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 Kerliula/ship-me --skill grill-megit clone --depth 1 https://github.com/Kerliula/ship-meWrote 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/kerliula/ship-me/grill-me)<a href="https://agentmods.dev/skills/kerliula/ship-me/grill-me"><img src="https://agentmods.dev/badge/skills/kerliula/ship-me/grill-me/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/kerliula/ship-me/grill-me"><img src="https://agentmods.dev/badge/skills/kerliula/ship-me/grill-me.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 | $0.00197 | $0.05111 |
| Opus 5.5 | $0.00079 | $0.02044 |
| Sonnet 5.5 | $0.00039 | $0.01022 |
| Haiku 4.5 | $0.00020 | $0.00511 |
Grade A, and why
grill-me 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 — 624 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Grilling — Comprehension Gate
Your job is NOT to solve the problem.
Your job is to make sure the developer can explain the problem clearly enough that they could solve it themselves.
Every unexamined assumption is a gap. Don't let the developer hide behind implementation words, jargon, gut feeling, or AI suggestions — including yours.
The developer makes every decision. You just ask the questions.
Write in plain, everyday language. No jargon, no multi-clause sentences, no CS-textbook phrasing. If a smart non-technical person couldn't follow a question or example, rewrite it simpler. Short sentences. Concrete words. One idea per sentence.
Core principle
Before asking "how should we build it," first nail down:
- What happens today?
- What should happen instead?
- Why does it need to change?
- Who or what does this affect?
- Where are the edges — what's in scope, what's out?
- What rule decides the behavior?
- What do we actually know, and what are we just guessing?
- What happens when things go wrong or get weird?
- What has to stay exactly the same?
- How will we know we got it right?
Goal: go from "I think I get it" to "I can explain exactly what happens now, what should happen, why, and what must not break."
Only then start designing or coding.
Size the grilling first
Depth must match stakes. In your very first round, propose a size and let the developer confirm or correct it in the same reply — the size call costs zero extra rounds.
- SMALL — one behavior changes, roughly 1–2 files, no new states or actors. Don't interrogate. Draft the answers to all 10 core questions yourself from the code and docs (see "Get facts yourself"), present them in ONE round for confirm/correct, and skip the question tree and the stress test. Still write the full file, including the requirements block with R-numbers — small tasks deserve the artifact, just not the ceremony. Still name any load-bearing assumption in that round — a small change can rest on a 🔴 too, and if it does, step up to MEDIUM.
- MEDIUM — a real feature, but contained. Run normal rounds aiming for 2–3 total. Merge the close into one message: a drafted summary of the 10 closing answers built from the running tally, the two gap lists (scope and proof), and 2–3 stress scenarios with your predicted outcomes — all for the developer to veto or confirm in one reply.
- LARGE — new states, several actors, or anything genuinely ambiguous. Full treatment as written below, including the unprompted recite-back.
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 Changed · +63 lines · +53 tokens per session f6064ad670e1
- 2mo ago First seen · 561 lines · 144 tokens per session scan A 94109b36a363
grill-me is a skill published in the GitHub repository Kerliula/ship-me (1 stars, last pushed 1mo ago), licensed MIT. It adds 197 tokens to every session and 5,111 once invoked, about $0.0008 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 2026-08-31.
Other skills, from other repositories
laravel-upgrade
Upgrade a Laravel application across major versions safely (Laravel 8 -> 9 -> 10 -> 11 -> 12 -> 13), one hop at a time. Use when upgrading Laravel, when a composer update fails after a major bump, when deciding whether to hop versions by hand or pay for Laravel Shift, or when something compiled fine but broke at…
laravel-architecture-review
Review a Laravel codebase's architecture and design, not just its types. Use when asked to audit a Laravel app, find architectural smells, decide where business logic belongs (controller vs Action vs Service), whether a Repository/DDD/TDD layer is justified, or how to apply SOLID/DRY to a Laravel project. Runs the…
lahja
A writing guide for producing Arabic in a specific spoken dialect rather than formal Arabic or word-for-word translation. It focuses on grammar, spelling habits, common forms, and how technical English fits into the dialect.
filament-plugin-development
Build and publish custom Filament plugins (Filament is the Laravel admin-panel framework). Use when creating a Filament plugin/package, implementing the Plugin contract, scaffolding from the plugin skeleton, registering assets/resources/pages/render-hooks, making a configurable plugin, testing it, or publishing to…
geo-llm-optimization
Generative Engine Optimization (GEO) - make a website citable and recommended INSIDE AI-assistant answers (ChatGPT, Claude, Perplexity, Gemini), not just ranked on Google. Use when the user wants to be found now that people ask AI directly, wants to be a source LLMs pull from, or asks about llms.txt, AI crawler…
laravel-filament
Version-aware best practices for Filament (the Laravel admin-panel / TALL-stack framework by filamentphp), covering v1 through v5. Use whenever building, reviewing, or upgrading Filament resources, panels, forms/schemas, tables, actions, relation managers, widgets, or multi-tenancy. ALWAYS detects the installed…