Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Designing deterministic, verifiable metrics v1.0.0

Use this resource to turn a fuzzy notion into a deterministic metric an agent can compute and optimise

New

  • 474 downloads
  • 171 likes
  • 137 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install deterministic-metric-invention

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent designs a metric against 44 rules: operationalising the construct, computable approximations, proven properties, determinism, construct validity and resistance to gaming. It reports uncertainty and supplies test vectors. You get a reliable measure that can be tracked and optimised without being distorted.

Mostly picked by

Updated November 2, 2026

Card for the “Designing deterministic, verifiable metrics” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • references/: cards on operationalisation and approximations with bounded error
  • references/: cards on aggregation, public versioning and test vectors

Good fit for

  • You want a code quality score that agents can optimise
  • An internal metric already rewards gamed behaviour

Not for

    Tags

    • metric
    • measurement
    • quality
    • determinism
    • validity

    About

    You say

    “Define a deterministic maintainability metric for our modules, with its proven properties and its limits.”

    Card for the “Designing deterministic, verifiable metrics” skill

    Skill published by pproenca under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to turn a fuzzy notion into a deterministic metric an agent can compute and optimise
    3. Review and adjust the result

    Content preview

    Files

    • AGENTS.md12.8 KB
    • assets/templates/_template.md1.9 KB
    • LICENSE1.1 KB
    • metadata.json2.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • README.md8.3 KB
    • references/_sections.md6.4 KB
    • references/agg-report-uncertainty-not-false-precision.md1.3 KB
    • references/agg-respect-scale-in-aggregation.md1.5 KB
    • references/agg-ship-reference-impl-and-test-vectors.md1.6 KB
    • references/agg-version-the-metric-publicly.md1.4 KB
    • references/comp-bound-approximation-error-explicitly.md1.6 KB
    • references/comp-choose-a-decidable-observational-equivalence.md2 KB
    • references/comp-design-a-proxy-with-a-proven-error-direction.md2 KB
    • references/comp-do-not-define-metric-as-uncomputable-ideal.md2 KB
    • references/comp-keep-the-metric-tractable.md1.9 KB
    • references/comp-prefer-monotone-confluent-transformations.md2.1 KB
    • references/comp-respect-rices-theorem-for-semantic-properties.md2 KB
    • references/def-anchor-to-the-decision.md1.7 KB
    • references/def-fix-unit-of-analysis.md1.9 KB
    • references/def-name-the-latent-construct.md2.1 KB
    • references/def-operationalize-behavior-and-size.md2.1 KB
    • references/def-separate-construct-from-proxy.md1.8 KB
    • references/def-write-falsifiable-operational-definition.md1.6 KB
    • references/det-control-floating-point-and-accumulation.md1.6 KB
    • references/det-make-the-metric-a-pure-function.md1.5 KB
    • references/det-pin-iteration-and-tie-break-order.md1.5 KB
    • references/det-pin-the-input-representation.md1.5 KB
    • references/det-version-and-record-the-toolchain.md1.4 KB
    • references/game-detect-reward-hacking-with-audits.md1.5 KB
    • references/game-hard-block-construct-violating-wins.md1.6 KB
    • references/game-make-cheapest-improvement-the-right-one.md1.6 KB
    • references/game-pair-with-guardrail-metrics.md1.6 KB
    • references/game-recognize-goodhart-variants.md1.6 KB
    • references/meas-avoid-ad-hoc-weighted-sums.md1.7 KB
    • references/meas-declare-the-scale-type.md1.6 KB
    • references/meas-establish-meaningful-zero-and-unit.md1.4 KB
    • references/meas-only-admissible-statistics.md1.5 KB
    • references/meas-preserve-the-empirical-relation.md1.6 KB
    • references/prop-check-weyuker-briand-axioms.md1.7 KB
    • references/prop-ensure-sensitivity-to-relevant-change.md1.6 KB
    • references/prop-prove-boundedness-and-handle-empty.md1.6 KB
    • references/prop-prove-invariance-under-irrelevant-transforms.md1.6 KB
    • references/prop-prove-monotonicity.md1.6 KB
    • references/prop-prove-or-disclaim-composability.md1.6 KB
    • references/valid-beat-the-trivial-baseline.md1.4 KB
    • references/valid-calibrate-thresholds-to-ground-truth.md1.5 KB
    • references/valid-converge-with-accepted-measure.md1.7 KB
    • references/valid-discriminant-not-just-loc.md1.6 KB
    • references/valid-predictive-validity-against-outcome.md1.5 KB
    • references/valid-validate-out-of-sample.md1.5 KB
    • SKILL.md13.9 KB

    Origin and licence

    Listed unmodified

    Author
    pproenca
    Source
    github.com/pproenca/dot-skills/tree/cf93c57cac89d6fc3e4194686000411567f5caf3/skills/.curated/deterministic-metric-design
    Licence
    MIT

    Notice

    « deterministic-metric-design » (pproenca, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.