yolo-debug

星标9

分支0

更新时间2026年7月2日 17:45

Use when a reported problem, failure, or bug needs a systematic root-cause investigation — reproduce → isolate → fix → verify — recorded as one durable artifact. Triggers on "fix the 500 on login", "X is broken", "why is Y failing?", "debug this", "track down this bug".

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

来源

CoriChui

CoriChui/yolo

打开 GitHub 仓库查看创作者相关仓库

下载

yolo-debug

Find the true cause of a failure and fix it with evidence — not by guessing. A stateless investigation act: it produces ONE debug record plus the fix, and stores no session state.

Inputs

A bug report or failing behaviour. Derive a kebab-case slug for the record filename — there is no brief (a debug is not a feature). Run on the current branch (see .claude/yolo/conventions.md — a debug record never touches status derivation). If the work turns out to add new capability rather than restore intended behaviour, hand off to yolo-feature.

Procedure

Work the arc in order; each step's output feeds the record.

Reproduce — get a deterministic repro (a failing test, a command, exact steps). No repro → no confirmed bug yet; say so rather than guessing.
Hypothesize — list candidate causes; rank by likelihood and cost-to-check.
Isolate — trace to the true cause with a concrete evidence chain (file:line → observed value → why it's wrong). Bisect or instrument as needed.
Fix — make the minimal change that addresses the root cause, not the symptom. Note the over-scope traps you avoided.
Verify — the failing reproducer now passes, the symptom is gone, and the wider suite stays green.

The Iron Law: no root-cause claim without a file:line evidence chain; no fix without a failing reproducer that the fix flips to green; never patch the symptom while the cause stands. Failing-test-first is the default — if a bug is genuinely not unit-testable (config, perf, flake, environment), record why in the artifact and substitute the strongest available evidence (a before/after command, a metric, a log excerpt).

Discipline (the traps that end investigations early)

"It's probably X" without checking is a guess, not a cause — confirm before you fix.
The first plausible cause is not always the cause — confirm it explains the symptom fully.
A green fix with no reproducer proves nothing — you may have moved the bug, not killed it.
Red flags to stop and re-examine: the fix touches unrelated files; the bug "can't be reproduced but is fixed"; the explanation needs a "this should never happen".

Output

Write exactly ONE artifact — workspace/debug/<YYYY-MM-DD>-<slug>.md (mkdir -p its dir) — with: symptom + reproducer, the evidence chain (file:line), a one-sentence root cause + confidence, the fix (files, the minimal change, over-scope avoided), and verification evidence. Commit the record yolo: debug <slug>; the fix lands as ordinary descriptive commit(s) on the current branch.

Constraints

Cite .claude/yolo/conventions.md for the commit form and the standalone-record rules; do not restate trailer strings. A debug record carries no trailer, no brief, no tracked state.
Stateless: no persistent session or run-state files — the committed record plus the fix commit are the entire trail.
Hand off to yolo-feature the moment the work becomes new capability rather than a fix.

同仓库更多 Skills

同仓库

yolo-feature

CoriChui/yolo

Use when the user wants to build, add, or implement a feature. Captures intent, drafts a brief, confirms, then composes research → plan → execute → verify → finish. In a YOLO-initialized repo this takes precedence over generic brainstorming for feature intent.

2026-07-079

yolo-decide

CoriChui/yolo

Use when facing a design or architecture decision that benefits from multiple perspectives before committing. Produces a decision record under workspace/decisions/. Triggers on "help me decide X vs Y", "which approach", "should we use A or B".

2026-07-079

yolo-init

CoriChui/yolo

Use when setting up YOLO in a project for the first time, or repairing the setup. Scaffolds workspace/config.yaml and the features/, decisions/, and debug/ dirs (plus docs/ for the docs engine), and installs the CLAUDE.md routing block. Triggers on "set up YOLO here", "init yolo", "yolo init".

2026-07-079

yolo-intake

CoriChui/yolo

Use when a feature should draw on the project's reference material for context. yolo-intake does NOT fetch, import, or copy anything — it points YOLO at a selected folder (default `docs/`) and READS from it to enrich a feature's context. Getting material INTO that folder (a Figma, a spec, a URL, a schema) is the docs engine's or another connector's job, done separately. Triggers on "use the docs as context", "reference the spec/schema in docs", "the feature should read our docs".

2026-07-079

yolo-research

CoriChui/yolo

Use before planning a YOLO change, to understand the codebase and any cited intake before a plan is written. Read-only exploration that produces context for yolo-plan. Triggers on "understand how X works", "explore the code before we change it", or as the research step of yolo-feature.

2026-07-079

yolo-finish

CoriChui/yolo

Use when a feature is implemented and verified, to land it. Default path is PR + CI check, with the ship gate confirming before the irreversible merge; fast-local is the escape hatch. Includes the risk classifier whose hard triggers always stop for a human. Triggers on "ship it", "land this", "open a PR", or as the finish step of yolo-feature.

2026-07-029

name	yolo-debug
description	Use when a reported problem, failure, or bug needs a systematic root-cause investigation — reproduce → isolate → fix → verify — recorded as one durable artifact. Triggers on "fix the 500 on login", "X is broken", "why is Y failing?", "debug this", "track down this bug".

yolo-debug

Find the true cause of a failure and fix it with evidence — not by guessing. A stateless investigation act: it produces ONE debug record plus the fix, and stores no session state.

Inputs

Procedure

Work the arc in order; each step's output feeds the record.

Reproduce — get a deterministic repro (a failing test, a command, exact steps). No repro → no confirmed bug yet; say so rather than guessing.
Hypothesize — list candidate causes; rank by likelihood and cost-to-check.
Isolate — trace to the true cause with a concrete evidence chain (file:line → observed value → why it's wrong). Bisect or instrument as needed.
Fix — make the minimal change that addresses the root cause, not the symptom. Note the over-scope traps you avoided.
Verify — the failing reproducer now passes, the symptom is gone, and the wider suite stays green.

Discipline (the traps that end investigations early)

"It's probably X" without checking is a guess, not a cause — confirm before you fix.
The first plausible cause is not always the cause — confirm it explains the symptom fully.
A green fix with no reproducer proves nothing — you may have moved the bug, not killed it.
Red flags to stop and re-examine: the fix touches unrelated files; the bug "can't be reproduced but is fixed"; the explanation needs a "this should never happen".

Output

Constraints

Cite .claude/yolo/conventions.md for the commit form and the standalone-record rules; do not restate trailer strings. A debug record carries no trailer, no brief, no tracked state.
Stateless: no persistent session or run-state files — the committed record plus the fix commit are the entire trail.
Hand off to yolo-feature the moment the work becomes new capability rather than a fix.