distill

distill is a command for Claude Code from nWave-ai/nWave. It costs 26 tokens per session (4,997 once invoked), scanned A, original, MIT.

Creates E2E acceptance tests in Given-When-Then format from requirements and architecture. Use when preparing executable specifications before implementation.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: reads .claude/ paths.

Part of the nw plugin — 12 commands shipped together

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 commands/nwave-ai/nwave/distill
Clone the repo
git clone --depth 1 https://github.com/nWave-ai/nWave

Made for: Claude Code.

Or install nw, the plugin that ships this one along with the rest of its 12 commands.

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 distill

README.md
[![agentmods](https://agentmods.dev/badge/commands/nwave-ai/nwave/distill.svg)](https://agentmods.dev/commands/nwave-ai/nwave/distill)
Your own site
<a href="https://agentmods.dev/commands/nwave-ai/nwave/distill"><img src="https://agentmods.dev/badge/commands/nwave-ai/nwave/distill.svg" alt="Measured on agentmods" height="20"></a>
Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,997 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown 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 $0.00026 $0.04997
Opus 5 $0.00013 $0.02499
Sonnet 5 $0.00005 $0.00999
Haiku 4.5 $0.00003 $0.00500

Measured today against content hash c6e90fa620ef, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

distill 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 today.

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.

plugins/nw/commands/distill.md · 407 lines

How it starts

The opening of the file, as written. The whole thing — 407 lines — stays where its author put it; the contents beside it link to each section on GitHub.

NW-DISTILL: Acceptance Test Creation and Business Validation

Wave: DISTILL (wave 5 of 6) | Agent: Quinn (nw-acceptance-designer)

Overview

Orchestrate acceptance test creation from prior wave artifacts, then gate the result through parallel reviews before handoff to DELIVER. You (main Claude instance) are the orchestrator. You dispatch agents and enforce gates.

The AT-completeness gate, MAX-PBT mandate, and Mandate-12 step-reuse metric are advisory.

REVIEW GATE SUMMARY (read this first)

After the acceptance designer produces scenarios, you MUST dispatch 4 parallel reviewers if scenario count exceeds 3 (Eclipse + Architect + Forge + Sentinel). Sentinel (@nw-acceptance-designer-reviewer) is the structural-correctness reviewer — it ALWAYS dispatches even on fast-path or under rigor.reviewer_model: "skip" (which only skips scale-sensitive cost-driven reviewers). This is the single most important orchestration step in DISTILL. The procedure is: dispatch designer -> count scenarios -> dispatch 4 reviewers in parallel -> AND-gate results -> handoff. Details in Phase 3 below.

Phase 1: Decisions and Context

Interactive Decision Points

Decision 1: Feature Scope

Question: What is the scope of this feature? Options:

  1. Core feature -- primary application functionality
  2. Extension -- modular add-on or integration
  3. Bug fix -- regression tests for a known defect
Decision 2: Test Framework

Question: Which test framework to use? Options:

  1. pytest-bdd -- Python BDD framework
  2. Cucumber -- Ruby/JS BDD framework
  3. SpecFlow -- .NET BDD framework
  4. Custom -- user provides details
Decision 3: Integration Approach

Question: How should integration tests connect to services? Options:

  1. Real services -- test against actual running services
  2. Test containers -- ephemeral containers for dependencies
  3. Mocks for external only -- real internal, mocked external services
Decision 4: Infrastructure Testing

Question: Should acceptance tests cover infrastructure concerns? Options:

  1. Yes -- include CI/CD validation, deployment smoke tests
  2. No -- functional acceptance tests only

Read the full file on GitHub · 407 lines

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. today First seen · 407 lines · 26 tokens per session scan A c6e90fa620ef

Subscribe to this mod's changes

distill is a command published in the GitHub repository nWave-ai/nWave (605 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 4,997 once invoked, about $0.0001 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 2026-09-06.