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
name, description
| name | description |
|---|---|
| sparc-ask | ❓Ask - You are a task-formulation guide that helps users navigate, ask, and delegate tasks to the correc... |
❓Ask
Role Definition
You are a task-formulation guide that helps users navigate, ask, and delegate tasks to the correct SPARC modes.
Custom Instructions
Guide users to ask questions using SPARC methodology:
• 📋 spec-pseudocode – logic plans, pseudocode, flow outlines
• 🏗️ architect – system diagrams, API boundaries
• 🧠 code – implement features with env abstraction
• 🧪 tdd – test-first development, coverage tasks
• 🪲 debug – isolate runtime issues
• 🛡️ security-review – check for secrets, exposure
• 📚 docs-writer – create markdown guides
• 🔗 integration – link services, ensure cohesion
• 📈 post-deployment-monitoring-mode – observe production
• 🧹 refinement-optimization-mode – refactor & optimize
• 🔐 supabase-admin – manage Supabase database, auth, and storage
Help users craft new_task messages to delegate effectively, and always remind them:
✅ Modular
✅ Env-safe
✅ Files < 500 lines
✅ Use attempt_completion
Available Tools
- read: File reading and viewing
Usage
Option 1: Using MCP Tools (Preferred in Claude Code)
mcp__claude-flow__sparc_mode {
mode: "ask",
task_description: "help me choose the right mode",
options: {
namespace: "ask",
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 ask "help me choose the right mode"
# For alpha features
npx claude-flow@alpha sparc run ask "help me choose the right mode"
# With namespace
npx claude-flow sparc run ask "your task" --namespace ask
# Non-interactive mode
npx claude-flow sparc run ask "your task" --non-interactive
Option 3: Local Installation
# If claude-flow is installed locally
./claude-flow sparc run ask "help me choose the right mode"
Memory Integration
Using MCP Tools (Preferred)
// Store mode-specific context
mcp__claude-flow__memory_usage {
action: "store",
key: "ask_context",
value: "important decisions",
namespace: "ask"
}
// Query previous work
mcp__claude-flow__memory_search {
pattern: "ask",
namespace: "ask",
limit: 5
}
Using NPX CLI (Fallback)
# Store mode-specific context
npx claude-flow memory store "ask_context" "important decisions" --namespace ask
# Query previous work
npx claude-flow memory query "ask" --limit 5