web-browse

Browse and interact with web pages headlessly. Use when agent needs to navigate websites, click elements, fill forms, read content, or take screenshots.

Quality

81%

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

78%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 highly actionable, well-organized single-file CLI reference with strong examples and error recovery. Its main weakness is the absence of progressive disclosure — substantial reference sections that could be split into bundle files are inlined, inflating token usage.

Suggestions

Move the Action Reference, Snapshot Control, and Macros sections into separate reference files (e.g. ACTIONS.md, SNAPSHOT.md, MACROS.md) and link to them one level deep from SKILL.md to reduce the inline token footprint.

Trim redundancy such as the click-wait vs click --wait-stable restatement and consolidate the snapshot-transform flag descriptions.

Add an explicit validate-before-proceed checkpoint to the Workflow Pattern (e.g. snapshot to confirm state before interacting) to strengthen workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is a dense CLI reference with executable signatures and return shapes rather than concept lectures, so most tokens earn their place; minor redundancy (e.g., click-wait re-explained as click --wait-stable, lengthy snapshot-transform descriptions) could be trimmed, placing it just above the midpoint anchor.	4 / 5
Actionability	Every action ships a copy-paste-ready bash invocation with full argument hints, plus a worked Workflow Pattern and an error-recovery table — fully executable guidance covering the common cases.	5 / 5
Workflow Clarity	A clear navigate → snapshot → interact → verify sequence and a MANDATORY bot-detection recovery loop provide good sequencing and feedback, but the main workflow pattern is brief and lacks an explicit validate-before-proceed checkpoint, leaving it just under the top anchor.	4 / 5
Progressive Disclosure	The file is well-sectioned with clear headers, but at ~530 lines it inlines large blocks (full Action Reference, Snapshot Control, Macros) that belong in separate reference files, and there are no bundle files or one-level-deep references to offload them.	3 / 5
	Total	16 / 20 Passed

Description

83%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 concise, third-person, and explicitly pairs a concrete capability statement with a clear "Use when" trigger clause. It is strong overall, with only minor room to broaden trigger synonyms and tighten the generic opener.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("navigate websites, click elements, fill forms, read content, or take screenshots") but the generic opener "Browse and interact with web pages headlessly" and omission of deeper actions (extract, evaluate) leave minor coverage gaps, so it sits between the 3- and 5-anchors.	4 / 5
Completeness	Clearly answers both: the "what" ("Browse and interact with web pages headlessly") and the "when" via an explicit "Use when agent needs to ..." clause with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Includes natural user phrases like "navigate websites", "click elements", "fill forms", "take screenshots" — good coverage — but misses common synonyms such as "scrape", "browser automation", or "headless browser" that a user might also say.	4 / 5
Distinctiveness Conflict Risk	Headless browser automation with these specific triggers is a fairly distinct niche with low conflict risk, but there is minor overlap with static web-fetch skills, keeping it just below the clearest-niche anchor.	4 / 5
	Total	17 / 20 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
skill_md_line_count	SKILL.md is long (535 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: ekipo-user/web-ctl
Path: skills/web-browse/SKILL.md
Commit: e00f409

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.