smith6jt-cop
- 171 skills
- 0 followers
- 3 repo stars
- 2 weeks ago last updated
- ▌ Integrated Risk Manager · smith6jt-copUse when implementing unified position sizing, Kelly criterion, or combining GARCH+Capital+Portfolio risk systems
- ▌ Multi Agent Integration · smith6jt-copIntegrate Claude agents into training and live trading (v3.0). Trigger when: (1) setting up multi-agent training, (2) adding agent consultation to live trading, (3) configuring orchestrator, (4) understanding agent roles and safety mechanisms.
- ▌ Persistence Layer Audit · smith6jt-copAudit SQLite persistence layer for unused tables and broken integrations. Trigger when: (1) checking database usage, (2) cleaning up schema, (3) finding missing methods.
- ▌ Position Reconciliation · smith6jt-copBroker position reconciliation pattern for live trading systems
- ▌ Training Archive Gating · smith6jt-copMandatory training archive with model gating (APPROVED/REVIEW/DROP). Trigger when: (1) training run completes, (2) need to decide which models to deploy, (3) want historical training reference, (4) need checkpoint recommendations for overfitting.
- ▌
- ▌ Channel Tif To Ome Tiff · smith6jt-copConvert single-channel TIF directories to pyramidal OME-TIFF with SubIFDs
- ▌ Cleanup Qc Review Guard · smith6jt-copGuard pattern for cleanup scripts: verify QC sentinel files and prompt for human review before destructive operations
- ▌ Cycle Dir Normalization · smith6jt-copNormalize long-form CODEX cycle folders to short form before notebooks run. Trigger: cyc001_reg001_*, hard-coded cyc paths breaking, staged CODEX raw data failing in Notebooks 1/2.
- ▌ Gpu Parallel Scheduling · smith6jt-copGPU-safe parallel processing patterns for KINTSUGI to prevent OOM crashes and ensure Jupyter-compatible progress output
- ▌ Old Code Data Detection · smith6jt-copDetect processed data produced with old/buggy KINTSUGI code via timestamp forensics
- ▌ Snakemake Skip Existing · smith6jt-copPer-channel skip-existing checks in Snakemake wrapper scripts for resuming interrupted SLURM jobs. Trigger: Snakemake interrupted job, skip existing channels, resume incomplete cycle, sentinel missing but outputs exist, per-channel completeness check, wrapper script resume logic.
- ▌ Shiny Feature Patterns · smith6jt-copPatterns for adding interactive features to modular Shiny apps with H5AD data, conditional UI, embedded panels, tissue scatter, Leiden clustering
- ▌ Github Copilot Pr Review · smith6jt-copReview and apply GitHub Copilot suggestions on pull requests. Trigger when reviewing PRs with Copilot suggestions.
- ▌ Hpc Dev Testing Workflow · smith6jt-copEfficient development testing workflow for HPC environments with VS Code tunneling. Trigger: testing code changes on HPC, development workflow with external data directories, VS Code Remote SSH development
- ▌ Broker Order Limitations · smith6jt-copAlpaca broker limitations: crypto shorts blocked (broker doesn't support), stock shorts allowed. Trigger when: (1) shorting gate blocks wrong assets, (2) SELL signals blocked, (3) order type confusion.
- ▌ Dashboard Cohesion Fixes · smith6jt-copIntegration fixes between backtest engine, FastAPI dashboard, Vue.js frontend, and live trader
- ▌ Dashboard Parallel Lists · smith6jt-copDashboard symbol_signals uses parallel lists (symbols[], signal_values[], gate_statuses[]) not dict keyed by symbol. Trigger when: (1) 'list' object has no attribute 'get', (2) .items() on symbol_signals fails.
- ▌ Empirical Config Builder · smith6jt-copDerive selection thresholds from market data instead of hardcoding. Trigger when: (1) reviewing hardcoded parameters, (2) volume/price thresholds seem arbitrary, (3) selection returns too many/few candidates.
- ▌ MCP Server Configuration · smith6jt-copMCP (Model Context Protocol) server configuration for Claude Code integration with trading platform. Trigger when setting up MCP servers, fixing alpaca-mcp-server issues, or adding new integrations.
- ▌ Multi Timeframe Training · smith6jt-copDEPRECATED in v5.6.0 — see joint-multi-tf-v560 skill. Documents the v5.2.0 dual-model approach (train separate 15Min/1Hour models, combine via weighted voting). Still relevant for: (1) loading legacy v5.5.0 dual models, (2) understanding the historical aggregation layer, (3) resampling pattern via origin='start'.
- ▌ Risk Sizing Not Blocking · smith6jt-copRisk managers should SIZE trades, not BLOCK them. Trigger when: (1) capital manager blocks trades, (2) risk manager rejects orders, (3) trades fail due to 'exceeds limit', (4) components don't work in harmony.
- ▌ Basic Caching Evaluation · smith6jt-copEvaluation of BaSiC illumination correction caching - NOT RECOMMENDED for sparse markers. Trigger: optimizing BaSiC, caching illumination correction
- ▌
- ▌ MCP Absolute Path Config · smith6jt-copFix MCP server 'Executable not found' by resolving absolute paths in .mcp.json
- ▌ Multi Project Batch Isolation · smith6jt-copMulti-project signal isolation with cascading recipe resolution
- ▌
- ▌ Slurm Edf Channel Naming · smith6jt-copSLURM EDF job must use marker names from CHANNELNAMES.txt for output files. Trigger: EDF output naming, SLURM job producing CH#_edf.tif instead of marker names, downstream _find_edf_file failures
- ▌ Stale Script Remediation · smith6jt-copBulk-update stale workflow scripts across KINTSUGI projects. Trigger: scripts not propagating, QC failures, batch stalls after code updates.
- ▌ Valis Registration Codex · smith6jt-copVALIS registration for CODEX: rigid + non-rigid with tuned smoothing. Earlier 'rigid-only' conclusion was caused by a parameter passing bug — all non-rigid tests ran with unsmoothed OpticalFlowWarper.
- ▌ Project Data Separation · smith6jt-copPattern for separating repository code from user data and notebooks. Trigger: organizing scientific pipelines, separating code from data
- ▌ Cross Modal Normalization · smith6jt-copScale alignment for RNA-protein cross-modal integration - BOTH modalities must be z-scored
- ▌ Large Cell Ratio Matching · smith6jt-copMaxFuse parameter tuning for datasets with large protein:RNA cell ratios (>100:1)
- ▌ Differential Sharpe Ratio · smith6jt-copUse when implementing risk-adjusted rewards, discussing Sharpe ratio in RL training, or tuning reward components for risk awareness
- ▌ Gpu Aware Training Config · smith6jt-copGPU-aware PPO training configuration for A100/H100. Trigger when training is slow or GPU utilization is low.
- ▌ Macro Calendar Modulation · smith6jt-copCPI/GDP/retail sales date awareness and VIX regime scaling for calendar features
- ▌ Persistent Sector Caching · smith6jt-copCache yfinance sector data permanently to avoid Yahoo Finance rate limits. Trigger when: (1) 401 errors during sector fetching, (2) sector updates take too long, (3) need selective sector refresh.
- ▌ Reward Function Hold Bias · smith6jt-copFix HOLD bias in RL reward function. Trigger when: (1) model learns to always HOLD, (2) trade rate is too low (<10%), (3) slippage penalty exceeds typical price moves.
- ▌
- ▌ H5ad Shiny Data Pipeline · smith6jt-copPatterns for H5AD-backed Shiny apps with Excel fallback, groovy data in .uns, scVI QC validation, Leiden clustering merge, per-donor tissue extraction
- ▌
- ▌ Plugin Structure Migration · smith6jt-copMigrate improperly structured plugins to correct .claude-plugin/ layout
- ▌ Colab Notebook Development · smith6jt-copPattern for creating new Colab notebooks. Trigger when: (1) Creating a new notebook for experiments, (2) Adding notebook-based functionality, (3) Agent or validation notebooks, (4) Any notebook that uses GPU training infrastructure.
- ▌ Crypto Database Population · smith6jt-copPopulate symbol database with crypto symbols before training. Trigger when: (1) live trader fails with 'CRYPTO VIOLATION', (2) no crypto models in models/rl_symbols/, (3) db.get_candidates(asset_types=['crypto']) returns 0, (4) starting fresh training without crypto.
- ▌ Crypto Websocket Bars · smith6jt-copUse Alpaca CryptoDataStream websocket for real-time crypto bars in live trading (OHLC only, volume excluded)
- ▌
- ▌ Shiny LLM Chat · smith6jt-copPatterns for integrating OpenAI-compatible LLM chat into Shiny apps. Covers UF Navigator, reasoning models, streaming, fallback logic, and credential management.
- ▌
- ▌ Parameter Scaling · smith6jt-copMaxFuse parameter tuning when protein panel size changes (26 → 59+ markers)
- ▌
- ▌
- ▌ Pypi Collision Fix · smith6jt-copFix PyPI package name collisions when local package name exists on PyPI
- ▌ Skilled Agent V500 · smith6jt-copSkilled agent architecture replacing multi-agent system for RL training. Trigger when: (1) planning agent-guided training, (2) implementing tool-augmented LLM consultations, (3) comparing skilled vs multi-agent approaches, (4) designing simulate-verify loops for training, (5) implementing prompt evolution / learnable parameters, (6) understanding Claude Agent SDK integration in training, (7) debugging SkilledTrainer consultations or tool calls, (8) configuring agent safety bounds for training actions.
- ▌ Gpu Memory Cleanup · smith6jt-copPreventing Jupyter cell hangs with explicit CuPy GPU memory cleanup
- ▌ Qc Plot PDF Export · smith6jt-copKINTSUGI quantification plots: Always save QC heatmaps and profiles as PDF
- ▌
- ▌ Repo Reorganization · smith6jt-copPython package reorganization with pyproject.toml inside package directory
- ▌ Ccxt Multi Exchange · smith6jt-copCCXT multi-exchange crypto broker integration for better liquidity during after-hours capital shift
- ▌ Joint Multi Tf V560 · smith6jt-copv5.6.0 joint multi-TF model: single model per symbol with broadcast 1Hour context replaces dual 15Min/1Hour models. Trigger: (1) replacing weighted-voting model aggregation, (2) adding broadcast features to vectorized env, (3) limited training data + worried about overfitting from doubling obs_dim, (4) backtest builder mismatch with newer feature counts.
- ▌ Training Resilience · smith6jt-copFix PPO training early-stop issues. Trigger when: (1) impossible drawdown values (>100%), (2) training stops too early, (3) need adaptive recovery instead of hard stop.
- ▌ Gpu Only Scheduling · smith6jt-copGPU-only SLURM scheduling for KINTSUGI: never use CPU fallback — measured speedups: stitch 25x, decon 5x, EDF 15x (~13x full cycle)
- ▌ Windows Cupy Nvrtc · smith6jt-copFix CuPy NVRTC compilation errors on Windows. Trigger: NVRTC_ERROR_BUILTIN_OPERATION_FAILURE, nvrtc-builtins64 not found
- ▌ Backtest Notebook V4 · smith6jt-copFix 8-action space bug in backtest engines and rewrite backtest notebook for Colab v4.0.0
- ▌ Backtest Persistence · smith6jt-copSave backtest results to SQLite database for comparison. Trigger when: (1) tracking backtest history, (2) comparing model performance, (3) querying best backtests.
- ▌ Colab Unzip Workflow · smith6jt-copColab notebook setup pattern. Trigger when: (1) Creating new Colab notebooks, (2) API key file not found, (3) file path errors in Colab, (4) repository extraction fails, (5) 'yfinance fallback' despite keys existing.
- ▌ Data Source Priority · smith6jt-copCRITICAL: Alpaca API is MANDATORY for all OHLCV data. yfinance is NOT a valid fallback - EVER. Trigger when: (1) any code attempts yfinance for price/volume data, (2) crypto volume filter fails, (3) zero-volume bars detected, (4) API key configuration issues, (5) fallback behavior proposed.
- ▌ Dual Folder Workflow · smith6jt-copWorkflow for private repos + Colab training. Trigger when: (1) training on Colab with private repo, (2) separating dev from production, (3) avoiding untested code in production.
- ▌ Multi Tf Backtesting · smith6jt-copMulti-timeframe backtesting combining 15Min + 1Hour model signals. Trigger when: (1) multi-TF backtest, (2) combining timeframe signals in backtest, (3) validating multi-TF strategy, (4) --multi-tf CLI flag.
- ▌ Reward Function V330 · smith6jt-copUse when tuning RL reward function weights, fixing reward hacking, or addressing P&L gradient issues
- ▌ Reward Function V410 · smith6jt-copv4.1.0 reward function redesign to fix overtrading and DSR dominance
- ▌ Channel Name Parsing · smith6jt-copMulti-format channel name parsing for KINTSUGI CHANNELNAMES.txt files