Skip to content

JPeetz/agent-skills / skills/development/finding-unknowns

finding-unknowns

Verified

Use when entering unfamiliar code or domain territory.

FileKindSize
CHANGELOG.mdmarkdown668 B
LICENSEother1.3 KB
SKILL.mdskill md4.8 KB
evals/evals.jsonconfig2.5 KB

evals/evals.json

View on GitHub
{
  "skill": "finding-unknowns",
  "version": "1.0.0",
  "cases": [
    {
      "id": "e1",
      "description": "Should trigger: blindspot-pass before unfamiliar work",
      "input": "I need to add authentication to our Express API. What am I missing?",
      "shouldTrigger": true,
      "expect": "Produces a blindspot report with landmines, hidden context, and quality calibration",
      "assertions": [
        "Does not start implementing",
        "Reports landmines and hidden context",
        "Ends with a rewritten request incorporating the findings"
      ]
    },
    {
      "id": "e2",
      "description": "Should trigger: context-audit for bloated config",
      "input": "My CLAUDE.md has grown to 200 lines. Audit it.",
      "shouldTrigger": true,
      "expect": "Inventories all context layers, finds contradictions, proposes cuts",
      "assertions": [
        "Scans all context layers reaching the model",
        "Classifies instructions as conflict/duplicate/obvious/gotcha",
        "Proposes specific deletions with before/after line counts"
      ]
    },
    {
      "id": "e3",
      "description": "Should trigger: implementation-notes during a build",
      "input": "Start building the user profile module per the plan. Keep notes.",
      "shouldTrigger": true,
      "expect": "Creates implementation-notes.md and logs deviations",
      "assertions": [
        "Creates a running notes file with Deviations, Edge Cases, Questions headings",
        "Logs every deviation from plan with rationale",
        "Ends with a 5-line summary"
      ]
    },
    {
      "id": "e4",
      "description": "Should trigger: change-quiz after a session",
      "input": "What did we actually change in this session? Give me a quiz.",
      "shouldTrigger": true,
      "expect": "Produces a change report and 5-8 question quiz",
      "assertions": [
        "Groups changes by intent, not by file",
        "Quiz targets interaction effects and edge cases, not trivia",
        "Misses get explained with the correct answer"
      ]
    },
    {
      "id": "e5",
      "description": "Should trigger: reference-hunt from existing code",
      "input": "Implement pagination like the admin panel does it.",
      "shouldTrigger": true,
      "expect": "Reads the reference code, produces semantics summary, reimplements",
      "assertions": [
        "Produces a semantics summary before writing code",
        "User confirms the summary before implementation",
        "Maps behaviors from reference to new implementation"
      ]
    }
  ]
}
receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementationStars: 322kUpdated Claude CodeCodexVerifiedInstall
requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirementsStars: 322kUpdated Claude CodeCodexVerifiedInstall
using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallbackStars: 322kUpdated Claude CodeCodexVerifiedInstall
verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions alwaysStars: 322kUpdated Claude CodeCodexVerifiedInstall