agentleFS
HolojermHolojerm/drawthesystem/skills/scenario/SKILL.md

scenario

Skill0 starsChanged 2 months agoMITView on GitHub

Create a company-flavoured system design scenario + session folder (prompt, hidden interviewer notes, blank Excalidraw canvas). Usage — /scenario <company-slug|generic> [seed # | topic] [--depth|--breadth] [--minutes 45]

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. scenario-labs/skills★ 3.5ksame name
    2. ffroliva/gflow-cli★ 367same name
    3. caoergou/windows-xp★ 5same name
    4. b-urge/scenario-skill★ 1same name
    5. nguyenquocan0101/skills★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository8 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read companies/<slug>/profile.md (skip for generic) and progress.md (past sessions, weak dimensions).

    line 18

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/scenario

Turn a company profile into a concrete, timed practice scenario.

Inputs

  • <company-slug> — must match companies/<slug>/profile.md; or generic for a classic problem with no company flavour.
  • Optional seed number (from the profile's Scenario seeds) or free-text topic. If omitted, pick the seed that best matches the role and that the user hasn't done yet (check progress.md).
  • --depth (one component, implementation-level) or --breadth (whole system). Default: breadth for the first session with a company, then alternate.
  • --minutes N default 45.

Steps

  1. Read companies/<slug>/profile.md (skip for generic) and progress.md (past sessions, weak dimensions).
  2. Create sessions/<YYYY-MM-DD>-<slug>-<short-topic>/ containing:
    • prompt.md — what the candidate sees (see template below).
    • interviewer.md — hidden: expected clarifying questions & good answers, key numbers, the 2–3 "crux" decisions, expected deep-dive targets, follow-up probes ordered by difficulty, red flags, and a rubric emphasis line pointing at the user's historically weak dimensions. Do not show this file to the user during the session.
    • canvas.excalidraw — a starter canvas: title text + a small legend of colours (client / service / db / cache / queue / storage). Write the spec to canvas-spec.json in the session dir (title + a notes block listing the legend; nodes/edges empty) and generate with bun scripts/excalidraw.mjs <dir>/canvas-spec.json <dir>/canvas.excalidraw; verify with bun scripts/read-excalidraw.mjs. Keeping the spec lets /mock append to it if it has to transcribe for the candidate.
    • notes.md — empty template with headings: Requirements / Estimates / API & data model / High-level / Deep dive / Trade-offs / Open questions.
  3. Print the prompt and how to open the canvas:
    • Workbench (preferred): bun run web, then http://localhost:7788/sessions/<dir> — autosaves to canvas.excalidraw.
    • excalidraw.com: bun scripts/to-clipboard.mjs sessions/<dir>/canvas.excalidraw, Cmd+V, then Save to… back over canvas.excalidraw.
  4. Offer /mock (--voice if the workbench is open) now, or /critique later.

prompt.md template

# <Title>
_Company: <name> · Mode: breadth|depth · Time: N min · Level: senior/staff_

## Setting
2–4 sentences framing the problem in the company's world, using their real product surface (from the profile) but not requiring insider knowledge.

## The ask
One clear design task. E.g. "Design the system that lets a rider see live ETAs for nearby drivers."

## Given constraints
- 3–6 bullets: scale numbers grounded in the profile (cite "public numbers suggest ~X"), latency/consistency/regulatory needs.

## Deliverables (what a strong answer covers)
Requirements → estimates → API/data model → high-level diagram → 1–2 deep dives → trade-offs. Say which deep-dives are most likely to be probed *without* giving away the crux.

## Rules
Think aloud. Draw in Excalidraw. Ask clarifying questions — I will answer as the interviewer.

Quality bar

  • The scenario should feel like it came from this company's interview loop, not a generic textbook question.
  • Include at least one non-obvious constraint that forces a real trade-off.
  • Numbers must be plausible and consistent with the profile.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.