resilient-localization
Skill4 starsChanged 23 days agoMITView on GitHub
Harden interfaces against slow networks, unexpected data, long text, localization, right-to-left layouts, and recoverable failures.
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Use when preparing a UI for real-world data, multiple languages or locales, slow or unreliable networks, permissions, asynchronous content, …
- Distinctive: line 20
Use stable message keys and complete messages rather than concatenating sentence fragments.
- Distinctive: line 24
Check right-to-left layouts when in scope.
- Distinctive: line 32
Check null and missing fields, unusually long strings, unexpected units, large and negative values, empty arrays, duplicate items, stale …
- Distinctive: line 36
Use intrinsic sizing and flexible constraints instead of fixed heights that clip translated text.
- Distinctive: line 44
Test with at least one long-string scenario, an alternate locale if supported, empty and malformed data, slow or failed requests, and …
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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
Resilience and localization
When to use
Use when preparing a UI for real-world data, multiple languages or locales, slow or unreliable networks, permissions, asynchronous content, varied input lengths, or production release hardening.
Workflow
1. Inspect the supported environment
Determine supported locales, date and number formats, text direction, browsers, input modes, network expectations, permissions, and data contracts. Reuse existing localization and error-handling architecture instead of introducing another layer.
2. Keep content localizable
Use stable message keys and complete messages rather than concatenating sentence fragments. Support placeholders, plural rules, contextual translator notes, and locale-aware formatting for dates, times, numbers, and currency. Never assume translated strings have the same length as English.
3. Support direction and expansion
Check right-to-left layouts when in scope. Prefer logical CSS properties such as inline-start and block-end when appropriate. Test long labels, compound words, different scripts, larger text settings, and missing translations. Avoid embedding essential text in images.
4. Model failure and delay
For asynchronous operations, define idle, loading, success, empty, stale, offline, partial, error, and retry states when relevant. Preserve safe user input across failure. Provide cancellation or retry when feasible. Do not show fake progress, fabricate data, or imply completion before the operation succeeds.
5. Harden against unpredictable data
Check null and missing fields, unusually long strings, unexpected units, large and negative values, empty arrays, duplicate items, stale records, expired sessions, permission denial, and partial responses. Truncate visually only when the full value remains discoverable where important; never silently destroy user input.
6. Make layouts resilient
Use intrinsic sizing and flexible constraints instead of fixed heights that clip translated text. Ensure tables, sidebars, dialogs, toasts, and navigation can accommodate expansion. Avoid layout shifts as fonts and media load. Keep essential actions available at narrow widths and with zoom.
7. Protect user trust
Explain what can be recovered and what was saved. Use precise timestamps or freshness labels when relevant. Confirm only real outcomes. Log technical details appropriately without exposing secrets or unnecessary personal data in user-facing messages.
8. Validate conditions
Test with at least one long-string scenario, an alternate locale if supported, empty and malformed data, slow or failed requests, and responsive layouts. Test RTL only when the product supports it. Clearly mark environments and cases that were not tested.
Quality gates
- Copy and formats use the project's localization conventions.
- Long translations and RTL behavior are considered when in scope.
- Network, permission, empty, stale, and failure states reflect actual system status.
- Unexpected and missing data does not break or misrepresent the interface.
- User input is preserved across recoverable errors where safe.
- Layouts remain usable at narrow widths and increased text sizes.
- Sensitive technical details are not exposed to end users.
- Verification claims match the environments and scenarios actually tested.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- max-designSkill
- accessibility-auditSkill
- data-visualizationSkill
- design-reviewSkill
- design-systemSkill
- forms-uxSkill
- frontend-designSkill
- information-architectureSkill
- interaction-motionSkill
- max-designSkill
- ui-testingSkill
- ux-writing-contentSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.

