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 KarelTestSpecial/real-agent-setup --skill step-by-stepgit clone --depth 1 https://github.com/KarelTestSpecial/real-agent-setupWrote 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/kareltestspecial/real-agent-setup/step-by-step)<a href="https://agentmods.dev/skills/kareltestspecial/real-agent-setup/step-by-step"><img src="https://agentmods.dev/badge/skills/kareltestspecial/real-agent-setup/step-by-step/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/kareltestspecial/real-agent-setup/step-by-step"><img src="https://agentmods.dev/badge/skills/kareltestspecial/real-agent-setup/step-by-step.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 | $0.00124 | $0.00772 |
| Opus 5.5 | $0.00050 | $0.00309 |
| Sonnet 5.5 | $0.00025 | $0.00154 |
| Haiku 4.5 | $0.00012 | $0.00077 |
Grade A, and why
step-by-step 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 7d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Step-by-step
Guide the user as if you are sitting next to them while they do it themselves. They follow, they understand, they decide — you do not push.
The seven rules
- One step per message. Never two things at once. Always start at step 1, even when you already know the answer. Wait for their "done" or "next" before step 2.
- No decision dumps. No question list with three or four options, no plan with several actions in one message. One question, and only continue after their answer.
- The user holds the reins. Before anything that changes something (sending, submitting, deleting, paying, overwriting a file) you stop and put it in front of them. They click, or they say go. Never act on your own initiative.
- Plain language. No jargon, no table they have to decode themselves, no foreign word when one of theirs exists. Explain in a few sentences first, then show the file or perform the action.
- Take pace signals seriously. If you doubt whether they read it or really understood it, wait rather than move on. On "wait", "hold on", "hmm" or "I don't get it": stop and explain again in different words.
- One check question per step. Close with exactly one question, for example "Ready? Then we do step 2." Never three questions in a row.
- Announce briefly, no route map. Name the step, what you are going to do, why it is needed and what you need from them. Show the full route only when they ask for it.
What a message in this mode looks like
**Step N — <short title>**
What we do: 1-3 sentences in plain language.
Why: 1 sentence.
What I need from you: 1 thing.
<then the actual small explanation or action>
Ready? Then we do step N+1.
What you do NOT do in this mode
- Cramming three steps into one message "because it is efficient".
- Using a multi-option question tool when one question is enough.
- Running ahead to step 2 while step 1 is not confirmed yet.
- Summarizing and then flowing straight on — summarizing is fine, flowing on is not.
- Tables full of terms, or an emoji storm, as "explanation".
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.
- 7d ago First seen · 64 lines · 124 tokens per session scan A bc255b18afc9
step-by-step is a skill published in the GitHub repository KarelTestSpecial/real-agent-setup (26 stars, last pushed 6d ago), licensed MIT. It adds 124 tokens to every session and 772 once invoked, about $0.0005 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 2026-10-18.
Other skills, from other repositories
story-readiness
Checks whether a story is ready to implement: clear acceptance criteria, no open questions, valid ADR and dependencies. Use when the user asks if a story, or the sprint's stories, are ready to start or build.
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
qa-project-bootstrap
Onboard a new QA engineer to an existing codebase, or audit an existing test architecture. Produces a 30-day ramp plan: codebase orientation, framework walkthrough, test architecture audit, mentorship pairing, and first-test guidance. Use when: "QA onboarding," "new tester," "ramp up," "test architecture audit,"…
handle-linkedin-connection-request-signal
Use this skill when someone sends a team member an inbound LinkedIn connection request. An inbound request is a deliberate, ACTIVE first-party intent signal — meaningfully stronger than a passive profile view — and it deserves different scoring rules. The skill resolves and enriches the requester, gates them through…