dev-browser

Fallback browser automation with persistent Chrome state. Use only when Browser Use is unavailable or blocked.

Quality

65%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

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

Quality

Content

67%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 highly actionable with concrete executable commands and a clear fallback workflow featuring real validation checkpoints. Its main weakness is redundancy between the Plate Defaults and Rules sections, which inflates tokens without adding guidance.

Suggestions

Merge the overlapping guidance from "Plate Defaults" and "Rules" (persistent debug browser, dedicated --user-data-dir, profile clone, macOS open -na) into one section to remove near-duplicate tokens.

Add an explicit post-clone validation step before launching the debug browser (e.g., confirm the copied profile opens and 9222 reports a healthy /json/version) to strengthen the destructive clone workflow.

Trim the repeated "dev-browser --connect http://127.0.0.1:9222" mentions to a single canonical statement plus the failure-fallback variant.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with executable commands, but the "Plate Defaults" and "Rules" sections restate the persistent-debug-browser, dedicated --user-data-dir, profile-clone, and macOS open -na guidance nearly verbatim, adding redundancy that could be tightened, fitting the 3 anchor.	3 / 5
Actionability	It provides concrete copy-paste-ready commands throughout (npm install, dev-browser --connect, curl health check, python Local State mapping, rsync clone, open -na launch, lsof), with only minor gaps such as omitted failure-case commands, matching the 4 anchor.	4 / 5
Workflow Clarity	The sequence is clear with explicit validation checkpoints (curl /json/version health check, websocket URL fallback, lsof to identify a wrong 9222 owner), but the destructive rsync --delete profile clone lacks an explicit post-clone validation step, fitting between 3 and 5 at 4.	4 / 5
Progressive Disclosure	No bundle files exist and the skill is a single well-organized file with clearly labeled sections (Installation, Plate Defaults, Rules, Preferred Shape, Google Chrome Path), giving good structure with only minor overlap between sections, matching the 4 anchor for a simple single-file skill.	4 / 5
	Total	15 / 20 Passed

Description

62%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 clearly states both what the skill does and the specific fallback condition under which to use it, which is its main strength. It is held back by limited trigger-term breadth and a single-action specificity rather than a list of concrete capabilities.

Suggestions

Expand the "what" with 2-3 concrete capabilities (e.g., drive a persistent debug Chrome via CDP, reuse a signed-in profile, run page scripts) to lift specificity toward 5.

Add more natural trigger phrases users would say (e.g., "when you need a persistent Chrome session", "debug Chrome on 9222", "CDP automation") to broaden trigger-term coverage.

Keep the fallback guard, but pair it with positive triggers so the skill surfaces when the user wants persistent Chrome automation, not only when the primary tool fails.

Dimension	Reasoning	Score
Specificity	"Fallback browser automation with persistent Chrome state" names the domain and a concrete capability (persistent Chrome state) but lists only one primary action rather than multiple specific actions, fitting the 3 anchor.	3 / 5
Completeness	It states both a clear "what" ("Fallback browser automation with persistent Chrome state") and an explicit "when" ("Use only when Browser Use is unavailable or blocked"), but the "when" is a guarded fallback condition rather than broad concrete trigger phrases, fitting between 3 and 5 at 4.	4 / 5
Trigger Term Quality	It includes "Use only when Browser Use is unavailable or blocked" which provides a trigger, but the natural keyword coverage is limited to "browser automation" and "Chrome" with no synonyms or user-natural phrasing variations, matching the 3 anchor.	3 / 5
Distinctiveness Conflict Risk	The explicit "Use only when Browser Use is unavailable or blocked" framing carves a clear niche as a fallback path with minimal conflict risk, though "browser automation" broadly still has minor overlap with sibling skills, matching the 4 anchor.	4 / 5
	Total	14 / 20 Passed

Validation

81%

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 — 13 / 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
relative_links	Relative link issues: 1 suspicious	Warning

	Total	13 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/dev-browser/SKILL.md
Commit: cce36d3

Reviewed: 10 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.