective
-Build a high-throughput Kafka producer in Go with batching, compression, and delivery guarantees.
-
-## Requirements
-
-1. **Producer Configuration**
-   - Async batching for throughput
-   - Compression (snappy/lz4)
-   - Configurable acks (0, 1, all)
-   - Idempotent producer for exactly-once
-
-2. **Message Publishing**
-   - Publish to specific partition (key-based)
-   - Headers support
-   - Async with callbacks
-   - Sync option for critical messages
-
-3. **Performance**
-   - Batch by size and time
-   - Connection pooling
-   - Metrics (throughput, latency)
-   - Handle backpressure
-
-4. **API**
-   - `POST /publish` - Publish single message
-   - `POST /publish/batch` - Publish batch
-   - `GET /metrics` - Producer metrics
-   - Health check endpoint
-
-## Technical Stack
-- Go 1.21+
-- confluent-kafka-go or segmentio/kafka-go
-- Prometheus metrics
-
-## Files to Create
-- `cmd/producer/main.go` - Entry point
-- `internal/kafka/producer.go` - Producer wrapper
-- `internal/kafka/config.go` - Configuration
-- `internal/handlers/publish.go` - HTTP handlers
-- `internal/metrics/metrics.go` - Metrics
-
-## Success Criteria
-- [ ] Achieves 100K+ messages/sec
-- [ ] Batching works correctly
-- [ ] Compression reduces size
-- [ ] Delivery callbacks work
-- [ ] Metrics exposed correctly

package/refs/vbenchmark/tasks/glue-code/queue/kafka-producer/task.yaml

@@ -1,27 +0,0 @@
-name: High-Throughput Kafka Producer
-category: glue-code
-subcategory: queue
-description: Implement high-throughput Kafka producer in Go
-difficulty: medium
-stack: go
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/producer_test.go
-
-docker:
-  compose: docker-compose.yaml
-
-tags:
-  - go
-  - kafka
-  - streaming
-  - producer
-  - high-throughput
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/queue/rabbitmq-consumer/PROMPT.md

@@ -1,50 +0,0 @@
-# Task: RabbitMQ Message Consumer
-
-## Objective
-Build a reliable RabbitMQ consumer with proper acknowledgment, retry, and dead letter handling.
-
-## Requirements
-
-1. **Consumer Setup**
-   - Connect with automatic reconnection
-   - Prefetch for flow control
-   - Multiple consumer instances (scaling)
-   - Graceful shutdown
-
-2. **Message Processing**
-   - Process different message types
-   - Validate message schema
-   - Handle processing errors
-   - Manual acknowledgment
-
-3. **Retry Logic**
-   - Retry failed messages with backoff
-   - Max retry count per message
-   - Move to dead letter after max retries
-   - Track retry count in headers
-
-4. **Dead Letter Queue**
-   - Capture failed messages
-   - Include error information
-   - API to inspect dead letters
-   - Replay dead letters
-
-## Technical Stack
-- Python 3.11+
-- aio-pika (async RabbitMQ)
-- Pydantic for validation
-- PostgreSQL for state
-
-## Files to Create
-- `consumer/main.py` - Entry point
-- `consumer/handlers.py` - Message handlers
-- `consumer/retry.py` - Retry logic
-- `consumer/dlq.py` - Dead letter handling
-- `consumer/models.py` - Message models
-
-## Success Criteria
-- [ ] Messages processed reliably
-- [ ] Failed messages retried with backoff
-- [ ] Dead letters captured after max retries
-- [ ] Consumer reconnects on connection loss
-- [ ] Graceful shutdown completes in-flight

package/refs/vbenchmark/tasks/glue-code/queue/rabbitmq-consumer/task.yaml

@@ -1,27 +0,0 @@
-name: RabbitMQ Message Consumer
-category: glue-code
-subcategory: queue
-description: Implement reliable RabbitMQ consumer with Python
-difficulty: medium
-stack: python
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/test_consumer.py
-
-docker:
-  compose: docker-compose.yaml
-
-tags:
-  - python
-  - rabbitmq
-  - queue
-  - messaging
-  - async
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/queue/sqs-batch-processor/PROMPT.md

@@ -1,47 +0,0 @@
-# Task: AWS SQS Batch Processor
-
-## Objective
-Build an efficient SQS batch processor with partial failure handling and visibility timeout management.
-
-## Requirements
-
-1. **Batch Receiving**
-   - Receive up to 10 messages per request
-   - Long polling for efficiency
-   - Parse message body (JSON)
-   - Extract message attributes
-
-2. **Batch Processing**
-   - Process messages in parallel
-   - Individual message error handling
-   - Partial batch failure support
-   - Report failed message IDs
-
-3. **Visibility Timeout**
-   - Extend timeout for long-running jobs
-   - Heartbeat mechanism
-   - Release messages on failure
-
-4. **Dead Letter Queue**
-   - Configure DLQ on max receives
-   - Move manually to DLQ on permanent failure
-   - Redrive from DLQ
-
-## Technical Stack
-- TypeScript/Node.js
-- AWS SDK v3
-- Zod for validation
-
-## Files to Create
-- `src/processor.ts` - Main processor
-- `src/handlers/index.ts` - Message handlers
-- `src/visibility.ts` - Visibility management
-- `src/dlq.ts` - DLQ operations
-- `src/types.ts` - Type definitions
-
-## Success Criteria
-- [ ] Batch receiving works efficiently
-- [ ] Partial failures don't block batch
-- [ ] Visibility extended for long jobs
-- [ ] Failed messages go to DLQ
-- [ ] Redrive from DLQ works

package/refs/vbenchmark/tasks/glue-code/queue/sqs-batch-processor/task.yaml

@@ -1,24 +0,0 @@
-name: AWS SQS Batch Processor
-category: glue-code
-subcategory: queue
-description: Process SQS messages in batches with TypeScript
-difficulty: medium
-stack: nodejs
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/sqs.test.ts
-
-tags:
-  - typescript
-  - aws
-  - sqs
-  - batch
-  - serverless
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/scheduler/cron-job-manager/PROMPT.md

@@ -1,52 +0,0 @@
-# Task: Distributed Cron Job Manager
-
-## Objective
-Build a distributed cron scheduling system that runs jobs across multiple workers with leader election.
-
-## Requirements
-
-1. **Job Definition**
-   - Cron expression parsing
-   - HTTP webhook jobs
-   - Function jobs (local execution)
-   - Job timeout configuration
-
-2. **Distributed Execution**
-   - Leader election for scheduler
-   - Worker pool for job execution
-   - Prevent duplicate execution
-   - Handle worker failures
-
-3. **Job Management**
-   - `POST /jobs` - Create job
-   - `GET /jobs` - List jobs
-   - `PUT /jobs/{id}` - Update job
-   - `DELETE /jobs/{id}` - Delete job
-   - `POST /jobs/{id}/run` - Trigger manually
-
-4. **Monitoring**
-   - Job execution history
-   - Success/failure rates
-   - Execution duration
-   - Next run time
-
-## Technical Stack
-- Go 1.21+
-- Redis for locking/leader election
-- PostgreSQL for job storage
-- robfig/cron for parsing
-
-## Files to Create
-- `cmd/scheduler/main.go` - Scheduler process
-- `cmd/worker/main.go` - Worker process
-- `internal/scheduler/leader.go` - Leader election
-- `internal/scheduler/cron.go` - Cron parsing
-- `internal/worker/executor.go` - Job execution
-- `internal/handlers/jobs.go` - API handlers
-
-## Success Criteria
-- [ ] Cron expressions parsed correctly
-- [ ] Only leader schedules jobs
-- [ ] Jobs execute exactly once
-- [ ] Failed jobs tracked
-- [ ] Manual trigger works

package/refs/vbenchmark/tasks/glue-code/scheduler/cron-job-manager/task.yaml

@@ -1,27 +0,0 @@
-name: Distributed Cron Job Manager
-category: glue-code
-subcategory: scheduler
-description: Build distributed cron scheduling system in Go
-difficulty: hard
-stack: go
-timeout: 900
-tokenLimit: 200000
-
-tests:
-  functional: tests/scheduler_test.go
-
-docker:
-  compose: docker-compose.yaml
-
-tags:
-  - go
-  - cron
-  - scheduler
-  - distributed
-  - redis
-
-evaluation:
-  functional: 0.5
-  quality: 0.25
-  cost: 0.1
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/scheduler/delayed-tasks/PROMPT.md

@@ -1,51 +0,0 @@
-# Task: Celery Delayed Task Executor
-
-## Objective
-Build a delayed task execution system using Celery for scheduling tasks at specific times or after delays.
-
-## Requirements
-
-1. **Task Scheduling**
-   - Execute after delay (countdown)
-   - Execute at specific time (eta)
-   - Recurring tasks (periodic)
-   - Cancel scheduled tasks
-
-2. **Task Types**
-   - Email reminders
-   - Report generation
-   - Data cleanup
-   - Notification delivery
-
-3. **API Endpoints**
-   - `POST /tasks/schedule` - Schedule task
-   - `GET /tasks/{id}` - Get task status
-   - `DELETE /tasks/{id}` - Cancel task
-   - `GET /tasks/pending` - List pending
-
-4. **Features**
-   - Task revocation
-   - Result storage
-   - Task chains/groups
-   - Retry on failure
-
-## Technical Stack
-- Python 3.11+
-- Celery with Redis broker
-- FastAPI for API
-- Redis for result backend
-
-## Files to Create
-- `app/main.py` - FastAPI app
-- `app/tasks/email.py` - Email tasks
-- `app/tasks/reports.py` - Report tasks
-- `app/routers/tasks.py` - API endpoints
-- `app/celery_app.py` - Celery config
-- `celeryconfig.py` - Celery settings
-
-## Success Criteria
-- [ ] Tasks execute after countdown
-- [ ] Tasks execute at specific eta
-- [ ] Periodic tasks run on schedule
-- [ ] Cancelled tasks don't execute
-- [ ] Task status queryable

package/refs/vbenchmark/tasks/glue-code/scheduler/delayed-tasks/task.yaml

@@ -1,27 +0,0 @@
-name: Celery Delayed Task Executor
-category: glue-code
-subcategory: scheduler
-description: Implement delayed task execution with Celery in Python
-difficulty: medium
-stack: python
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/test_delayed_tasks.py
-
-docker:
-  compose: docker-compose.yaml
-
-tags:
-  - python
-  - celery
-  - delayed-tasks
-  - redis
-  - async
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15