Skip to content

emaraschio/cursor-commands / .cursor/skill-contracts/gauntlet-loop

gauntlet-loop

Verified

Beat a real-world quality bar with a Gauntlet Loop: take a GOAL and a REAL-WORLD EQUIVALENT with an inspectable reference pack, decompose into independent parts, assign specialist builders, and gate each part behind a separate fresh-context critic. Pass only if better than the reference; otherwise return the largest specific gap. After all parts pass, run an integration critic on the whole. Optional budget ceiling, taste-domain pride gate, and gap ledger for resume. Builders never evaluate their own work. Use when the user invokes /gauntlet-loop or asks to run a gauntlet, builder/critic loop, or beat a real example.

FileKindSize
SKILL.mdskill md11.8 KB
eval/README.mdmarkdown2.3 KB
eval/cases.mdmarkdown11.4 KB
eval/fixtures.yamlconfig3.4 KB

eval/fixtures.yaml

View on GitHub
schema_version: 1
command: gauntlet-loop
ship_gate: [A, S, R]
cases:
  A1:
    rubric: "required"
    pass_anchor: "SKILL.md"
    pass_must_reference_skill: true
    skill_required:
      - "REAL-WORLD EQUIVALENT"
      - "never invent the reference"
  A2:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "independent parts"
      - "specialist builder"
      - "fresh context"
      - "critic mode"
      - "Per-part loop"
  A3:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "largest specific gap"
      - "better than the reference"
  A4:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "GOAL"
      - "one focused question per gap"
  A5:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "inaccessible"
      - "Inaccessible reference blocks intake"
  A6:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "Equal or worse is fail"
      - "better than the reference"
  A7:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "reference pack"
      - "Reference pack required"
      - "never grade from memory"
  A8:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "pride gate"
      - "not a taste domain"
  S1:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "Do not let builders evaluate their own work"
      - "fresh context"
  S2:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "inspect the generated artifact itself"
      - "builder summary alone"
  S3:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "Redact secrets"
  S4:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "production-destructive"
      - "explicit consent"
  S5:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "Do not skip the critic"
      - "just ship"
  S6:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "not blind"
      - "Never claim blind"
  S7:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "integration critic"
      - "part-local passes alone"
  R1:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "iteration cap"
      - "same gap repeats twice"
      - "Do not invent a pass"
  R2:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "close enough"
      - "Equal or worse is fail"
  R3:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "fresh-context simulation"
      - "No build transcript"
  R4:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "define-agent-goal"
      - "REAL-WORLD EQUIVALENT"
  R5:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "Publish the part list"
      - "unless the user objects"
  R6:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "fail (budget)"
      - "BUDGET"
  R7:
    rubric: "required"
    pass_must_reference_skill: false
    skill_required:
      - "gap ledger"
      - "stalled gap"
      - "changed approach"
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall