Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add racecraft-lab/racecraft-plugins-public --skill typed-judgmentsgit clone --depth 1 https://github.com/racecraft-lab/racecraft-plugins-publicWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/racecraft-lab/racecraft-plugins-public/typed-judgments)<a href="https://agentmods.dev/skills/racecraft-lab/racecraft-plugins-public/typed-judgments"><img src="https://agentmods.dev/badge/skills/racecraft-lab/racecraft-plugins-public/typed-judgments/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/racecraft-lab/racecraft-plugins-public/typed-judgments"><img src="https://agentmods.dev/badge/skills/racecraft-lab/racecraft-plugins-public/typed-judgments.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00145 | $0.02172 |
| Opus 5.5 | $0.00058 | $0.00869 |
| Sonnet 5.5 | $0.00029 | $0.00434 |
| Haiku 4.5 | $0.00015 | $0.00217 |
Grade A, and why
typed-judgments scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 21d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Typed judgments with Jev
The evaluate tool answers a question you define, from state you supply, with
a probability attached. Nothing to parse and no prompt to maintain.
When to reach for it
The trigger is a judgment you are about to make by intuition and then state as fact. In a working session that is usually one of:
| Moment | Question type |
|---|---|
| Is this change ready to merge? | score over readiness levels |
| Which of these options fits? Who owns this? | choice |
| How severe or risky is this? | score |
| Is this failure real or unrelated to the change? | noul |
| Does this satisfy the constraint stated earlier? | noul |
| Classify or rank many items the same way | one call per item |
Use it before asserting the judgment, not to confirm one already given.
When not to
- Writing code, prose, commit messages, or anything generative.
- A search, a lookup, or reading a file: those are cheaper done directly.
- Any question whose answers cannot be listed up front. The answer space must be enumerable, at most 255 options.
- A judgment the user already made. Their decision stands.
The three primitives
noul— the probability a yes/no condition holds. Optionalcriteriais{"true": "...", "false": "..."}.choice— one option from a requiredcriteriamap of option to description.score— a position on an ordered array of at least two level descriptions. The answer is probability-weighted, so2.13means mostly level 2 with some weight on 3. That fraction is the signal; do not round it away.
Learning to ask a better question
The rules below are the short version. TypeSafe publishes the long one, and it is live rather than bundled here, so it stays current as the model does. Read a page when the judgment in front of you is hard to frame, not routinely.
Mintlify serves Markdown by appending .md to a page path. Resolve relative
links against https://docs.typesafe.ai, and read targeted pages rather than
the whole site.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 21d ago First seen · 185 lines · 145 tokens per session scan A 2567348f2d90
typed-judgments is a skill published in the GitHub repository racecraft-lab/racecraft-plugins-public (5 stars, last pushed yesterday), licensed MIT. It adds 145 tokens to every session and 2,172 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-26.
Other skills, from other repositories
agent-platform-endpoint-management
Manages Agent Platform serving endpoints. Use when you need to create, list, describe, update, or delete serving endpoints for model deployment on Agent Platform. Also use when troubleshooting endpoint permission, quota, or resource busy errors. Don't use for deploying models to endpoints or for running model…
gpt-image-2
A skill for generating or editing images with GPT Image 2 across local, host-provided, or advisory setups.
ml-kit-genai-prompt-api
Analyzes Android codebases to implement ML Kit GenAI Prompt API. Use this skill to send natural language requests on-device to Gemini Nano, use structured output with Prompt API, implement prefix caching, optimize the current prompt, or apply best practices.".
LQF_Machine_Learning_Expert_Guide
LQF Machine Learning Expert Guide - Routed skill for ML/Statistical Modeling with Critical Discussion Mode. Triggers on: machine learning, modeling, prediction, training, classification, regression, clustering, deep learning, neural network, model evaluation, feature engineering, hyperparameter tuning, overfitting…
similarity-search-patterns
Implement efficient similarity search with vector databases. Use when building semantic search, implementing nearest neighbor queries, or optimizing retrieval performance.
nowait-reasoning-optimizer
Implements the NOWAIT technique for efficient reasoning in R1-style LLMs. Use when optimizing inference of reasoning models (QwQ, DeepSeek-R1, Phi4-Reasoning, Qwen3, Kimi-VL, QvQ), reducing chain-of-thought token usage by 27-51% while preserving accuracy. Triggers on "optimize reasoning", "reduce thinking tokens"…