xiduzo/wtf / skills/wtf.hotfix
wtf.hotfix
VerifiedThis skill should be used when something is broken in production and needs an immediate fix that bypasses the normal Epic→Feature→Trace planning flow — for example "production is down", "hotfix needed", "critical bug in prod", "emergency fix for
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.5 KB |
| evals/evals.json | config | 4.4 KB |
| evals/fixtures/narrow-fix.json | config | 1.1 KB |
| evals/fixtures/pr-body-completeness.json | config | 1.1 KB |
| evals/fixtures/scope-gate-fires.json | config | 1.4 KB |
evals/evals.json
View on GitHub{
"skill_name": "wtf.hotfix",
"evals": [
{
"id": 1,
"prompt": "Test mode: skip steps 0–3 (gh setup, bug identification, scope confirmation, and TECH.md load are already done). Read evals/fixtures/narrow-fix.json and use its bug object as the live bug context. Run step 4 only: determine the hotfix branch name for this bug, print the branch name, and explain how the slug was derived from the bug title. Do not start implementing the fix.",
"expected_output": "A hotfix branch name in the format hotfix/<bug-number>-<slug>, where the slug is a 2–4 word kebab-case summary derived from the bug title (e.g. hotfix/88-settlement-rounding-truncation). The output explains the slug derivation and confirms the branch targets main. No code or tests are written.",
"files": ["evals/fixtures/narrow-fix.json"],
"expectations": [
"The branch name starts with 'hotfix/' followed by the bug number 88 and a kebab-case slug (matches the pattern ^hotfix/88-[a-z0-9-]+$).",
"The slug contains only lowercase letters, digits, and hyphens — no special characters, spaces, or capital letters.",
"The output states that the branch targets main (not a feature branch or any other base).",
"The output explains how the slug was derived from the bug title words.",
"The output does not contain any code implementation, test code, or file edits — step 4 only."
]
},
{
"id": 2,
"prompt": "Test mode: skip steps 0–5 (gh setup, bug identification, scope confirmation, TECH.md load, branch setup, and codebase exploration are already done). Read evals/fixtures/scope-gate-fires.json and use its codebase_findings as the exploration result. Run step 6 only: evaluate whether the scope gate fires for this fix, describe what was found, and present the keep/proceed vs switch-to-normal-flow choice to the user. Do not write any code or tests.",
"expected_output": "The scope gate fires. The output describes the findings: 6 files affected (exceeds the 3–4 file threshold), one schema migration (add_currency_code to settlements table), and two API contract changes. The user is presented with a binary choice: proceed as hotfix (accepting the larger scope) or switch to the normal Trace workflow (write-feature + feature-to-traces). No code or tests are written.",
"files": ["evals/fixtures/scope-gate-fires.json"],
"expectations": [
"The output explicitly states that the scope gate fires — it does not silently proceed to implementation.",
"The output reports the number of files affected (6) and notes this exceeds the hotfix threshold of 3–4 files.",
"The output describes the schema migration (adding currency_code to the settlements table) and the API contract change.",
"The user is presented with two distinct options: proceed as hotfix or switch to the normal Trace workflow (write-feature / feature-to-traces).",
"No code, test, or file edit appears in the output — step 6 only."
]
},
{
"id": 3,
"prompt": "Test mode: skip steps 0–7 (gh setup, bug identification, scope confirmation, TECH.md load, branch setup, codebase exploration, scope gate, and implementation are already done). Read evals/fixtures/pr-body-completeness.json and use its bug and fix_context objects as the live context. Run step 8 only: produce the full PR title and PR body that would be passed to gh pr create.",
"expected_output": "A PR title in the format 'fix(settlement): use integer arithmetic for settlement amount calculation to avoid floating-point truncation' and a PR body containing five sections: Related (with Closes #88), Impact, Root cause, Fix, and Risk. The body is ready to be written to a temp file and passed to gh pr create --body-file.",
"files": ["evals/fixtures/pr-body-completeness.json"],
"expectations": [
"The PR title matches the pattern ^fix\\([a-z-]+\\): [a-z] — it starts with 'fix(' followed by a lowercase scope, closing paren, colon, space, and a lowercase description.",
"The PR body contains a 'Closes #88' line in the Related section.",
"The PR body contains an Impact section that describes what was broken and who was affected.",
"The PR body contains a Root cause section that explains the floating-point truncation in calculateSettlementAmount().",
"The PR body contains a Risk section describing what could be affected by the change."
]
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |