koubou
Skill1 starsChanged 16 days agoMITView on GitHub
Design and iterate on App Store screenshots with Koubou. Use when creating or redesigning store screenshots, framing existing app captures, or changing their copy, layout, device size, or localization.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- bitomule/Kouboutheir own version, first seen Oct 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- bitomule/Koubou★ 211same 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 repository57 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-loginSkill
- animateSkill
- app-intentsSkill
- app-store-connectSkill
- architectSkill
- arenaSkill
- blast-radiusSkill
- codebase-designSkill
- code-qualitySkill
- correctSkill
- diagnosing-bugsSkill
- domain-modelingSkill
- eli5Skill
- engineering-principlesSkill
- gauntlet-loopSkill
- gh-stackSkill
- goal-swarmSkill
- grillingSkill
- grill-with-docsSkill
- handoffSkill
- howSkill
- improve-codebase-architectureSkill
- ios-hapticsSkill
- mergeSkill
- orchestrationSkill
- prototypeSkill
- prSkill
- react-best-practicesSkill
- react-doctorSkill
- reframeSkill
- repo-cleanupSkill
- researchSkill
- retroSkill
- review-fix-loopSkill
- reviewSkill
- review-sweepSkill
- seoSkill
- setup-sreestackSkill
- show-dont-tellSkill
- showroomSkill
- snipsnap-do-choresSkill
- snip-snapSkill
- swift-testing-modernizationSkill
- swiftuiSkill
- tddSkill
- to-specSkill
- to-ticketsSkill
- triageSkill
- ui-designSkill
- uikitSkill
- verifySkill
- wayfinderSkill
- web-component-inspirationSkill
- whySkill
- writing-for-agentsSkill
- xcode-security-auditSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
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
- Distinctive: line 16
Read style intake. Inspect the app, captures, icon, assets, UI tokens, product docs, and existing marketing before asking for direction.
- Distinctive: line 27
Read the design guide. Draft the slide story and two or three copy options before CSS.
- Very specific: line 39
Use the project's existing Koubou installation, including a virtual environment when present.
- Distinctive: line 42
Read YAML configuration. Keep configuration, templates, and source captures in the app repo's established location.
- Distinctive: line 48
Use the returned screen bounds, margins, orientation, and canvas class.
- Distinctive: line 65
Prefer the host's shared browser tools for inspecting it.
- Distinctive: line 70
Read every generated slide at full resolution and thumbnail size.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Koubou
Build a screenshot campaign from the app's actual visual language and useful content. Use HTML/CSS templates, real device frames, and measured output. Each slide communicates one concrete benefit. User direction sets the scope and style; readability, honest product claims, and correct canvas geometry remain the quality bar.
Read the app and settle the story
- Read style intake. Inspect the app, captures, icon, assets, UI tokens, product docs, and existing marketing before asking for direction. Summarize the palette, contrast, density, typography, tone, and distinctive motifs that should carry into the campaign.
- Resolve only material uncertainty using the style interview. Reuse answers and references already supplied. Establish the audience, main benefit, requested slides, target devices, and any localization. When captures are missing or weak, capture useful app states with the project's documented run path and the host's available device tools. Check assets' redistribution terms and retain their required notices.
- Read the design guide. Draft the slide story and two or three copy options before CSS. Choose useful, realistic app content that makes each claim visible. Keep the app UI accurate.
- Record the campaign direction: product signals, style, copy voice, backgrounds, device composition, and variation across slides. Ground each choice in the app or user direction. For a full set, vary the first three compositions; repeated upright centered phones are a weak default. For step-by-step work, finish one requested slide and return its checked preview before expanding the set.
Prepare and compose
- Use the project's existing Koubou installation, including a virtual
environment when present. Follow setup; prepare HTML
support with
kou setup-html. Linking this skill does not install the CLI. - Read YAML configuration. Keep configuration, templates, and source captures in the app repo's established location; respect its ignored-output and machine-specific configuration rules.
- Select the real device with
kou list-frames, readproject.deviceandproject.output_size, and inspect its geometry before writing CSS:kou inspect-frame "<device>" --output-size <size> --output json. Use the returned screen bounds, margins, orientation, and canvas class. - Plan the text and device zones. Write HTML/CSS that fits the actual canvas,
copy length, and frame proportions. In HTML, annotate each headline, any
supporting copy, and the primary device when present with
data-kou-idanddata-kou-roleso the output can be measured. Use the design guide's canvas-specific typography and scale. For highlights, zoom callouts, gradients, or content-mode composition, consult capabilities. - For each checked export, use a fresh, empty, ignored
project.output_dir; preserve earlier accepted outputs. Scope a targeted export to the requested slides/languages. Generate withkou generate <config.yaml> --output jsonand retain stdout/stderr. Reject per-slide or per-language generation errors even if the CLI reports success. Verify every expected slide × language PNG and HTML sidecar was produced in this run; existing files alone are not proof (see YAML configuration). Usekou live <config.yaml>when the user wants to iterate in a live preview; read its asset-watch limits. Prefer the host's shared browser tools for inspecting it.
Check and iterate
Read every generated slide at full resolution and thumbnail size. Account for
every requested language and slide from the current export. For HTML outputs, inspect each sibling
*.layout.json, using layout_path when reported, for objective bounds,
proportions, and mathematical overlaps. Judge appearance from the image.
Content-mode YAML does not emit HTML annotation sidecars: verify its configured
positions and inspected frame geometry against each PNG, and report that
element-level sidecar measurements were unavailable. Choose HTML when that
measured verification is required.
Apply the design guide's rejection checklist to every requested output:
- Copy is specific, readable, truthful, and supported by the visible app state.
- Device frame and screen align; cropping is intentional and useful.
- Text fits its zone, stays legible, and avoids unintended device overlap.
- Empty space, visual hierarchy, and composition serve the slide's benefit.
- A set has deliberate variation while keeping the app's visual identity.
- HTML layout sidecars contain the required annotated elements; empty sidecars require an annotation fix before claiming measured verification.
Fix and regenerate failed slides before presenting them. A successful renderer exit is not design acceptance. Return the checked visual preview, full-size output paths, and any verification gap. For follow-ups, change only the scope requested and recheck the affected outputs. Uploading or publishing belongs to the user's authorized release workflow.
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
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.

