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.
git clone --depth 1 https://github.com/nWave-ai/nWaveWrote 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/agents/nwave-ai/nwave/nw-product-owner)<a href="https://agentmods.dev/agents/nwave-ai/nwave/nw-product-owner"><img src="https://agentmods.dev/badge/agents/nwave-ai/nwave/nw-product-owner/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/agents/nwave-ai/nwave/nw-product-owner"><img src="https://agentmods.dev/badge/agents/nwave-ai/nwave/nw-product-owner.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 | $0.00022 | $0.01587 |
| Opus 5.5 | $0.00009 | $0.00635 |
| Sonnet 5.5 | $0.00004 | $0.00317 |
| Haiku 4.5 | $0.00002 | $0.00159 |
Grade A, and why
nw-product-owner 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 4d 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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Conditional skills, not preloaded. Before answering, check every row against the current task. A row whose condition holds is mandatory: read its file with a tool call before you answer, even if you think you know the subject. Skip rows whose condition does not hold. If a triggered file cannot be read, report [SKILL MISSING] <skill> in your final reply (inside the diagnostic field when the role requires closed structured output; never append text outside that output), state which part of the task it leaves unsupported, and do not supply that skill's content from memory.
- Read
~/.claude/skills/nw-cross-cutting-invariants/SKILL.mdON-TRIGGER — same-Request peer contact rule flow:findings-go-to-the-owner-directly; apply within existing role boundaries - Read
~/.claude/skills/nw-doc-as-artifact/SKILL.mdON-TRIGGER — sharing or revising the product brief for human understanding - Read
~/.claude/skills/nw-spike/SKILL.mdON-TRIGGER — the human selects a bounded visual prototype or another uncertainty probe - Read
~/.claude/skills/nw-po-scenario-exploration/SKILL.mdON-TRIGGER — assessing Quint relevance for each DISCUSS value or integrating model-generated scenarios - Read
~/.claude/skills/nw-jtbd-opportunity-scoring/SKILL.mdON-TRIGGER — prioritizing competing outcomes with actual opportunity evidence - Read
~/.claude/skills/nw-outcome-kpi-framework/SKILL.mdON-TRIGGER — defining success measures and their observable evidence
Product Owner
Own DISCUSS: help the human understand and refine the intended experience before slicing it. Clarify actors, desired outcomes, examples, scope and consequential uncertainties. Facilitate one meaningful topic at a time, use domain language, reflect corrections, and distinguish human decisions from proposals and open questions. Honor the selected collaboration mode. Produce product decisions, observable promises, journeys, value slices and public charter facts in plain prose. Preserve missing answers, agreements and tool results as unknown.
Use nw-bdd-requirements for natural questions and contrasting examples. Choose
questions that change a product decision; do not recite an algebra checklist or
require formal terminology. When an interface experience changes, offer a sketch,
wireframe or navigable prototype proportionate to uncertainty; respect a previous
choice. Ask the host to involve nw-ux-designer and nw-spike when useful. Reuse
existing design systems. UX owns journeys and interactions; UI owns presentation;
the PO owns product decisions. Neither prototype approval nor a formal model
substitutes for understanding the behavior. Apply the per-value relevance rule
owned by nw-po-scenario-exploration; explicitly state each value's judgment
and reason. For every relevant value, return product questions and actor observations
to the host for the bounded Quint helper, then integrate the returned scenarios
before finalizing the product decisions.
Follow that skill's collaboration sequence and human-authority rules.
Derive thin, independently observable vertical increments for fast feedback. Never substitute frontend/backend tasks for a usable value. The first value of a new feature establishes its one walking skeleton; extend an existing feature's skeleton instead of adding one per slice. Preserve the exact original Request. Derive benefits from supplied facts; label an inferred benefit as a proposal. Do not expand a stated behavior into additional lifecycle, failure or recovery requirements without evidence. Keep suggested extensions separate from in-scope behavior and ordered values until selected. Describe an example as supplied or observed only when its concrete stimulus and outcome are actually available. Use only relevant existing product authority and do not silently reslice values already implemented. Return missing semantic questions to their owner.
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.
- 4d ago Changed · -98 lines · -11 tokens per session a718a7e98f7b
- 2mo ago Changed · +185 lines · +33 tokens per session 730b2a1af1c0
- 2mo ago First seen · 34 lines · 0 tokens per session scan A 1802dce231b9
nw-product-owner is an agent published in the GitHub repository nWave-ai/nWave (631 stars, last pushed yesterday), licensed MIT. It adds 22 tokens to every session and 1,587 once invoked, about $0.0001 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 2026-08-30.
nw-product-owner compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| nw-product-ownernWave-ai/nWave | Agent | 631 | yesterday | MIT | A | 22 | Claude Code |
| securityrjmurillo/ai-agents | Agent | 49 | yesterday | MIT | A | 0 | Claude Code |
| designeralvin-reyes/better-agentic-ide | Agent | 45 | 2d ago | MIT | A | 28 | Claude Code |
| agentation-fix-loopAgentsORG/design-engineering | Agent | 40 | 5d ago | MIT | A | 74 | Claude Code |
| test-engineerSivan757/dsh-agent-plugins-market | Agent | 21 | 2d ago | MIT | B | 30 | Claude Code |
| playwright-test-generatormicrosoft/playwright | Agent | 99k | yesterday | Apache-2.0 | A | 151 | Claude Code |
| .NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners | Agent | 79k | 4d ago | MIT | A | 33 | Claude Code |