Browse Skills

Dashi PPT turns natural language requirements into offline-editable HTML slide decks using preset visual themes, and supports exporting to PPTX or PDF. Use it when making presentations, reports, or slide decks without hand-writing slide markup.

Agent Identifier helps you create and configure Claude Code agents, including YAML frontmatter, triggering descriptions, system prompts, model, color, and tool settings. Use it when setting up autonomous subagents, writing agent frontmatter, or designing agent behavior.

Architecture Design defines the standard code architecture for machine learning projects using Factory and Registry patterns. Use it when creating new registrable ML components such as datasets, models, or modules that require decorators and factory wiring.

Bug Detective provides a systematic debugging workflow for investigating and resolving code errors, exceptions, and failures. Use it when debugging, fixing errors, or troubleshooting broken code.

Citation Verification provides reference guidance for verifying academic citations during paper writing. Use it when checking citation accuracy, preventing fake citations, or integrating verification into an ML paper workflow.

Code Review Excellence provides principles, checklists, and techniques for reviewing diffs and pull requests, writing actionable feedback, and establishing team review standards. Use it when auditing code quality, mentoring developers, or improving review cycles.

Command Development guides creating Claude Code slash commands, covering Markdown structure, YAML frontmatter, dynamic arguments, file references, and bash execution. Use it when building reusable, interactive, or namespaced commands for agent workflows.

Daily Coding provides a quality assurance checklist for writing or modifying source code. Use it before, during, and after coding tasks to ensure minimal changes, type safety, security, and clean execution.

Daily Paper Generator discovers, screens, and summarizes recent arXiv and bioRxiv papers for any research topic, producing bilingual Chinese/English structured notes with a Top 10 to Top 3 to Top 1 selection workflow.

Defuddle extracts clean Markdown from web pages so agents can ingest them into a knowledge base, typically under Sources/Web/.

Doc Co-Authoring guides users through a structured three-stage workflow—context gathering, iterative refinement, and reader testing—to collaboratively produce documentation, proposals, technical specs, decision docs, or RFCs.

Expression structures responses with conclusion-first logic, concrete evidence, risk disclosure, and actionable next steps. Use it for task reports, file-operation summaries, research discussion, study-note synthesis, planning, and writing feedback.