A lean, highly actionable release-prep runbook: specific paths, naming conventions, a verbatim scan command, and a mid-workflow validation loop over changelog Unreleased sections. Its main gaps are the absence of a final verification step and a couple of steps that stop just short of copy-paste commands.

Suggestions

Add a final validation step at the end of the workflow (e.g., re-run the rg Unreleased scan, confirm every changelog has a `## {new_version}` section, and verify the package builds/tests pass) to close the workflow-clarity gap.

Make 'Run pub get in /client dir' executable by stating the exact command (e.g., `cd client && dart pub get` or `flutter pub get`) as is already done for the rg scan.

Consider surfacing one verbatim changelog-entry example or pulling item-format detail into the existing write-changelog-entry reference more explicitly, so the drafting steps are as copy-paste ready as the rest of the skill.

Dimension	Reasoning	Score
Conciseness	The body is dense, imperative instruction with zero concept explanation — every bullet states an operation ('Set new version in packages/flet/pubspec.yaml', 'Pull the latest main and create a new branch named prepare-release-{new_version}'). Repeated mentions of the two sibling skills are inline reminders at point of use, so they earn their tokens; not verbose enough to drop to 4.	5 / 5
Actionability	Guidance is concrete: exact file paths (packages/flet/pubspec.yaml, website/sidebars.yml, sdk/python/packages/*/CHANGELOG.md), a copy-paste `rg` scan command, a branch naming pattern, and a fixed section ordering. Not a 5 because a few steps stop short of executable commands — 'Run pub get in /client dir' does not specify the exact command (dart vs flutter pub get), and the changelog item-drafting steps delegate without a verbatim example.	4 / 5
Workflow Clarity	The sequence is clear (version → branch → pubspec → pub get → changelogs → website docs → Unreleased conversion → sorting) and includes real checkpoints: the rg scan across all changelogs with an explicit remediation ('convert that section... Do not leave duplicate release content') and the milestone verify-and-fix loop. Not a 5 because there is no final end-to-end validation of the release state (e.g., verify the build passes, changelog format check, or milestone audit completion).	4 / 5
Progressive Disclosure	No bundle files (references/, scripts/, assets/) exist and none are needed: the body is a self-contained ~58-line document organized into Inputs / Related Skills / Steps. The only external links are one-level-deep, clearly signaled references to sibling skills (../flet-deprecation/SKILL.md, ../write-changelog-entry/SKILL.md), matching the simple-skill exception in the rubric notes.	5 / 5
	Total	18 / 20 Passed

Description

82%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.

A concise, well-formed description with an explicit 'Use when' trigger, third-person voice, and a clearly distinguishable Flet-specific niche. Its only weakness is thin action coverage — naming just two of the release-prep tasks the body actually performs.

Dimension	Reasoning	Score
Specificity	The description names the domain ('prepare new Flet release') and two concrete actions ('bumping versions and author release notes'), matching the 'names domain and 1-2 concrete actions' anchor. It is not a 4 because coverage is limited to those two actions — changelog assembly, milestone updates, and website docs from the body are absent.	3 / 5
Completeness	It explicitly answers both questions: what ('prepare new Flet release by bumping versions and author release notes') and when ('Use when asked to prepare new Flet release') with a concrete trigger phrase. Not a 4 because the 'when' clause is fully explicit, not merely implied or in need of more specificity.	5 / 5
Trigger Term Quality	Natural user phrases are present — 'asked to prepare new Flet release', 'bumping versions', 'release notes' — which a user would plausibly say verbatim. Not a 5 because a few natural synonyms ('cut a release', 'publish Flet', 'changelog') are missing.	4 / 5
Distinctiveness Conflict Risk	'Flet release' is a clear niche with distinct triggers; no other generic release/versioning skill would be wrongly triggered by this phrasing. Minimal conflict risk, matching the top anchor.	5 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 suspicious	Warning

	Total	15 / 16 Passed

Repository: flet-dev/flet
Path: .agents/skills/prepare-flet-release/SKILL.md
Commit: 9748de3

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.