agentleFS
hushh-labshushh-labs/hushh-research/.agents/skills/mobile-plugin-contracts/SKILL.md

mobile-plugin-contracts

Skill40 starsChanged 4 days agoApache-2.0View on GitHub

Use when implementing or reviewing Capacitor plugin contracts, native plugin registration, or platform bridge alignment across iOS and Android.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Avoid overlap with mobile-parity-audit and frontend-architecture.

    line 12
  2. Distinctive:

    Verify the TypeScript bridge, plugin registration, and native implementation paths first.

    line 58
  3. Distinctive:

    Keep plugin method names and service-layer contracts aligned across platforms.

    line 59
  4. Distinctive:

    Treat service-layer native/web branching as part of the same contract, not a follow-up.

    line 60
  5. Very specific:

    If the request is still broad or ambiguous, route it back to mobile-native.

    line 69
  6. Very specific:

    If the task is parity or release-readiness rather than plugin contracts, use mobile-parity-audit.

    line 70
  7. Very specific:

    If the work becomes broad web-frontend or backend architecture, route to frontend or backend.

    line 71

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Hussh Mobile Plugin Contracts Skill

Purpose and Trigger

  • Primary scope: mobile-plugin-contracts
  • Trigger on Capacitor plugin contracts, native plugin registration, platform bridge alignment, and service-to-plugin interface work.
  • Avoid overlap with mobile-parity-audit and frontend-architecture.

Coverage and Ownership

  • Role: spoke
  • Owner family: mobile-native

Owned repo surfaces:

  1. hushh-webapp/lib/capacitor
  2. hushh-webapp/ios/App/App/Plugins
  3. hushh-webapp/android/app/src/main

Non-owned surfaces:

  1. mobile-native
  2. frontend
  3. backend

Do Use

  1. Creating or updating native plugins and their TypeScript bridge contracts.
  2. Plugin registration and method-name alignment across iOS, Android, and service-layer callers.
  3. Reviewing plugin boundary drift between native and web paths.

Do Not Use

  1. Broad native parity or release-readiness work.
  2. Broad frontend or backend intake.
  3. Cross-domain repo mapping.

Read First

  1. docs/development/guides/new-feature.md
  2. hushh-webapp/ios/App/App/MyViewController.swift
  3. hushh-webapp/android/app/src/main/AndroidManifest.xml
  4. hushh-webapp/docs/ios-native-surface-adoption.md

Workflow

For custom native presentation changes, follow the current renderer boundaries in hushh-webapp/docs/ios-native-surface-adoption.md. Prove a feasible whole-layer composition first; retained hosts and ordered bridge receipts cannot synchronize React and native animation frames. Choose one renderer before reveal and keep it for that presentation lifetime. Do not reopen historical Debug flags as adoption.

  1. Verify the TypeScript bridge, plugin registration, and native implementation paths first.
  2. Keep plugin method names and service-layer contracts aligned across platforms.
  3. Treat service-layer native/web branching as part of the same contract, not a follow-up.
  4. Separate retained presentation, current action binding, interaction/accessibility eligibility and privacy. Capability/layout receipts are admission evidence, not rendered transition acceptance; use the adoption reference's qualification rules.
  5. Treat keyboard movement as eligibility/geometry, not automatic host retirement. Validate deliberate focus cancellation separately from uncertainty; require fresh system geometry for keyboard acceptance. Refresh compatible admission without replacing unchanged presentation. Fence superseded layout failures as well as receipts; current failures remain fail-closed. Native drawers own complete motion locally; do not synchronize them with per-frame React poses. Historical pose protocols remain rollback-only, not product qualification.
  6. Characterize renderer bootstrap against the installed SDK's final WebView creation path. Configuration hooks can be overwritten before document creation; use a host regression plus real renderer admission, not merely a mocked eligibility check. See the adoption reference's bootstrap evidence.
  7. A private editor's final committed text can arrive after coverage revokes actions. Reconcile it through the editor's owner/revision fence without reviving actions, clearing a newer draft or logging its body. Acknowledgment of a trigger is not proof that its owned popup finished dismissal.
  8. Transient feedback over a native pane belongs to its acknowledged owned presenter and accessible subtree, never a window-level sibling. Preserve the shared notification/promise owner, distinguish pending admission from retirement, and bind expiry/dismissal to the captured form, destination and revision. Prove retained-host replacement and delivered touches outside the bounded banner; an isHittable receipt alone is insufficient. Follow the adoption reference's private feedback boundary.

Handoff Rules

  1. If the request is still broad or ambiguous, route it back to mobile-native.
  2. If the task is parity or release-readiness rather than plugin contracts, use mobile-parity-audit.
  3. If the work becomes broad web-frontend or backend architecture, route to frontend or backend.

Required Checks

cd hushh-webapp && npm run typecheck
cd hushh-webapp && npm run cap:build

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.