CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

plate-plugins

Use before adding or changing a plugin's stage fields or options, such as read middleware, contributions, transfer vetoes or landing policy, and for a feature that spans Plite, Plate, the registry and docs. Load it before any edit to a Base plugin file under packages/platejs/src, such as packages/platejs/src/features/*/lib/Base*Plugin.ts, even when the task reads as Plite core, editor behavior or a bug fix. Builds Plate plugins and entrypoints with semantic ownership, inference, scoped capabilities and package proof, and delivers a feature across its package, UI, docs and release. Use it also when a change touches a plugin's Markdown or HTML encode or decode mapping or its preserve claims.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A dense, well-structured instruction skill with an explicit workflow, validation proofs, and clean reference routing. Its main weakness is bundle integrity: three of the referenced rule files that anchor the required reading are missing, so the navigation the body promises partially fails.

Suggestions

Add the missing ./rules/creation-flow.md, ./rules/typing.md, and ./rules/capabilities.md files (or correct their paths) so the Required Reads and Workflow step 1 are executable as written.

Add an explicit feedback loop after the step-7 proofs (e.g., "if a proof fails, fix and re-run before proceeding to step 8") to reach top workflow-clarity marks.

Inline one minimal definePlugin/toReactPlugin code sketch so the semantic-layer decision rules are actionable without first opening the reference files.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence throughout: no concept explanations, no library tutorials, and terse declarative rules like "Colocation is the default" and "Prove the smallest honest surface". Every token instructs, matching the 5 anchor rather than the 4 anchor, which still expects minor over-explanation to trim.

5 / 5

Actionability

Concrete, runnable guidance is present — exact commands ("pnpm --filter platejs build", "node tooling/scripts/check-plate-feature.mjs <plan>"), exact paths, and named APIs ("definePlugin", "toReactPlugin") — but the mechanics of authoring are deferred to reference files and no inline code examples are given. This sits between the 4 anchor (mostly executable, minor gaps) and the 5 anchor (copy-paste ready coverage of common cases), landing at 4.

4 / 5

Workflow Clarity

The 8-step Workflow is clearly sequenced with explicit validation checkpoints (step 7 lists typecheck, behavior tests, and a build with its failure rationale "TS4094 errors the typecheck skips") plus a step-8 audit checklist. It stops short of the 5 anchor because there is no explicit fix-and-re-validate feedback loop; it is above the 3 anchor since checkpoints are explicit rather than implicit.

4 / 5

Progressive Disclosure

Signaling is excellent ("Read this reference when…", one-level-deep references such as package-host.md and feature/manifest.md), but scored against the actual bundle: the required reads ./rules/creation-flow.md, ./rules/typing.md, and ./rules/capabilities.md do not exist, and ../verify/references/testing.md points outside the skill. Broken required-read navigation exceeds the 4 anchor's "minor organization gaps" while the clear structure and surfaced references keep it above the cluttered/buried pattern of lower anchors.

3 / 5

Total

16

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An exemplary description: it states the capability concretely and gives multiple explicit, file-path-level trigger conditions, including edge cases like tasks that merely read as core or bug-fix work. Density is high but every clause carries distinct trigger information rather than padding.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions and artifacts: "Builds Plate plugins and entrypoints", "adding or changing a plugin's stage fields or options, such as read middleware, contributions, transfer vetoes or landing policy", and "delivers a feature across its package, UI, docs and release". Coverage is comprehensive within its niche, so it matches the 5 anchor rather than the 4 anchor, which reserves room for gaps.

5 / 5

Completeness

It explicitly answers "what" ("Builds Plate plugins and entrypoints with semantic ownership, inference, scoped capabilities and package proof") and "when" through three concrete trigger clauses ("Use before adding or changing…", "Load it before any edit to a Base plugin file…", "Use it also when a change touches…"). Both halves are explicit with concrete trigger phrases, matching the 5 anchor rather than the 4 anchor's softer "when".

5 / 5

Trigger Term Quality

Natural trigger vocabulary is thorough and includes path-level triggers users would actually encounter: "Base plugin file under packages/platejs/src", "packages/platejs/src/features/*/lib/Base*Plugin.ts", "stage fields", "encode or decode mapping", "Plite, Plate, the registry and docs". No natural synonym or file-pattern variation is missing, which fits the 5 anchor rather than the 4 anchor's "a few natural terms missing".

5 / 5

Distinctiveness Conflict Risk

The niche is unambiguous: Plate/Plite plugin authoring scoped by exact file globs and plugin-specific terms ("stage fields", "transfer vetoes", "preserve claims"), so it is unlikely to fire for unrelated skills. This matches the 5 anchor's "clear niche with distinct triggers; minimal conflict risk" rather than the 4 anchor, which still allows overlap with closely related skills.

5 / 5

Total

20

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 5 missing, 4 deeper-than-1-level, 1 suspicious

Warning

Total

13

/

16

Passed

Repository

udecode/plate

Path

.agents/skills/plate-plugins/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.