From dbud
Baseline coding behavior. Use before writing, reviewing, debugging, or refactoring code. Establishes ground rules for assumptions, simplicity, surgical edits, and verification.
How this skill is triggered — by the user, by Claude, or both
Slash command
/dbud:coding-guidelinesThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
These are baseline rules for coding tasks. Apply them before writing, reviewing, debugging, refactoring, or testing code.
These are baseline rules for coding tasks. Apply them before writing, reviewing, debugging, refactoring, or testing code.
Bias toward small, correct, maintainable, verified changes.
For trivial tasks, do not add unnecessary process. For non-trivial tasks, define the goal, make a focused change, and verify it.
Do not assume silently.
State assumptions when they materially affect:
Ask for clarification only when the answer would meaningfully change the implementation. For low-risk ambiguity with a conventional default, state the assumption and proceed instead of blocking.
Push back when the requested approach appears overcomplicated, risky, or misaligned with the stated goal.
Implement the minimum code that correctly solves the requested problem.
Do not add:
Prefer existing project utilities and standard library functionality over new code or dependencies.
If the solution is much larger than the problem, simplify before proceeding.
Touch only what the request requires.
When editing existing code:
Mention unrelated issues separately instead of changing them.
Clean up only artifacts caused by your own change:
Every changed line should trace directly to the user's request.
Default to no comments. Add one only when the why is non-obvious — a hidden constraint, subtle invariant, or workaround.
Do not:
// removed X for deleted codeFor non-trivial tasks, define success before editing.
Use a short plan:
1. [Step] -> verify: [check]
2. [Step] -> verify: [check]
3. [Step] -> verify: [check]
Convert vague goals into verifiable goals:
Do not broaden the task silently.
Verify before claiming work is done. Do not claim something works without actually running it.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub tkxkd0159/dbud