agent-browser

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

82%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 body is an exemplary dense command reference: fully executable, well-structured, and supported by copy-paste worked examples. Its main weaknesses are minor redundancy across sections, absent verification steps for destructive/batch operations, and a templates table whose linked scripts are missing from the bundle.

Suggestions

Add a verification step (e.g. re-snapshot or get text) after destructive/batch commands like 'cookies clear' and 'storage local clear' so the workflow_clarity feedback loop covers them.

Either add the missing templates/ scripts (form-automation.sh, authenticated-session.sh, capture-workflow.sh) referenced in the 'Ready-to-use templates' table, or remove that table to avoid broken links.

De-duplicate overlapping options (e.g. --cdp vs connect, --full, --json) by keeping each in one canonical section and cross-referencing rather than restating.

Dimension	Reasoning	Score
Conciseness	The body is a lean command reference with inline comments and no concept over-explanation, but it is long and somewhat repetitive (e.g. --full, --cdp/connect, and --json each appear in multiple sections) and could be tightened.	4 / 5
Actionability	Nearly every line is an executable command with concrete flags, and the two worked examples (form submission, authenticated session with saved state) are copy-paste ready and cover the common cases.	5 / 5
Workflow Clarity	A clear 4-step core workflow (Navigate, Snapshot, Interact, Re-snapshot) is reinforced by examples with wait/re-snapshot checkpoints, but batch/clearing operations like 'cookies clear' and 'storage local clear' lack explicit verification steps.	4 / 5
Progressive Disclosure	The overview links five real, one-level-deep reference files in a well-signaled table, but the templates table points to templates/*.sh paths that do not exist in the bundle, a broken navigation link.	4 / 5
	Total	17 / 20 Passed

Description

88%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 strong: it pairs a concrete capability list with an explicit, natural-language 'Use when' trigger clause covering the common cases. It is mostly distinct and comprehensive, with only minor broadness in the 'what' framing and some repetition across clauses.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions in both clauses ('web testing, form filling, screenshots, and data extraction' and 'navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information'), giving comprehensive coverage of the domain.	5 / 5
Completeness	Clearly states what the skill does ('Automates browser interactions for...') and explicitly when to use it ('Use when the user needs to navigate websites...'), answering both with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural user phrases ('navigate websites', 'fill forms', 'take screenshots', 'test web applications', 'extract information') with good synonym coverage, but no distinguishing extension/token and several terms repeat across the what and when clauses.	4 / 5
Distinctiveness Conflict Risk	The trigger list carves a clear browser-automation niche, but the 'Automates browser interactions' framing is somewhat broad and could overlap with general web/search skills.	4 / 5
	Total	18 / 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: 3 missing	Warning

	Total	15 / 16 Passed

Repository: freekmurze/dotfiles
Path: config/claude/skills/agent-browser/SKILL.md
Commit: a7899fe

Reviewed: 3 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.