potion-yjs-dev-browser-test

Use when a Slate Yjs collaboration behavior is suspicious and Potion should be used as a live reference implementation through dev-browser, especially for offline/reconnect, selection, or history scenarios.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/potion-yjs-dev-browser-test/SKILL.md

Quality

Content

77%

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

The body is highly actionable and workflow-clear, with executable CDP code and explicit validation/recovery checkpoints for a fragile network-isolation procedure. The main gaps are minor redundancy between the workflow and repro template and a lack of file-level split for the reusable CDP pattern.

Suggestions

Consolidate the repro sequencing so it appears once (either the Workflow or the One-Off Repro Template) and have the other reference it, removing the overlap.

Move the 'Per-Page CDP Pattern' snippet and the repro template into a reference file (e.g. references/cdp-offline-pattern.md) signaled from the body, improving progressive disclosure for the moderately-sized skill.

Add a one-line preflight assertion in the template that confirms both pages reached '[data-slate-editor="true"]' before going offline, making the existing wait step an explicit checkpoint.

Dimension	Reasoning	Score
Conciseness	Lean and assumes competence (never explains what Yjs, Slate, or CDP are) and every section earns its place, but the Workflow step 6 and the 'One-Off Repro Template' overlap on the repro sequence, leaving it 'mostly efficient but could be tightened' rather than fully lean.	2 / 3
Actionability	Provides a fully executable per-page CDP code block with exact calls and parameters, a concrete endpoint (127.0.0.1:9222), named pages, and copy-paste-ready probe checks ('A fetch should succeed', 'B fetch should fail while offline').	3 / 3
Workflow Clarity	A clear numbered 7-step workflow plus an 8-step repro template, with explicit validation (network-isolation probe before editing), feedback loops (force online + retry once, resume from printed step on timeout), and a cleanup checkpoint ('Restore B online and detach CDP sessions before exiting').	3 / 3
Progressive Disclosure	Single self-contained file with well-labeled sections and no deep nesting, but at ~83 lines it exceeds the 'under 50 lines' simple-skill allowance and keeps reusable material (CDP pattern, repro template) inline rather than signaling one-level-deep reference files.	2 / 3
	Total	10 / 12 Passed

Description

57%

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 gives an explicit trigger and a highly distinctive niche, but it blends the capability statement into the 'when' clause and uses scenario names rather than a list of concrete actions. Tightening into a clean 'what it does' statement followed by 'Use when...' would raise specificity and completeness.

Suggestions

Lead with a crisp capability statement (e.g. 'Drives a persistent debug Chrome via dev-browser to replay Slate/Yjs collaboration repros against Potion as a live oracle.'), then add the 'Use when...' clause separately.

Expand trigger terms with common user phrasings such as 'Yjs sync', 'reconnect divergence', 'cursor/selection drift', and 'undo/redo history mismatch'.

Name concrete actions the skill performs (isolate one tab offline via per-page CDP, replay repro steps, compare convergence/selection/history) rather than only naming scenarios.

Dimension	Reasoning	Score
Specificity	Names the domain ('Slate Yjs collaboration', 'Potion as a live reference implementation through dev-browser') and scenario types ('offline/reconnect, selection, or history') but lists scenario triggers rather than multiple discrete concrete actions, matching the 'names domain and some actions but not comprehensive' anchor.	2 / 3
Completeness	Opens with an explicit 'Use when...' trigger and embeds the capability ('Potion should be used as a live reference implementation through dev-browser'), but the 'what' is woven into the 'when' clause rather than stated as a crisp standalone capability list, so it lands at the 'has what, but when is implied/blended' level rather than a clean explicit-both anchor.	2 / 3
Trigger Term Quality	Includes relevant natural terms a user in this niche would say ('Slate Yjs collaboration', 'offline/reconnect', 'selection', 'history', 'dev-browser') but leans on internal product names (Potion, Slate Yjs) and misses common phrasing variations, fitting the 'some relevant keywords but missing common variations' anchor.	2 / 3
Distinctiveness Conflict Risk	A clear niche with distinct triggers ('Slate Yjs collaboration behavior is suspicious', 'Potion... live reference', 'offline/reconnect, selection, or history') that is very unlikely to fire for an unrelated skill.	3 / 3
	Total	9 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/potion-yjs-dev-browser-test/SKILL.md
Commit: 45116fd

Reviewed: about 14 hours 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.