offline-ux-kit
Skill0 starsChanged 2 months agoMITView on GitHub
Add local UX telemetry, categorized feedback and self-contained HTML reports to offline HTML or static web prototypes. Use for offline playtest logs, form usability tests and task paths; not native-app instrumentation or advertising analytics.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Fable-Forge/fableforge-agent-skillsword for word, first seen Sep 21, 2026
Other versions1 project uses a version of this.
Word for word · 1
- Fable-Forge/fableforge-agent-skills★ 1100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
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.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Read entrypoints, routing, completion handlers, storage and network calls.
- Distinctive: line 24
Read privacy-and-acceptance.md before shipping.
- Very specific: line 28
Use OfflineUX.report(run) and OfflineUX.reviewer().
- Very specific: line 32
Run node scripts/build.cjs, then node --test tests/core.test.cjs and python tests/package.test.py from the kit.
≈ “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
Offline UX Kit
Instrument an existing browser application without changing its business decisions. Reuse assets/offline-ux.js; do not regenerate the recorder from scratch.
Inspect and map
Read entrypoints, routing, completion handlers, storage and network calls. Preserve unrelated files. Identify the test variant: offline-scripted or local-html-with-ai. An HTML entrypoint alone does not mean offline.
Produce an event map: operation → stable task/step ID → exact call site → allowed metadata → completion/abandonment condition. Read integration.md and schema.md. Extend the metadata allowlist and tests together when an adapter needs new fields.
Integrate
Copy or inline the runtime. Obtain informed test consent and an anonymous code. Use a stable per-run session ID for refresh recovery, and a new ID for new tests. Never use names, emails or input text as identifiers.
Await OfflineUX.create() before instrumented interaction. Call the task, step, choice, back, reply and endTask methods that correspond to real transitions in the event map; do not emit artificial events for interactions the host does not have. data-ux-id captures explicit clicks and rapid-click signals; it does not infer task completion.
Wire an accessible feedback selector and finish/export button. Call endTask('completed') only at real success. Await finish() before report(). Keep raw JSON available. Show persistence warnings; never promise recovery in memory mode. Do not delete data after export.
Read privacy-and-acceptance.md before shipping. Do not collect input text, credentials, request bodies, URLs, error messages or imported business data. This release supports categorized feedback, not comments. Inventory host network calls; do not patch global fetch to fake offline safety.
Report
Use OfflineUX.report(run) and OfflineUX.reviewer(). Keep builds, batches and variants separate. The generic funnel is session-level ordered reach, not task conversion. Use task-run IDs for custom task-level conversion. Missing events are missing evidence, not zero failures.
Validate and deliver
Run node scripts/build.cjs, then node --test tests/core.test.cjs and python tests/package.test.py from the kit. Test the modified application separately in a real browser: completion, back/change, feedback, export, refresh, background timing, fallback, report import and hostile input. Automated tests do not validate browser persistence or visual rendering.
Deliver the modified app, reviewer, sample data and test results with unverified platforms clearly marked. Do not install, publish, upload logs or overwrite the original unless requested.
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.