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 klimentij/klimkit --skill klimkit-grill-megit clone --depth 1 https://github.com/klimentij/klimkitWrote 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/klimentij/klimkit/klimkit-grill-me)<a href="https://agentmods.dev/skills/klimentij/klimkit/klimkit-grill-me"><img src="https://agentmods.dev/badge/skills/klimentij/klimkit/klimkit-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/klimentij/klimkit/klimkit-grill-me"><img src="https://agentmods.dev/badge/skills/klimentij/klimkit/klimkit-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 | $0.00097 | $0.01022 |
| Opus 5 | $0.00048 | $0.00511 |
| Sonnet 5 | $0.00019 | $0.00204 |
| Haiku 4.5 | $0.00010 | $0.00102 |
Grade A, and why
klimkit-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 9d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Klimkit Grill Me
Interview the user relentlessly about every important part of the plan until both sides can see the same shape of the decision. Walk down the design tree branch by branch, resolve dependencies between decisions one at a time, and provide your recommended answer for each question.
Keep the session direct and human. This is a working conversation, not a form: ask the question you would actually ask if you were trying to save the plan from a hidden flaw, a vague assumption, or a weak tradeoff. The Klimkit addition is that every approved decision gets a short written trail.
Workflow
- Resolve the active work folder. If the repo has no
docs/work/layout, useklimkit-setupfirst. - Create or reuse a grilling phase folder inside the current work folder under
docs/work/, for exampledocs/work/015-140126-checkout-redesign/002-150126-grilling/. - Create or reuse a number-prefixed agent-authored Markdown note in that folder, for example
001-grilling-session.md. - Start the note with a short title, the source request, and the current decision target.
- Prepare under the hood before asking:
- Inspect the relevant codebase, docs, task notes, and existing decisions.
- Search the web for current outside context when market, product, API, framework, legal, pricing, or best-practice facts could affect the answer.
- Draft a private top-question list, estimate each question's importance, and rank contradictions or complete requirement gaps first.
- Before asking, write a compact
Question Triagegrid into the grilling-session note with up to 10 considered questions:- one-sentence TLDR of each candidate question.
- importance estimate such as
critical,high,medium, orlow. - one-sentence reason for the ranking.
- mark the question chosen now and why it is top priority at this moment.
- Ask exactly one question at a time: the current highest-importance question from the ranked list.
- For each question, include your recommended answer unless repo/web research or user context makes that premature.
- After the user answers, append a short tracking memo to the grilling-session note:
Q:the question asked.Approved decision:the user's decision, or your concise restatement if the user approved your recommendation.Open follow-up:only when the answer leaves a real unresolved branch.
- Rebuild and rerank the top-question list after every answer, using the new state and any new facts discovered from the repo or web; refresh the
Question Triagegrid before the next question so the user can audit what was considered and why. - Continue until the remaining branches are low-risk, explicitly deferred, or converted into concrete tasks.
- End with a compact decision summary and the path to the grilling-session note.
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.
- 9d ago First seen · 73 lines · 97 tokens per session scan A fc3c0adda1f2
klimkit-grill-me is a skill published in the GitHub repository klimentij/klimkit (7 stars, last pushed 29d ago), licensed MIT. It adds 97 tokens to every session and 1,022 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
LinkedIn automation via the Linked API CLI - fetch profiles, search people and companies, send messages, manage connections, create posts, react, comment, and run Sales Navigator and custom workflows. Use when the user wants to interact with LinkedIn.
gws-modelarmor
Google Model Armor: Filter user-generated content for safety.
persona-event-coordinator
Plan and manage events — scheduling, invitations, and logistics.
persona-researcher
Organize research — manage references, notes, and collaboration.
persona-team-lead
Lead a team — run standups, coordinate tasks, and communicate.
recipe-save-email-attachments
Find Gmail messages with attachments and save them to a Google Drive folder.