evaluation-design

A process for designing evaluation datasets: collections of test examples used to measure whether an AI system behaves correctly and safely. It also documents how to review examples, choose measurements, and use evaluation tools.

In plain words
What is it for?
It helps define evaluation scope, balance test difficulty, create dataset contracts, review samples, select metrics, and recommend supporting tools.
Why use it?
It gives teams a consistent way to test an AI system's useful behavior, refusals, and responses under pressure instead of relying on anecdotes.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/microsoft/hve-core/evaluation-design
Any agent
npx skills add microsoft/hve-core --skill evaluation-design
Clone the repo
git clone --depth 1 https://github.com/microsoft/hve-core

Made for: Claude Code, Codex.

Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,250 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00058 $0.01250
Opus 5 $0.00029 $0.00625
Sonnet 5 $0.00012 $0.00250
Haiku 4.5 $0.00006 $0.00125

Measured 2d ago against content hash 7dcef9291bc4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evaluation-design 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 2d 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.

.github/skills/data-science-engineering/evaluation-design/SKILL.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Evaluation Dataset Design

Goal

Produce an evaluation dataset and its supporting documentation that measure whether an AI system does its job, refuses what it should refuse, and behaves acceptably under pressure. The dataset is a durable customer artifact, so its scope, balance, and rationale are recorded rather than implied.

Flow

  1. Run the scoping interview from the interview reference. Ask one question at a time and wait for the answer; do not batch the interview into a single prompt.
  2. Present a structured summary of what you heard and obtain explicit confirmation before generating anything.
  3. Derive the difficulty distribution from the confirmed scope, adjusting the defaults when the system's risk profile warrants it.
  4. Generate the dataset against the contract template in both machine-readable forms.
  5. Walk a representative sample through the user, gather consolidated feedback, and revise before finalizing the full set.
  6. Produce one sectioned evaluation guide containing curation notes, metric selection with rationale, and tooling recommendations.
  7. Route every durable write through the workstream's scan gate before it lands in a customer location.

Inputs

  • The system under evaluation, its purpose, and its intended users
  • Its grounding sources, tools, and response-format expectations
  • Known risks, refusal requirements, and prohibited content areas
  • The team's development approach and evaluation cadence
  • A caller-confirmed destination for the dataset and documents

Success criteria

  • Every interview area is answered or explicitly recorded as unknown before generation begins.
  • The dataset meets its size floor and its confirmed category balance, and the recorded difficulty counts match the actual rows.
  • Every pair records the confirmed user populations it exercises, and every confirmed population carries a count, including the ones with no pairs.
  • Each pair states its category, difficulty, expected behavior, and, where relevant, the tools the system should invoke.
  • Refusal and safety pairs assert the specific action expected, not merely that the system declines.
  • The sample review happened and its feedback is reflected in the final set.
  • Metric selection is justified from the system's actual grounding, tool use, and risk profile rather than applied uniformly.
  • The evaluation guide states who reviewed the content and when it should be revisited.

Read the full file on GitHub · 88 lines

Files

What ships with it

5 files 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. 2d ago First seen · 88 lines · 58 tokens per session scan A 7dcef9291bc4

Subscribe to this mod's changes

evaluation-design is a skill published in the GitHub repository microsoft/hve-core (1,411 stars, last pushed today), licensed MIT. It adds 58 tokens to every session and 1,250 once invoked, about $0.0003 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 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens