cli-anything-browser

Browser automation CLI using DOMShell MCP server. Maps Chrome's Accessibility Tree to a virtual filesystem for agent-native navigation.

Quality

68%

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

Fix and improve this skill with Tessl

tessl review fix ./browser/agent-harness/cli_anything/browser/skills/SKILL.md

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 a strong, actionable CLI reference with copy-paste examples and clear sequences. Main improvements are removing the JSON/daemon-mode duplication and adding an explicit error-recovery feedback loop.

Suggestions

De-duplicate the JSON output and daemon-mode coverage (each appears in both Usage Examples and Agent-Specific Guidance) by keeping one canonical section and cross-referencing it.

Add a short validate→fix→retry loop for failed commands (e.g. on 'MCP call failed': verify extension is running, then retry) to raise workflow clarity.

Clarify the install path for `cd browser/agent-harness` (state what repo/directory it is relative to) so the setup step is unambiguous.

Dimension	Reasoning	Score
Conciseness	The body is an efficient command reference with copy-paste examples and no padding of concepts Claude already knows; the only tightening needed is minor redundancy where JSON output and daemon mode each appear in two sections.	4 / 5
Actionability	Fully executable, copy-paste-ready commands cover the common cases (navigation, search/click, form fill, JSON output, daemon mode, REPL) with concrete path syntax and explicit error-handling commands.	5 / 5
Workflow Clarity	Multi-step sequences are clear (daemon mode is numbered start→run→stop) with an is_available() preflight checkpoint and a concrete error list, but there is no explicit validate→fix→retry feedback loop for failed operations.	4 / 5
Progressive Disclosure	No bundle files exist; the single SKILL.md is well-organized with clear section headers and no nested references, and keeping a CLI reference inline is reasonable — only minor gaps versus ideal content splitting.	4 / 5
	Total	17 / 20 Passed

Description

53%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 conveys a clear, technically specific niche but lacks any 'when to use' trigger guidance and leans on jargon over natural user phrasing. Adding a 'Use when…' clause with everyday trigger terms would lift completeness and trigger-term quality.

Suggestions

Append a 'Use when…' trigger clause, e.g. 'Use when navigating web pages, clicking elements, or filling forms via the browser from the command line.'

Replace some jargon ('Accessibility Tree', 'agent-native') with natural terms users actually say ('navigate web pages', 'click page elements', 'browser automation').

Enumerate 1-2 concrete filesystem-style actions (e.g. 'ls, cd, cat, and click page elements') to raise specificity from generic 'automation/navigation' to concrete capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain ('Browser automation CLI') and a concrete mechanism ('Maps Chrome's Accessibility Tree to a virtual filesystem for agent-native navigation'), but the actions themselves (automation, navigation) are generic and no command-level actions are enumerated — between anchor 2 (minimal actions) and anchor 3 (1-2 concrete actions).	3 / 5
Completeness	It clearly states what the skill does, but there is no 'Use when…' clause or equivalent explicit trigger guidance, capping completeness at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	'Browser automation' and 'navigation' are relevant natural terms, but the description leans heavily on jargon ('DOMShell MCP server', 'Accessibility Tree', 'agent-native') and omits common synonyms users would say (navigate web pages, click, scrape, control browser).	3 / 5
Distinctiveness Conflict Risk	The DOMShell + Accessibility-Tree-to-VFS niche is clearly distinct with minimal conflict risk, but the generic 'Browser automation CLI' opener and absence of explicit trigger phrases leave minor overlap risk with general web-automation skills rather than a fully distinct trigger surface.	4 / 5
	Total	13 / 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: HKUDS/CLI-Anything
Path: browser/agent-harness/cli_anything/browser/skills/SKILL.md
Commit: 34f5195

Reviewed: 1 day 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.