agent-browser

Use the agent-browser CLI to exercise web interfaces, inspect rendered accessibility state, verify interactions, and capture screenshots. Use for browser-based UI testing and evidence collection. Do not use as a substitute for deterministic Playwright coverage when an automated regression test is the requested deliverable.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%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 tight, well-structured instruction skill whose verification loop is explicit and validation-rich. The main gap is that it intentionally routes CLI syntax to an external guide rather than inlining copy-paste-ready commands, capping actionability at 4.

Suggestions

Inline one or two fully-formed agent-browser command examples (e.g. opening a named session and taking an accessibility snapshot) so the core loop is executable without first running the guide-lookup command.

Tighten the opening paragraph to avoid restating the description and save tokens.

Dimension	Reasoning	Score
Conciseness	Mostly lean and dense with earned, non-obvious operational guidance (re-snapshot after navigation, wait-for-selector over fixed delays, inspect screenshots before reporting); the opening sentence slightly restates the description and a few lines could be trimmed.	4 / 5
Actionability	Provides concrete executable commands (agent-browser skills get core --full, the dev-bypass open URL, agent-browser console/errors/network requests), but deliberately defers most CLI syntax to the version-matched guide rather than showing copy-paste examples for each loop step.	4 / 5
Workflow Clarity	A clearly sequenced 7-step verification loop with explicit validation checkpoints (wait then re-snapshot, verify observable result/console/network, inspect the saved image before reporting success), plus error-recovery guidance via console/errors checks.	5 / 5
Progressive Disclosure	Under 50 lines, single-purpose, with no bundle files and no need for them; content is organized into well-signaled sections (Verification loop, Interaction guidance, Evidence) and delegates detailed syntax one level out to the CLI's own skills mechanism.	5 / 5
	Total	18 / 20 Passed

Description

87%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 strong, specific description with explicit what/when guidance and a useful negative boundary. Its only real weakness is the imperative second-person voice, which the rubric penalizes on specificity.

Suggestions

Rewrite in third person ('Exercises web interfaces, inspects accessibility state, verifies interactions, and captures screenshots') to avoid the second-person specificity penalty.

Add a couple of natural trigger synonyms users might say (e.g. 'end-to-end UI testing', 'click through') to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('exercise web interfaces, inspect rendered accessibility state, verify interactions, and capture screenshots') which would anchor at 5, but the imperative second-person voice ('Use the agent-browser CLI') triggers the -1 specificity penalty.	4 / 5
Completeness	Clearly answers both what (exercise/inspect/verify/screenshot) and when ('Use for browser-based UI testing and evidence collection'), plus a negative boundary clause, matching the anchor for explicit what-and-when with concrete triggers.	5 / 5
Trigger Term Quality	Good natural keyword coverage ('browser-based UI testing', 'web interfaces', 'screenshots', 'accessibility state') with an explicit 'Use for' trigger, though a few synonyms users might say (e.g. 'click', 'end-to-end') are absent.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (the agent-browser CLI for rendered verification) with distinct triggers and an explicit non-use boundary against Playwright regression coverage, giving minimal conflict risk.	5 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: emdash-cms/emdash
Path: skills/agent-browser/SKILL.md
Commit: 9ca2de5

Reviewed: 3 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.