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 agentmods add skills/assisjp/specflow/grillnpx skills add assisjp/specflow --skill grillgit clone --depth 1 https://github.com/assisjp/specflowWhat 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 | $0.00121 | $0.01220 |
| Opus 5 | $0.00060 | $0.00610 |
| Sonnet 5 | $0.00024 | $0.00244 |
| Haiku 4.5 | $0.00012 | $0.00122 |
Grade A, and why
grill 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 2d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Grill
Interview the user relentlessly until you both reach a shared understanding they can build from. Walk down every branch of the decision tree, resolving dependencies between decisions one at a time.
The goal is not to be agreeable. It is to find the ambiguity, the contradiction, and the unexamined trade-off before they become code.
The loop
-
One question at a time. Asking several at once is bewildering and lets the hard one hide behind the easy ones. Ask, wait for the answer, then ask the next.
-
Recommend an answer to every question. Never ask a bare question. State your recommended choice and the reason, so the user is reacting to a proposal, not staring at a blank field. A question without a recommendation is you offloading the thinking you were asked to do.
-
Facts you look up. Decisions you ask. If something can be discovered by exploring the environment — filesystem, git, tools, docs, existing code — look it up instead of asking. Only genuine decisions, where the answer is the user's judgment and not a fact, get put to them. Asking the user a question you could have answered yourself wastes the session.
-
Resolve dependencies in order. Find the most upstream decision — the one the others hang off — and settle it first. Do not ask a downstream question whose framing depends on an answer you do not have yet.
-
Do not act until the user confirms shared understanding. Grilling produces alignment, not artifacts. It never writes source or tests — however settled the design feels, however broad the user's go sounds ("yes", "all of it", "go ahead"). Building is
spec-execution's work, and only after a spec exists. Withdocs, the artifacts you may write are exhaustivelyCONTEXT.mdand ADRs; nothing else. Present the resolved picture and hand off — the go you are waiting for is the go to the next step, never a licence to build.
The four sharpeners
These are what separate a real grilling from a polite Q&A. Apply all four throughout.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 67 lines · 121 tokens per session scan A d818b69ad4fe
grill is a skill published in the GitHub repository assisjp/specflow (17 stars, last pushed 23d ago), licensed MIT. It adds 121 tokens to every session and 1,220 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…