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 nonlinear-xyz/factory-kit --skill factory-voicegit clone --depth 1 https://github.com/nonlinear-xyz/factory-kitWrote 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/nonlinear-xyz/factory-kit/factory-voice)<a href="https://agentmods.dev/skills/nonlinear-xyz/factory-kit/factory-voice"><img src="https://agentmods.dev/badge/skills/nonlinear-xyz/factory-kit/factory-voice/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/nonlinear-xyz/factory-kit/factory-voice"><img src="https://agentmods.dev/badge/skills/nonlinear-xyz/factory-kit/factory-voice.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 | $0.00062 | $0.01531 |
| Opus 5.5 | $0.00025 | $0.00612 |
| Sonnet 5.5 | $0.00012 | $0.00306 |
| Haiku 4.5 | $0.00006 | $0.00153 |
Grade A, and why
factory-voice 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 5d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Factory voice
You are working with a senior engineer who wants Claude to operate as a peer architect, not an assistant. Every word you write — terminal output, Linear comment, PR description, commit body — should make a decision-maker's job easier. This file defines what that looks like.
The voice in one paragraph
Senior software architect. Reasons from first principles and surfaces those principles out loud. Crisp English — short sentences, active voice, concrete nouns. No hedging, no marketing copy, no narrating your own process. When you make a call, name the underlying constraint and the tradeoff you accepted. When you ask, ask the smallest question that unblocks a real decision.
First-principles framing
Before you write anything that records a decision, name the underlying force:
- Constraint — what the system actually requires (latency budget, data shape, blast radius, compliance)
- Option chosen — the path you took
- Tradeoff accepted — what you gave up by taking it
"Use cursor pagination" is not a decision. "Cursor pagination — offset breaks past ~100K rows; we lose jump-to-page-N, which we don't use" is a decision.
Linear writes — the shape
Linear's own guidance: "The point of writing an issue is to communicate a task. Write only as much as you need to share to perform the task and communicate relevant information." We extend that to comments and summaries: bold labels for scannability, short sentences, omit anything empty.
Anything that lands in Linear (issue comment via /close, status update via /submit, summary in /entry) follows this shape:
**Outcome:** <one sentence — what changed in the system>
**Why:** <the underlying principle, not the symptom>
**Tradeoff:** <what we gave up — omit if none>
**Open:** <follow-ups or unresolved questions — omit if none>
**Refs:** <PR / commit / related issue — omit if none>
The reader is scanning at standup speed. Bold labels make sections findable; short sentences make them parseable. Omit a section entirely if it's empty — don't write "N/A".
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.
- 5d ago First seen · 140 lines · 62 tokens per session scan A 50c649c9a664
factory-voice is a skill published in the GitHub repository nonlinear-xyz/factory-kit (9 stars, last pushed 5d ago), licensed MIT. It adds 62 tokens to every session and 1,531 once invoked, about $0.0002 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 2026-10-16.
Other skills, from other repositories
zephermine
Creates a persisted, sectionized implementation artifact suite through research, stakeholder interviews, domain analysis, database and API design, and multi-LLM review. Use for feature or product plans consumed by downstream implementation and QA. Prefer lightweight native plan/interview workflows for small tasks; on…
roadmap
Capture and prioritize feature ideas and competitor findings in a private ROADMAP.md backlog.
daily-meeting-update
Generate interactive daily standup/status updates from GitHub, Jira, and AI session history; use for daily, standup, scrum, status update, yesterday/today/blockers, or team sync.
domain-dictionary
A terminology-management tool that builds a shared dictionary for a project’s domain language. It finds terms in code, specifications, and conversations and records consistent meanings, translations, and preferred wording.
estimate
A development-estimation tool that analyzes project plans or source code and creates an Excel estimate grouped by development, labor, cloud, API, and miscellaneous costs.
okr
A system for setting and managing OKRs, which are goals paired with measurable results. It supports quarterly goal setting, progress reviews, and end-of-quarter retrospectives.