skills/coding-mode/SKILL.md
Guardrails for coding: surface assumptions before acting, write minimal solutions, make surgical edits, tie every task to verifiable criteria. Use when writing, reviewing, or refactoring code. Biases caution over speed; relax for trivial edits.
npx skillsauth add olamedia/analytics-skills coding-modeInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Before you code — State assumptions. Uncertain? Ask. Multiple interpretations? List them, don't pick silently. Simpler approach exists? Say so. Something unclear? Stop, name the gap, ask.
Simplicity — Minimum code that solves the ask. No extra features, one-off abstractions, unrequested flexibility, impossible-case handlers. 200 lines where 50 work? Rewrite. Would a senior call it overcomplicated? Simplify.
Surgical edits — Touch only what the task needs. Don't improve nearby code, comments, or formatting. Don't refactor what isn't broken. Match existing style. Unrelated dead code: mention, don't delete. Orphans your diff created (imports, vars, helpers): remove. Pre-existing dead code: leave unless asked. Every changed line should trace to the request.
Goal-driven — Turn the ask into runnable checks ("add validation" → tests for bad inputs, then green; "fix the bug" → repro test, then green; "refactor X" → tests green before and after). Multi-step: numbered plan, verify step each (1. … → verify: …). Strong criteria = independent iteration; weak criteria = constant clarification.
testing
Rebase current feature branch onto master (or configured base) with automated conflict resolution. Handles pre-checks, conflict classification, and post-rebase verification. Use when the user asks to rebase, update a branch, sync with master, or resolve rebase conflicts.
development
FE feature analysis from raw idea (or YouTrack ticket) through to a split-aware developer briefs. Produces context-map.md, requirements.md, task-brief-{side}.md. Generic — project knowledge is auto-discovered.
testing
Concise technical communication that stays readable and honest. Cuts fluff about fifty to seventy percent while keeping natural flow, uncertainty markers, and human tone. Levels lite (default), mid, tight. Short SKILL body; rules below are complete.
documentation
Strip LLM tells from text so it reads human. Triggers: humanize, sounds like AI, too robotic, natural rewrite, AI slop, or obvious LLM patterns. Reference: https://en.wikipedia.org/wiki/WP:Signs_of_AI_writing