agentleFS
t1mericksont1merickson/skills/transitional-interfaces/SKILL.md

transitional-interfaces

Skill1 starsChanged 2 months agoMITView on GitHub

Design structural continuity between two defined interface states when identity, containment, position, or set membership must survive the change. Use for an actual state-transition design problem. Do not use for general navigation architecture, motion curves or animation implementation, isolated visual polish, or static screens whose relationship is already clear.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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
  1. Distinctive:

    Start with the truth the connection must carry:

    line 12
  2. Distinctive:

    If the states have no defensible relationship, motion that fabricates one is worse than a cut: the user spends attention decoding a …

    line 23
  3. Distinctive:

    Push/navigate: when the destination is a distinct place reachable through other routes.

    line 27
  4. Distinctive:

    Expand in place: when the detail belongs to the selected item and return to the set is the likely next move.

    line 28
  5. Distinctive:

    Sheet/popover: when the task concerns the item but its origin should remain visible and cheaply recoverable.

    line 29
  6. Distinctive:

    Do not gate response on a network round trip.

    line 35
  7. Distinctive:

    Place optimistic content only where it can remain, or avoid promising placement before the server decides order.

    line 36
  8. Distinctive:

    Use the project's existing primitives first: platform transitions, layout animation, shared-element APIs, or a measured FLIP implementation.

    line 44
  9. Distinctive:

    Do not specify milliseconds here.

    line 46

≈ “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.

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

Transitional interfaces

This skill decides one edge on the map from spatial-interfaces. A transition is not decoration applied after a state change; it is the channel that tells the user how the two states are related. Sometimes deciding the transition changes what the screens themselves must be. This skill does not choose curves, durations, or library syntax.

Name the relationship

Start with the truth the connection must carry:

RelationshipStructural choice
same object, more detailpreserve identity; expand or transform its container
container and contentsshow entering/leaving that container
peers on an ordered axiscarry direction and position among peers
member joining/leaving a setshow both membership and the set making/closing room
different contexts with no spatial relationfade through or cut without inventing continuity
genuinely unrelated statesa cut may be correct

If the states have no defensible relationship, motion that fabricates one is worse than a cut: the user spends attention decoding a connection that does not exist. If they are related and identity vanishes in one frame, the connection is underspecified.

Choose the structural form

  • Push/navigate when the destination is a distinct place reachable through other routes.
  • Expand in place when the detail belongs to the selected item and return to the set is the likely next move.
  • Sheet/popover when the task concerns the item but its origin should remain visible and cheaply recoverable.

For insertion, let the set make room and the member occupy it. For removal, let the member leave and then close the gap. These are two events—the set changing shape and the member arriving or leaving. They may overlap, but collapsing them into one simultaneous blur discards the half that explains the other and makes a membership change look like a re-sort.

Preserve truth under real conditions

  • Do not gate response on a network round trip. Reconcile optimistic state explicitly; failure should not masquerade as user removal.
  • Place optimistic content only where it can remain, or avoid promising placement before the server decides order.
  • Match skeleton geometry to the content that replaces it.
  • Reduced Motion keeps identity and containment in the resting states, using a calmer transition such as a crossfade.
  • Deep links, restoration, missed motion, and time away mean the destination still needs orientation and an exit in a screenshot. Continuity may replace redundant chrome, never necessary chrome.
  • Focus, accessibility announcements, and input ownership move with the visual state; continuity that strands keyboard or screen-reader users is not continuity.

Implement after the decision

Use the project's existing primitives first: platform transitions, layout animation, shared-element APIs, or a measured FLIP implementation. Check current support when choosing a web or native API; browser versions and framework capability belong in implementation research, not this stable skill.

Do not specify milliseconds here. Hand curve, spring, and reduced-motion implementation to the platform motion guidance, then verify the rendered interaction can be interrupted and does not delay input.

The current failure mode is not too little animation everywhere; it is motion everywhere, each piece defensible alone and exhausting in aggregate. Ask the functional question: does this transition communicate a true relationship? If not, it is embellishment on the critical path, and a cut is often more honest.

Output

State the two endpoints, their relationship, chosen structural form, identity/containment cue, enter and exit paths, arbitrary-entry fallback, reduced-motion equivalent, focus/accessibility handoff, data-latency behavior, and the implementation evidence still needed.

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.