// Skill family for agent-assisted software development workflows. Includes workspace isolation (git-worktrees), parallel debugging (dispatching-parallel-agents), structured multi-task development (subagent-driven-development), and lightweight code review (requesting-code-review). Install individual skills based on workflow needs.
Mechanical writing conventions for formal standards documents across all SDOs (3GPP, ETSI, ITU-T). Use when drafting or revising text to ensure consistent spelling (-ize/-ise), hyphenation, number formatting, capitalization, abbreviations, list formatting, proper names, and non-discriminatory language. Complements sdo-writing-style (prose quality) with mechanical correctness rules. Sources: ITU English Language Style Guide (2024).
Cross-platform operational guardrails and deterministic helper scripts for officecli OOXML edits on .docx files. Use when editing .docx with officecli raw-set, fixing tab run issues, inserting bookmarks or field codes (SEQ/REF), validating XPath targeting, or preventing destructive raw-set mistakes. Applies across all SDOs -- concrete style names provided by the per-SDO drafting skill.
Comprehensive Python development skill covering coding standards, CLI development, linting, testing, debugging, refactoring, code review, auditing, documentation, project planning, and bulk operations. Use when writing, reviewing, refactoring, debugging, or documenting Python code; configuring linters; setting up CLI tools; planning features; performing code audits; checking Python antipatterns, forbidden methods, or bad style; or handling bulk operations (10+ files) that benefit from batch workflows instead of per-file iteration.
Parse, convert, chunk, and analyze documents using Docling. Use this skill when the user provides a document (PDF, DOCX, PPTX, HTML, image) as a file path or URL and wants to: extract text or structured content, convert to Markdown or JSON, chunk the document for RAG ingestion, analyze document structure (headings, tables, figures, reading order), or run quality evaluation with iterative pipeline tuning. Triggers: "parse this PDF", "convert to markdown", "chunk for RAG", "extract tables", "analyze document structure", "prepare for ingestion", "process document", "evaluate docling output", "improve conversion quality".
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.
Insert SVG images into Word .docx files. Use when the user or coding agent needs to add, embed, or update an SVG graphic inside a .docx document. Triggers include "insert SVG into docx", "embed SVG in Word document", "add vector graphic to docx", "SVG image in template", "requests to convert SVG to PNG for docx", or any request explicitly mentioning embedding or inserting SVG graphics into .docx files.
| name | agentic-coding |
| description | Skill family for agent-assisted software development workflows. Includes workspace isolation (git-worktrees), parallel debugging (dispatching-parallel-agents), structured multi-task development (subagent-driven-development), and lightweight code review (requesting-code-review). Install individual skills based on workflow needs. |
| license | MIT |
A modular skill family for agent-assisted software development workflows. Each skill addresses a distinct phase or pattern—install only what you need.
| Skill | Purpose | When to Use |
|---|---|---|
| using-git-worktrees | Workspace isolation | Starting feature work that needs clean branch environment |
| dispatching-parallel-agents | Parallel debugging | 2+ completely independent failures across different domains |
| subagent-driven-development | Structured workflow | Multi-task implementation with integrated two-stage review |
| requesting-code-review | Standalone review | Ad-hoc reviews outside structured workflows |
using-git-worktrees → subagent-driven-development
↓ ↓
Create isolated Execute plan with
feature branch per-task subagents
+ integrated review
dispatching-parallel-agents
↓
Multiple subagents
investigate unrelated
failures simultaneously
requesting-code-review
↓
Lightweight review
after single task
or before merge
Install individual skills based on your workflow:
# Need workspace isolation?
saddle-cli install agentic-coding/using-git-worktrees
# Need parallel debugging?
saddle-cli install agentic-coding/dispatching-parallel-agents
# Need structured development workflow?
saddle-cli install agentic-coding/subagent-driven-development
# Need standalone code review?
saddle-cli install agentic-coding/requesting-code-review
digraph which_skill {
"Starting feature work?" [shape=diamond];
"Need clean workspace?" [shape=diamond];
"using-git-worktrees" [shape=box];
"Multiple failures?" [shape=diamond];
"Independent domains?" [shape=diamond];
"dispatching-parallel-agents" [shape=box];
"Single failure path" [shape=box];
"Have implementation plan?" [shape=diamond];
"Multi-task plan?" [shape=diamond];
"subagent-driven-development" [shape=box];
"requesting-code-review" [shape=box];
"Starting feature work?" -> "Need clean workspace?" [label="yes"];
"Starting feature work?" -> "Have implementation plan?" [label="no"];
"Need clean workspace?" -> "using-git-worktrees" [label="yes"];
"Need clean workspace?" -> "Have implementation plan?" [label="no"];
"Have implementation plan?" -> "Multi-task plan?" [label="yes"];
"Have implementation plan?" -> "Multiple failures?" [label="no"];
"Multi-task plan?" -> "subagent-driven-development" [label="yes"];
"Multi-task plan?" -> "requesting-code-review" [label="no"];
"Multiple failures?" -> "Independent domains?" [label="yes"];
"Multiple failures?" -> "Single failure path" [label="no"];
"Independent domains?" -> "dispatching-parallel-agents" [label="yes"];
"Independent domains?" -> "Single failure path" [label="no"];
}