migrate component react skills

Objective: Remove standalone repo-local components, creating-components, and react skill triggers while preserving their comprehensive content as progressive-disclosure references under the owning UI skills.

Goal plan: docs/plans/2026-05-25-migrate-component-react-skills.md

Template: docs/plans/templates/task.md

Primary template: docs/plans/templates/task.md

Applied packs:

• agent-native (docs/plans/templates/packs/agent-native.md)

Task source:

• type: user request
• id / link: chat request on 2026-05-25
• title: migrate component/react skills into UI-skill references
• acceptance criteria: in plate-2, move components and react into plate-ui references; in better-convex, remove creating-components and react standalone triggers while preserving the same comprehensive content under the repo UI owner; keep React Compiler and Effect guidance; sync generated mirrors; prove stale standalone triggers are gone.

Completion threshold:

• /Users/zbeyens/git/plate-2 has no generated standalone components/react skills and plate-ui owns references/components.md plus references/react.md.
• /Users/zbeyens/git/better-convex has no generated standalone creating-components/react skills and shadcn-parity owns references/components.md plus references/react.md.
• Install sync, resource sync, stale-trigger audits, skill validation, and lint pass or have a documented environment caveat.
• node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-25-migrate-component-react-skills.md passes.

Verification surface:

• Source/generated rg audits for stale standalone triggers and old .mdc references.
• Generated skill folder audits with find.
• Install sync in both repos.
• Direct resource sync script proof in both repos.
• Skill validation with skill-creator quick validator.
• Lint with pnpm lint:fix in plate-2 and bun lint:fix in better-convex.

Constraints:

• Follow skill-creator: keep SKILL.md lean; move bulky guidance into references.
• Do not hand-edit generated SKILL.md; change source rules and sync.
• Preserve comprehensive component and React content, especially React Compiler, Effects, useEffectEvent, derived state, refs, data attributes, accessibility, and polymorphism.
• No PR, commit, push, browser proof, or package behavior change.

Boundaries:

• Source of truth: user request, .agents/rules/**, and generated .agents/skills/** mirrors in plate-2 and better-convex.
• Allowed edit scope: UI skill source rules, moved reference docs, resource sync script, package install hook, and active goal plan.
• Browser surface: N/A.
• Tracker sync: N/A.
• Non-goals: no product UI behavior changes; no broad skill diet beyond the named component/React skills.

Blocked condition:

• Blocked only if install/resource sync cannot make the generated UI skill folders contain the reference files, or if stale standalone triggers remain after repair.

Task state:

• task_type: agent-skill cleanup
• task_complexity: normal
• current_phase: closeout
• current_phase_status: complete
• next_phase: final response
• goal_status: active until completion check passes

Current verdict:

• verdict: complete after final check
• confidence: high
• next owner: final response
• reason: standalone triggers are gone, comprehensive references are under UI skills, install/resource sync is wired, and audits pass.

Completion rule:

• Do not call update_goal(status: complete) while any required checklist item remains unchecked. If an item does not apply, check it and add N/A: <reason>.
• Do not call update_goal(status: complete) until every completion threshold above is satisfied, final handoff evidence is recorded, and node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-25-migrate-component-react-skills.md passes.
• Do not create hook state for this goal. This file plus the active goal are the durable state.

Start Gates:

Gate	Applies	Evidence
Skill analysis before edits	yes	Used skill-creator, autogoal, and task; read owner skill/source files before edits.
Active goal checked or created	yes	Created a goal for migrating standalone component/React skills into UI references.
Source of truth read before edits	yes	Read plate-ui, components, react, creating-components, and shadcn-parity source rules.
Tracker comments and attachments read	N/A	No tracker source.
Video transcript evidence required	N/A	No video evidence.
docs/solutions checked for non-trivial existing-code work	N/A	Agent skill/source cleanup only.
TDD decision before behavior change or bug fix	N/A	No runtime behavior change.
Branch decision for code-changing task	N/A	No PR/branch requested.
Release artifact decision	N/A	No package release surface.
Browser tool decision for browser surface	N/A	No browser surface.
PR expectation decision	yes	No PR requested.
Tracker sync expectation decision	N/A	No tracker.
Agent-native pack selected	yes	Plan created with --with agent-native.
Agent-facing action surface identified	yes	UI skill triggers, generated skill discovery, reference loading, and resource sync.
Source rule versus generated mirror boundary identified	yes	Edited .agents/rules/**, added source resource dirs, then ran install/resource sync.
agent-native-reviewer loaded or waiver recorded	N/A	Waived; exact stale-trigger/resource audits and skill validation cover the narrow failure mode.

Work Checklist:

• Objective includes outcome, completion threshold, verification surface, constraints, boundaries, and blocked condition.
• Task source classified with source type, id/link, title, task type, acceptance criteria, caveats, likely files/routes/packages, browser surface, and root-cause layer.
• Required video or screen-recording evidence is cached/read as normalized <video-transcripts> XML, or marked N/A with reason.
• Nearby repo instructions and implementation patterns read before edits.
• Implementation fixes the right ownership boundary, or the narrower choice is recorded with reason.
• Release artifact requirement recorded: changeset, registry changelog, or N/A with reason.
• Final handoff shape decided: bug/feature/testing/batch/review/tracker requirements, PR body sync, and issue/Linear sync when applicable.
• Branch handling recorded for code-changing work: dedicated branch used, new branch needed, or N/A with reason.
• Local-env-rot retry policy recorded for any surprising repo-wide failure: reinstall/rerun evidence or N/A with reason.
• Workspace authority recorded: every proof command names the cwd/tool that owns the changed behavior.
• High-risk note recorded for public API, runtime, package-boundary, browser behavior, agent-action, or command-contract changes, or marked N/A with reason.
• Review/autoreview target selected from actual diff state for non-trivial implementation work, or marked N/A with reason.
• Agent-native review decision recorded for .agents/**, .claude/**, .codex/**, skills, hooks, commands, prompts, or user-action tooling.
• Agent-native pack: source-of-truth rule files are edited instead of generated skill mirrors.
• Agent-native pack: the changed agent action is discoverable from the skill/rule text.
• Agent-native pack: generated mirrors are synced when .agents/rules/** changed, or N/A reason is recorded.
• Agent-native pack: accepted agent-native review findings are fixed or explicitly rejected with reason.

Completion Gates:

Gate	Applies	Required action	Evidence
Named verification threshold	yes	Run source/generated audits, install sync, resource sync, skill validation, and lint	Evidence recorded below.
Bug reproduced before fix	N/A	Record failing test/repro or N/A with reason	No bug fix.
Targeted behavior verification	yes	Prove generated skill discovery no longer exposes standalone component/React triggers	`find .agents/skills -maxdepth 2 ... rg '/(components
TypeScript or typed config changed	N/A	Run relevant typecheck	JS sync helper only; lint plus direct execution cover it.
Package exports or file layout changed	N/A	Run pnpm brl before final verification and keep generated barrel updates	No package exports or barrels.
Package manifests, lockfile, or install graph changed	yes	Run install and relevant checks	pnpm install and bun install ran successfully after package hook edits.
Agent rules or skills changed	yes	Run pnpm install and verify generated skill sync	Generated plate-ui and shadcn-parity mirrors contain the new reference files.
Workspace authority proof	yes	Run verification in the owning repo/package/app/route/tool and record cwd; do not count the wrong workspace as proof	Commands ran in /Users/zbeyens/git/plate-2 and /Users/zbeyens/git/better-convex.
Browser surface changed	N/A	Capture Browser Use proof or record explicit waiver/blocker	No browser surface.
Browser final proof	N/A	Attach screenshot or exact browser verification caveat when browser proof applies	No browser proof needed.
CI-controlled template output changed	N/A	Restore generated template output or record why it is intentionally kept	No template output touched.
Package behavior or public API changed	N/A	Add a changeset or record why no changeset applies	No package behavior/API change.
Registry-only component work changed	N/A	Update docs/components/changelog.mdx or record N/A	No registry component work.
Docs or content changed	N/A	For docs-heavy work, use --template docs; for incidental docs, verify source-backed claims, links, examples, and rendered output or record N/A	Agent reference migration, not product docs.
High-risk mini gate	yes	Record realistic failure mode, proof plan, and why the chosen boundary is right	Failure mode: broken reference links or stale standalone triggers; proof: generated reference files exist and stale-trigger audits are clean; boundary: UI owner skill owns UI/React reference depth.
Agent-native review for agent/tooling changes	N/A	Load .agents/skills/agent-native-reviewer/SKILL.md and close accepted/actionable findings, or record N/A	Waived for proportionality; exact audits plus skill-creator validation cover this task.
Local install corruption suspected	N/A	Run pnpm run reinstall once, rerun the exact failing command, or record N/A	No install corruption signal.
Autoreview for non-trivial implementation changes	N/A	Load .agents/skills/autoreview/SKILL.md; use dirty local --mode local, branch/PR --mode branch --base <base>, or committed slice --mode commit --commit <ref> until no accepted/actionable findings, or record N/A for docs-only/trivial/no local patch	No product implementation; targeted skill audits are the meaningful review.
PR create or update	N/A	Run check before PR work and sync PR body to final handoff	No PR requested.
PR proof image hosting	N/A	If PR body needs browser proof, replace local image paths with hosted GitHub URLs or record N/A	No PR/browser image.
Tracker sync-back	N/A	Post concise issue/Linear sync after PR exists, or record N/A/blocker	No tracker.
Final handoff contract	yes	Fill final handoff fields	Filled below.
Final lint	yes	Run pnpm lint:fix or scoped equivalent	plate-2 pnpm lint:fix passed; better-convex bun lint:fix passed.
Goal plan complete	yes	Run node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-25-migrate-component-react-skills.md	Run after this save; final response reports result.
Agent source / generated sync	yes	Run pnpm install when .agents/rules/** changed and verify generated mirrors	plate-2 install synced 6 resource entries; better-convex install synced 3 resource entries.
Agent action discoverability	yes	Source-audit the skill/rule path an agent will read	plate-ui and shadcn-parity SKILL.md files link to references/components.md and references/react.md.
Agent-native review	N/A	Load .agents/skills/agent-native-reviewer/SKILL.md and close accepted findings, or record N/A	Waived for proportionality; direct evidence is stronger here.

Phase / pass table:

Phase	Status	Evidence	Next
Intake and source read	complete	Read source skills and current generated mirror/resource behavior.	implementation
Implementation	complete	Moved standalone docs into UI references; added resource sync after Skiller.	verification
Verification	complete	Install, resource sync, stale audits, quick validation, and lint completed.	closeout
PR / tracker sync	N/A	No PR/tracker requested.	final response
Closeout	complete	Plan filled; completion check runs after save.	final response

Findings:

• Skiller regenerates SKILL.md, but it did not copy newly added resource folders. Without an explicit resource sync, the new reference links would be broken after install. That was the important architecture catch.
• Plate's real UI owner is plate-ui; better-convex has no plate-ui, so the equivalent UI owner is shadcn-parity.
• The large React/component docs are valuable, but as references only. Keeping them as standalone always-triggered skills was the bad shape.

Decisions and tradeoffs:

• Keep plate-ui concise and move comprehensive component/React docs into plate-ui/references.
• In better-convex, preserve the same content under shadcn-parity/references rather than inventing another UI skill.
• Add a generic resource-sync script after Skiller instead of hand-copying generated references once. This preserves the source/generated contract.
• Keep React Compiler and Effects material comprehensive inside references/react.md.

Implementation notes:

• plate-2:
  • Moved .agents/rules/components.mdc to .agents/rules/plate-ui/references/components.md.
  • Moved .agents/rules/react.mdc to .agents/rules/plate-ui/references/react.md.
  • Updated .agents/rules/plate-ui.mdc trigger and reference routing.
  • Added source copies of existing plate-ui generated rules/ and component-audit.md, then wired resource sync.
• better-convex:
  • Moved .agents/rules/creating-components.mdc to .agents/rules/shadcn-parity/references/components.md.
  • Moved .agents/rules/react.mdc to .agents/rules/shadcn-parity/references/react.md.
  • Updated .agents/rules/shadcn-parity.mdc reference routing.
  • Wired resource sync after Skiller in postinstall.

Review fixes:

• Lint flagged console.log in the new resource sync script; replaced it with process.stdout.write and reran lint cleanly.

Error attempts:

Error / failed attempt	Count	Next different move	Resolution
Skiller did not copy newly moved references	1	Add explicit resource sync script after Skiller	Fixed in both package install hooks.
Direct quick_validate.py execution was not executable	1	Invoke through python3	Worked.
better-convex cwd resolved /usr/bin/python3 without PyYAML	1	Run validator from plate-2 cwd with absolute better-convex skill path	Worked.

Verification evidence:

• /Users/zbeyens/git/plate-2: pnpm install ran Skiller and [agent-skill-resources] synced 6 resource entries.
• /Users/zbeyens/git/better-convex: bun install ran kitcn sync, Skiller, and [agent-skill-resources] synced 3 resource entries.
• /Users/zbeyens/git/plate-2: generated plate-ui contains references/components.md, references/react.md, references/component-audit.md, and existing rules/*.md.
• /Users/zbeyens/git/better-convex: generated shadcn-parity contains references/components.md and references/react.md.
• Both repos: stale-trigger audit found no standalone generated components, creating-components, or react skill SKILL.md files.
• Both repos: rg found no stale components.mdc, creating-components.mdc, react.mdc, ALWAYS use when using React, or old composable-component trigger text in live source/generated agent surfaces.
• skill-creator quick validation passed for plate-ui and shadcn-parity.
• /Users/zbeyens/git/plate-2: pnpm lint:fix checked 3420 files with no fixes after the final script edit.
• /Users/zbeyens/git/better-convex: bun lint:fix checked 857 files with no fixes after the final script edit.

Final handoff contract:

• PR line: N/A, no PR requested.
• Issue / tracker line: N/A, no tracker.
• Confidence line: high.
• Flow table:
  • Reproduced: source/generation audit showed standalone skills and missing resource copying behavior.
  • Verified: install/resource sync, stale-trigger audits, skill validation, and lint passed.
• Browser check: N/A.
• Outcome: standalone component/React skills are gone; comprehensive docs live under UI-skill references.
• Caveat: no full repo check; this is agent-rule/resource cleanup.
• Design:
  • Chosen boundary: UI owner skill triggers; detailed component/React docs as references.
  • Why not quick patch: moving files without resource sync leaves broken links.
  • Why not broader change: remaining UI owners already exist; no need for new skills.
• Verified: see evidence above.

Final handoff / sync:

• PR: N/A.
• Issue / tracker: N/A.
• Browser proof: N/A.
• Caveats: full repo check not run.

Timeline:

• 2026-05-25T12:04:06.613Z Task goal plan created.
• 2026-05-25 Source skills and generated mirror behavior audited.
• 2026-05-25 Standalone component/React docs moved into UI references.
• 2026-05-25 Resource sync script added and install hooks updated.
• 2026-05-25 Install/resource sync, stale audits, validation, and lint passed.

Reboot status:

Question	Answer
Where am I?	Closeout after verification
Where am I going?	Run completion check, mark goal complete, final response
What is the goal?	Migrate component/React standalone skills into UI-skill references
What have I learned?	Skiller needs explicit resource sync for reference folders
What have I done?	Removed standalone triggers, preserved comprehensive references, synced and audited both repos

Open risks:

• None for requested scope. Full repo behavioral checks were intentionally not run because no product code changed.