coding-mode
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.
More from olamedia/analytics-skills
analyze-project
Use when starting work on any project to produce or update living documentation (TechStack.md, ProjectStructure.md) that bootstraps context for any AI agent session. Run before any feature work, or periodically to keep docs current.
13humanizer
>-
12architect
>-
12goal-definition
Use when you have a raw idea or request and need to define a clear goal with success criteria before exploring solutions. Use when requirements are vague, when "what does done look like" is unclear, or when assumptions need surfacing.
11prd
Use when you have a chosen direction and need to formalize requirements into a Product Requirements Document. Use when user stories, acceptance criteria, and scope boundaries need to be written down before architecture or implementation.
10analyze
Use when you have a raw idea or request and want to run the full analytics pipeline automatically — from research through to an interlinked task list. Best for straightforward problems where the full pipeline can flow with minimal back-and-forth.
10