Skip to content

borghei/Claude-Skills / engineering/focused-fix

focused-fix

Verified

Constrain a bugfix to the smallest reviewable change: reproduce, patch, add a regression test, check the diff for scope creep. Use when asked for a minimal fix, hotfix or release-blocker; diagnose unknown causes with bug-diagnosis.

evals/test_cases.json

View on GitHub
{
  "skill": "focused-fix",
  "version": "1.0.0",
  "cases": [
    {
      "id": "minimal-scope-basic",
      "prompt": "Login fails when an email address contains a plus sign. Fix it.",
      "expected": {
        "format": "markdown",
        "must_contain": [
          "minimal",
          "regression test"
        ],
        "must_not_contain": [
          "refactor",
          "while I'm here",
          "cleanup"
        ]
      },
      "rubric": {
        "structure": "Output identifies a minimal change set, references the change_scope_analyzer tool or its workflow, and requires a regression test",
        "accuracy": "Diagnosis focuses on the email-parsing layer; does not propose broader auth refactors",
        "scope": "Stays inside the focused-fix workflow — does not pivot to feature work"
      },
      "weight": 1.0
    },
    {
      "id": "reject-scope-creep",
      "prompt": "I'm fixing a CSS overflow bug on mobile and noticed our color tokens are inconsistent across the file. Should I clean those up in the same PR?",
      "expected": {
        "format": "markdown",
        "must_contain": [
          "separate PR",
          "/refactor:?/"
        ],
        "must_not_contain": [
          "yes, go ahead and clean",
          "while you're in there"
        ]
      },
      "rubric": {
        "structure": "Output recommends a separate PR for the token cleanup and explains why",
        "accuracy": "Cites the 'while I'm here' anti-pattern or the diff-readability/blast-radius reasoning",
        "scope": "Holds the line on minimal change discipline"
      },
      "weight": 1.0
    },
    {
      "id": "structural-fix-escalation",
      "prompt": "The analyzer says the fix is 'structural' — it spans 8 files across 3 modules. What do I do?",
      "expected": {
        "format": "markdown",
        "must_contain": [
          "stop",
          "/scope|design/"
        ],
        "must_not_contain": [
          "proceed with the bugfix",
          "ship it"
        ]
      },
      "rubric": {
        "structure": "Output stops the focused-fix workflow and recommends scoping a refactor or design discussion first",
        "accuracy": "Treats 'structural' as a halt condition per the documented workflow, not advice",
        "scope": "Does not attempt to fix the bug inside the focused-fix workflow"
      },
      "weight": 1.0
    },
    {
      "id": "missing-regression-test",
      "prompt": "I made a one-line fix that's obviously correct. Can I skip the regression test this time?",
      "expected": {
        "format": "markdown",
        "must_contain": [
          "/regression test/",
          "no"
        ],
        "must_not_contain": [
          "you can skip",
          "fine to skip"
        ]
      },
      "rubric": {
        "structure": "Output declines and explains the silent-regression risk on future refactors",
        "accuracy": "References the every-fix-commit-adds-a-test rule from the skill's anti-patterns",
        "scope": "Stays on focused-fix discipline, not test strategy in general"
      },
      "weight": 0.5
    }
  ]
}
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 322kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 322kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 322kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall