eval-and-safety-harness

eval-and-safety-harness is a skill for Claude Code from impactbrussels/AINativeOS. It costs 176 tokens per session (1,458 once invoked), scanned A, original, Apache-2.0.

A method for testing whether an AI feature is safe and reliable enough to release, especially in health, food, or finance.

In plain words
What is it for?
Use it to create pass-or-fail safety rules, evaluation examples, and a release gate for features where incorrect answers could harm people or mislead customers.
Why use it?
It replaces confidence from a successful demo with measurable checks for citations, numbers, uncertainty, and refusal when the system is not confident enough.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the ai-native-os plugin — 25 skills, 6 agents shipped together

Good fit Use it to create pass-or-fail safety rules, evaluation examples, and a release gate for features where incorrect answers could harm people or mislead customers.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/impactbrussels/ainativeos/eval-and-safety-harness
Install

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.

Any agent
npx skills add impactbrussels/AINativeOS --skill eval-and-safety-harness
Clone the repo
git clone --depth 1 https://github.com/impactbrussels/AINativeOS

Made for: Claude Code.

Or install ai-native-os, the plugin that ships this one along with the rest of its 25 skills, 6 agents.

Wrote 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.

agentmods badge for eval-and-safety-harness

README.md
[![agentmods](https://agentmods.dev/badge/skills/impactbrussels/ainativeos/eval-and-safety-harness/github.svg)](https://agentmods.dev/skills/impactbrussels/ainativeos/eval-and-safety-harness)
Your own site
<a href="https://agentmods.dev/skills/impactbrussels/ainativeos/eval-and-safety-harness"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/eval-and-safety-harness/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.

agentmods 80×15 button for eval-and-safety-harness

Your own site · 80×15
<a href="https://agentmods.dev/skills/impactbrussels/ainativeos/eval-and-safety-harness"><img src="https://agentmods.dev/badge/skills/impactbrussels/ainativeos/eval-and-safety-harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 176 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,458 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00176 $0.01458
Opus 5 $0.00088 $0.00729
Sonnet 5 $0.00035 $0.00292
Haiku 4.5 $0.00018 $0.00146

Measured 10d ago against content hash f43d73c50b3f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

eval-and-safety-harness 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 10d 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.

skills/eval-and-safety-harness/SKILL.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Eval and Safety Harness

A demo is not evidence. You picked the input, the moment, the clean case, and the model performed because you stacked the deck. In hard mode that demo is a liability with a clean interface: the wrong answer does not stay on the screen, it walks out into a person. Trust here is engineered before you earn the right to act on a human, never assumed after. This skill builds the harness that turns generation into something a tired clinician can sign their name under.

The method

Take one real feature, the one where being wrong costs the most, and build these five layers around it. Full method, how to design the eval set, the release gate, a worked harness, anti-patterns and a copyable template: references/eval-harness-method.md.

  1. Write the safety rubric. Turn "is it safe" into pass/fail checks a machine can enforce: every claim cites a real, resolvable source; every number traces to an input; uncertain claims are flagged uncertain; confidence below a threshold refuses rather than guesses. Map each check to a pillar of the six-pillar trust substrate. If you cannot write five sharp binary checks, your safety bar is still a vibe.

  2. Pair a generator with a separate judge. One agent drafts. A second, running fresh with its own clean context and the written rubric, scores that draft before anything reaches a human. An author cannot proofread their own typo, and a model defending the context that produced its answer is the same blind spot amplified. Everything turns on the rubric: "is this good?" passes everything; the sharp check catches the hallucinated citation every run.

  3. Set the human-on-the-loop boundary. Draw the line: what agents execute unattended, and what a human must review or approve. Put human-in-the-loop on the irreversible (a dose, a large trade, a product released to market). Put human-on-the-loop on everything else, where the human reviews the exceptions queue, not every output. Too low and you have built an expensive autocomplete; too high and a bad outcome reaches a person before anyone notices.

Read the full file on GitHub · 69 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 10d ago First seen · 69 lines · 176 tokens per session scan A f43d73c50b3f

Subscribe to this mod's changes

eval-and-safety-harness is a skill published in the GitHub repository impactbrussels/AINativeOS (1 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 176 tokens to every session and 1,458 once invoked, about $0.0009 per session on Opus 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 2026-08-31.

Related

Other skills, from other repositories

imaging-data-commons

Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index. Use for accessing large-scale radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required. Query by metadata, visualize in browser, check licenses.

foryourhealth111-pixel/Vibe-Skills · 62 tokens

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…

foryourhealth111-pixel/Vibe-Skills · 152 tokens

scientific-data-preprocessing

⚠️ CRITICAL USER EXPERIENCE-BASED SKILL - ALWAYS CONSULT BEFORE DATA PREPROCESSING ⚠️ Prevents catastrophic errors (88.9% error rate in V1.0 case study) through multi-level feature analysis, data leakage detection, and semantic validation. MANDATORY for: data preprocessing, feature engineering, standardization…

foryourhealth111-pixel/Vibe-Skills · 137 tokens

embedding-strategies

Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.

foryourhealth111-pixel/Vibe-Skills · 37 tokens

perplexity-search

Perform AI-powered web searches with real-time information using Perplexity models via LiteLLM and OpenRouter. This skill should be used when conducting web searches for current information, finding recent scientific literature, getting grounded answers with source citations, or accessing information beyond the model…

foryourhealth111-pixel/Vibe-Skills · 98 tokens

similarity-search-patterns

Implement efficient similarity search with vector databases. Use when building semantic search, implementing nearest neighbor queries, or optimizing retrieval performance.

foryourhealth111-pixel/Vibe-Skills · 30 tokens