CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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.

65

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is agent-browser in vercel-labs/agent-browser

SKILL.md
Quality
Evals
Security

Quality

Content

75%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 well-organized, highly actionable CLI reference with a clear core workflow, worked examples, and properly split deep-dive documentation. Main issues: the three referenced template scripts are missing from the bundle, the 'state' command is used but undocumented in the command reference, and a few duplicated option entries could be trimmed.

Suggestions

Fix or remove the broken template references — templates/form-automation.sh, templates/authenticated-session.sh, and templates/capture-workflow.sh are linked but do not exist in the bundle.

Document the 'state save/load' command in the Commands section, since the authentication example depends on it but it appears nowhere in the reference.

Deduplicate global options that repeat in section bodies (--full, proxy flags, --cdp vs. connect) to tighten token usage.

DimensionReasoningScore

Conciseness

The body is dense, executable command syntax with no conceptual padding — it assumes Claude's competence throughout. Minor redundancy could be trimmed: '--full' appears in both the Global options and Screenshots sections, '--cdp 9222' duplicates the 'connect 9222' command, and proxy options are listed twice.

4 / 5

Actionability

Commands are copy-paste ready with worked examples covering common cases (form submission, authentication with saved state). One minor gap: 'state save auth.json' / 'state load auth.json' is used in the auth example but never documented in the Commands section, leaving the reader without the full state-management command surface.

4 / 5

Workflow Clarity

The four-step core workflow (open → snapshot -i → interact via @refs → re-snapshot after DOM changes) is clear, and examples include implicit validation (wait --load networkidle, then 'snapshot -i # Check result'). Validation is shown in examples rather than stated as an explicit checkpoint in the workflow itself, keeping it just below the top anchor.

4 / 5

Progressive Disclosure

Good structure with a well-signaled, one-level-deep references table whose five files all exist in the bundle. However, the three linked template scripts (templates/form-automation.sh, templates/authenticated-session.sh, templates/capture-workflow.sh) do not exist in the bundle, which is a broken reference that prevents a top score.

4 / 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.

A strong description that explicitly answers both what the skill does and when to use it, with concrete action lists and natural trigger phrases. Minor improvements possible: add common synonyms like 'web scraping' and mention additional capabilities (recording, PDF export) for fuller coverage.

Suggestions

Add 'web scraping' or 'scrape data from websites' to the trigger phrases — it is the most natural term users say for the data-extraction use case and is currently missing.

Consider mentioning recording or PDF export among the actions to better reflect the skill's actual capability set documented in the body.

DimensionReasoningScore

Specificity

The description lists several concrete actions — 'web testing, form filling, screenshots, and data extraction' plus 'navigate websites, interact with web pages' — in proper third-person voice, but coverage has minor gaps (no mention of recording, PDF export, cookies, or network mocking capabilities documented in the body).

4 / 5

Completeness

Both questions are explicitly answered: a clear 'what' ('Automates browser interactions for web testing, form filling, screenshots, and data extraction') followed by an explicit 'Use when the user needs to...' clause enumerating concrete trigger phrases — matching the top anchor's example pattern.

5 / 5

Trigger Term Quality

Natural phrases users would say are well covered ('fill forms', 'take screenshots', 'navigate websites', 'test web applications', 'extract information from web pages'), but common synonyms like 'web scraping', 'scrape', or 'browser automation' are missing.

4 / 5

Distinctiveness Conflict Risk

The browser-automation niche is clear with distinct triggers, but phrases like 'extract information from web pages' and 'test web applications' carry minor overlap risk with generic web-fetching or data-extraction skills.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 3 missing

Warning

Total

15

/

16

Passed

Repository

Kiz8-Team/open-agents

Path

.agents/skills/agent-browser/SKILL.md

Reviewed

Table of Contents

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.