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 QuantumByteOSS/quantumbyte --skill first-run-opengit clone --depth 1 https://github.com/QuantumByteOSS/quantumbyteWrote 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/quantumbyteoss/quantumbyte/first-run-open)<a href="https://agentmods.dev/skills/quantumbyteoss/quantumbyte/first-run-open"><img src="https://agentmods.dev/badge/skills/quantumbyteoss/quantumbyte/first-run-open/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/quantumbyteoss/quantumbyte/first-run-open"><img src="https://agentmods.dev/badge/skills/quantumbyteoss/quantumbyte/first-run-open.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00102 | $0.00917 |
| Opus 5.5 | $0.00041 | $0.00367 |
| Sonnet 5.5 | $0.00020 | $0.00183 |
| Haiku 4.5 | $0.00010 | $0.00092 |
Grade A, and why
first-run-open 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 1mo 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
First-Run — the opening
The opening's job: get to a substantive, qualifiable intent without ever facing the user with a blank box. Show a guess; the correction is the data.
Posture by modality (set this first)
- Text → show-led. Lead with the visible sketch and tappable starting-point chips. A blank text field is the highest-drop-off move there is.
- Voice → say-led. Lead with a spoken hypothesis and one open question ("tell me what's eating your time"). Surface the blueprint/buttons on a pause — never narrate a slide the user can't look at while talking. Grounding is brittle in voice: if you mishear a name, repair it immediately, before anything else.
Two entries
- User-initiated (a real first message exists) → no cold-start needed. Apply the modality posture and hand straight back to the spine's T1 (appraise + sketch).
- AI-initiated / cold start (no prompt; user signed up and landed) → run the show-to-learn loop below.
Cold start: show a provisional dream + a guess about the person
Branch on passive context:
- Context-rich (firmographics resolved — company, industry from
get_user_context): render a provisional Dream Preview — a concrete sketch from the firmographics — and a hypothesis about who they are. "I pulled up — looks like a . Most s your size lose the most time to , so here's a starting sketch. I'm guessing you run the — close? And is that the real headache, or something else?" - Context-poor (nothing to seed from): present starting-point chips — a few concrete, high-magnet use-cases by archetype — with a free-text escape hatch. Recognition over recall; never "what do you want to build?"
The first preview is provisional by design. Firmographics give you the company, not the person — and the person decides which dream is even right (a solo founder and a champion get different vacations). Open with the safe company-level guess, then re-aim once they tell you who they are.
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.
- 1mo ago First seen · 47 lines · 102 tokens per session scan A 32976f7f7eb1
first-run-open is a skill published in the GitHub repository QuantumByteOSS/quantumbyte (294 stars, last pushed 2d ago), licensed Apache-2.0. It adds 102 tokens to every session and 917 once invoked, about $0.0004 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.
Other skills, from other repositories
factory-rereview
Re-review a pull request after a push — reconcile the previous review against the new commits, look for new defects the push introduced, then a fresh pass over the whole PR, and finish with a verdict on the PR.
understand-pr
Review a PR the way a seasoned maintainer would — build an independent model before seeing the diff, investigate the implementation and history, verify material claims, and return a concise briefing. Use for any PR review, self-review, or re-review.
herdr-kit-workflows
Set up and operate Herdr Kit safely: install and configure the plugin and Mastra Code integration, manage Review and Work repository scope, open primary repository workspaces, synchronize managers, materialize or dematerialize managed worktrees, and update Herdr Kit settings or shortcuts.
factory-triage
Triage a Factory work item's issue — trace history, understand architecture, diagnose root cause, then advance the stage.
codemod-patterns
This guide helps AI agents efficiently create codemods with optimal performance and consistency.
factory-gitlab-review
Review a GitLab merge request for a Factory work item using brokered source-control tools.