Agent Skills by Qifan Wen
Showing 51 vetted skills indexed across 2 GitHub repositories.
ml-project-migration
Categorize ML project files for Git versus cloud storage migration.
data-validation
Validate ML and RL pipeline data with four-layer schema, drift, label, and environment checks.
storage-audit
Scan disk usage and propose confirmed cleanup actions for storage.
pseudo-code-algorithm
Generate publication-quality LaTeX or Markdown pseudo-code for ML, RL, and EA algorithms.
project-preference
Consolidate project documentation conventions into a portable setup blueprint.
latex-doc-audit
Diagnose and fix LaTeX layout issues with a phase-based workflow.
drawio
Generate draw.io diagrams with embedded XML from prompts.
doc-db-query
Generate SQLite queries for Doc DB plugin documentation filtering.
academic-poster
Generate academic posters from project results using the tikzposter LaTeX class.
project-audit
Audit project directories for clutter and generate categorized cleanup plans.
doc-db-guide
Guide developers in modifying and debugging the Doc Database VSCode extension.
babysit-training
Monitor ML training jobs and auto-restart from the last checkpoint on crash.
ml-workflow
Standardize ML experiment workflows with dataclass configs and W&B tracking.
computation-graph
Generate interactive HTML computation graphs with draggable nodes and SVG edges.
pkm-guide
Guide development of the Doc Database VSCode extension for commands, UI, and schema changes.
latex
Generate LaTeX documents for ACL conference papers and academic reports.
ml-config-system
Define hierarchical ML experiment configurations with logging, checkpointing, and MLflow tracking.
internal-comms
Generate internal company communications from predefined markdown templates.
pkm-query
Generate SQLite queries to search and filter Doc Database documentation entries.
scientific-slides
Automates scientific presentation slide creation for PDF and PowerPoint formats.
todoist-tasks
Create, find, update, and complete Todoist tasks via the MCP server.
tmux-gsd
Manages persistent Tmux sessions for GSD phase executions, data collection, and training runs.
mlflow-monitor
Launch the MLflow UI to monitor and compare training experiment metrics.
init-project
Initialize a new project with standard directories and boilerplate files.