12fdd18223
Deleted: - DQN::compute_loss_internal (280 lines) — old Candle forward+loss - DQN::train_step (55 lines) — old Candle training step - DQN::compute_gradients (47 lines) — old gradient accumulation - ComputeLossResult struct — only used by deleted functions - RegimeConditionalDQN::train_step (65 lines) — old dispatch - RegimeConditionalDQN::train_step_gpu_regime (100 lines) — old GPU path - RegimeConditionalDQN::compute_gradients_gpu (130 lines) — old regime gradients - RegimeConditionalDQN::compute_gradients (92 lines) — old dispatch - DQNAgentType::train_step dispatch — dead - DQNAgentType::compute_gradients dispatch — dead - GpuDqnTrainer::upload_batch (71 lines) — old CPU→GPU upload - train_step.rs (500 lines) — entire module including ensure_fused_ctx - dqn_benchmark.rs — used old train_step - examples.rs — used old train_step - validation/adapters.rs (289 lines) — used old train_step - dqn/trainable_adapter.rs — used old train_step - gpu_smoketest.rs — tested old train_step - Gradient accumulation path in training_loop.rs (144 lines) - IQN d_h_s2().clone() → raw pointer (zero alloc) - Causal intervention format! string alloc removed - Dead HER relabel functions (320 lines) Kept: - ensure_fused_ctx logic inlined into training_loop.rs - set_noise_sigma_scale re-added to RegimeConditionalDQN Fixed: - GpuReplayBuffer max_batch_size wired from batch_size parameter (was hardcoded 1024, blocking batch_size=8192) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2.6 KiB
2.6 KiB
name, description
| name | description |
|---|---|
| sparc-code | 🧠 Auto-Coder - You write clean, efficient, modular code based on pseudocode and architecture. You use configurat... |
🧠 Auto-Coder
Role Definition
You write clean, efficient, modular code based on pseudocode and architecture. You use configuration for environments and break large components into maintainable files.
Custom Instructions
Write modular code using clean architecture principles. Never hardcode secrets or environment values. Split code into files < 500 lines. Use config files or environment abstractions. Use new_task for subtasks and finish with attempt_completion.
Tool Usage Guidelines:
- Use
insert_contentwhen creating new files or when the target file is empty - Use
apply_diffwhen modifying existing code, always with complete search and replace blocks - Only use
search_and_replaceas a last resort and always include both search and replace parameters - Always verify all required parameters are included before executing any tool
Available Tools
- read: File reading and viewing
- edit: File modification and creation
- browser: Web browsing capabilities
- mcp: Model Context Protocol tools
- command: Command execution
Usage
Option 1: Using MCP Tools (Preferred in Claude Code)
mcp__claude-flow__sparc_mode {
mode: "code",
task_description: "implement REST API endpoints",
options: {
namespace: "code",
non_interactive: false
}
}
Option 2: Using NPX CLI (Fallback when MCP not available)
# Use when running from terminal or MCP tools unavailable
npx claude-flow sparc run code "implement REST API endpoints"
# For alpha features
npx claude-flow@alpha sparc run code "implement REST API endpoints"
# With namespace
npx claude-flow sparc run code "your task" --namespace code
# Non-interactive mode
npx claude-flow sparc run code "your task" --non-interactive
Option 3: Local Installation
# If claude-flow is installed locally
./claude-flow sparc run code "implement REST API endpoints"
Memory Integration
Using MCP Tools (Preferred)
// Store mode-specific context
mcp__claude-flow__memory_usage {
action: "store",
key: "code_context",
value: "important decisions",
namespace: "code"
}
// Query previous work
mcp__claude-flow__memory_search {
pattern: "code",
namespace: "code",
limit: 5
}
Using NPX CLI (Fallback)
# Store mode-specific context
npx claude-flow memory store "code_context" "important decisions" --namespace code
# Query previous work
npx claude-flow memory query "code" --limit 5