CtrlK
BlogDocsLog inGet started
Tessl Logo

changeset

Command: changeset

49

Quality

55%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/changeset/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

The content is a strong, highly actionable rules skill: concrete templates, wrong/correct contrasts, and a pre-ship red-flags checklist make it easy to execute correctly. Its only real weakness is repetition in the 'relative to main' rule and a slightly long single-file layout.

DimensionReasoningScore

Conciseness

The body is mostly lean and imperative with good wrong/correct examples, but section 3 restates the same 'relative to main, never the last commit' directive at least three times ('NEVER write a changeset relative to the last commit...', 'Use main / origin/main as the release baseline', 'Write changesets for the user-visible delta from the current main branch') plus overlapping bullet lists — 'mostly efficient but could be tightened'. It is above 2 because there is no conceptual padding or explanation of things Claude already knows, and below 4 because the repetition in Rule 3 goes beyond minor trimming.

3 / 5

Actionability

Fully copy-paste-ready: complete templates for simple, API-change, and breaking changesets; wrong/correct frontmatter YAML; concrete file-path examples ('.changeset/core-fix-types.md'); before/after code examples; and runnable registry commands ('node tooling/scripts/generate-ui-changelog-entries.mjs --write'). Not below 5 because the common cases are all concretely covered.

5 / 5

Workflow Clarity

For this single-purpose skill the action is unambiguous, and it includes an explicit pre-ship validation checklist ('## Red Flags: Before shipping') plus a verification requirement ('check whether any named removed/renamed API actually exists on main before writing removal or migration prose') — a genuine feedback loop. It meets the 5 anchor's 'checklists for complex processes' and the simple-skill exception.

5 / 5

Progressive Disclosure

The body is well-sectioned and appropriately delegates registry detail to the sibling 'registry-changelog' skill rather than duplicating its schema, and there are no bundle files to organize (references/, scripts/, assets/ are absent). It is 4 rather than 5 because the ~230-line body is self-contained where a reference file could carry the template gallery, and the Registry Changelog guidance is split across two sections (Rule 4 and the trailing '## Registry Changelog'), a minor organization gap.

4 / 5

Total

17

/

20

Passed

Description

28%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.

The description is effectively just a title: it names the domain but communicates no capabilities, no usage triggers, and no scope. A user or model scanning skill descriptions would have almost nothing to decide whether to invoke it. The strong content body is badly undersold by this frontmatter.

Suggestions

Rewrite the description to state concrete actions, e.g. 'Create and edit .changeset/*.md files that document package changes (feat, fix, breaking) for release.'

Add an explicit 'Use when...' clause with natural trigger phrases: creating changeset files, documenting package changes, preparing a release, or version bump decisions for @platejs/* packages.

Add a boundary to reduce conflict risk, e.g. 'Not for commit messages, PR descriptions, or registry changelog entries.'

DimensionReasoningScore

Specificity

The description is 'Command: changeset' — it names the domain (changesets) but states zero concrete actions, matching the anchor 'names the domain but actions are minimal or generic'. It cannot score 3 because no action verbs or capabilities are listed at all, and it is above 1 only because 'changeset' is a concrete domain term rather than pure abstraction.

2 / 5

Completeness

It has an extremely vague 'what' (something about changesets) and no 'when'/usage guidance whatsoever, fitting 'has a vague what and no when'. It is not 1 because a specific domain is at least named, and it cannot be higher because neither the purpose nor any trigger conditions are explicit.

2 / 5

Trigger Term Quality

The only keyword is the bare word 'changeset' with no natural trigger phrases a user would say (e.g. 'write a changeset', 'release notes', 'version bump'), matching 'one or two generic keywords; missing the natural phrases users say'. It falls short of 3-4 because there are no synonyms, variations, or context terms at all.

2 / 5

Distinctiveness Conflict Risk

'Changeset' is a reasonably distinct niche term with limited conflict risk, but the description gives no scope boundaries, so it could overlap with adjacent skills (commit messages, PR descriptions, changelog generation) — 'somewhat specific but could still overlap with similar skills'. Not 4-5 because the description provides no distinguishing trigger phrasing to separate it from those neighbors.

3 / 5

Total

9

/

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

description_field

'description' is very short (18 chars), consider making it more detailed

Warning

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

13

/

16

Passed

Repository
udecode/plate
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.