CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-browser

Browser automation through the repo agent-browser CLI. Explicit helper for navigation, forms, screenshots, scraping, and web-app checks. Prefer Browser Use or Playwright when available. Do NOT load for: sharing URLs, embedding links, or editing screenshot files.

70

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

The canonical home for this skill is agent-browser in Chachamaru127/claude-code-harness

SKILL.md
Quality
Evals
Security

Quality

Content

72%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with concrete commands and a recommended snapshot workflow, and it uses progressive disclosure well by pointing to two real reference files. Its main weaknesses are inline duplication of reference material and the absence of explicit validation/error-recovery checkpoints in the workflow.

Suggestions

Add explicit validation checkpoints to the workflow (e.g. after login/form submission, snapshot and confirm expected URL or element before proceeding) and a brief retry-on-failure loop for auth flows.

Trim the Quick Reference command tables, which overlap heavily with references/browser-automation.md, to reduce token cost and duplication; keep only the highest-frequency commands inline.

Consolidate the 'MCP browser tool usage' and 'Notes' sections or move them into a reference, since they are secondary to the core workflow.

DimensionReasoningScore

Conciseness

The body is mostly efficient and avoids explaining generic concepts, but it duplicates a large command reference inline (Quick Reference tables) that largely repeats content delegated to the references files, which could be tightened.

2 / 3

Actionability

Every step is concrete and executable: exact commands (`agent-browser snapshot -i -c`, `agent-browser click @e2`), real example output with refs, and copy-paste-ready command tables.

3 / 3

Workflow Clarity

Steps are clearly sequenced (Step 0 through Step 3) and snapshot-before-act is emphasized, but there are no explicit validation/verification checkpoints or error-recovery feedback loops for risky operations (e.g. login/auth, form submission), capping clarity at 2 per the destructive/batch-operations note.

2 / 3

Progressive Disclosure

The SKILL.md is a clear overview with two well-signaled, one-level-deep references (browser-automation.md, ai-snapshot-workflow.md) that both exist as real files in ./references/, and detail is appropriately split rather than deeply nested.

3 / 3

Total

10

/

12

Passed

Description

100%

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 specific, third-person, and clearly delineates both the skill's capabilities and its trigger/non-trigger conditions, with strong natural-language trigger coverage. It is concise without padding and directly answers both 'what' and 'when'.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 'navigation, forms, screenshots, scraping, and web-app checks' — plus an explicit tool name (agent-browser CLI), matching the anchor for multiple specific concrete actions.

3 / 3

Completeness

It states what the skill does (browser automation via agent-browser CLI) and when to use it (an explicit 'Use when'-style guidance: 'Prefer Browser Use or Playwright when available' plus 'Do NOT load for: ...'), explicitly answering both what and when.

3 / 3

Trigger Term Quality

Natural user phrases are well covered in both Japanese (「ページを開いて」「スクリーンショットを撮って」) and English ('open this page', 'click on', 'fill the form', 'screenshot'), giving good coverage of terms users would actually say.

3 / 3

Distinctiveness Conflict Risk

A clear niche (agent-browser CLI automation) with explicit anti-triggers ('Do NOT load for: sharing URLs, embedding links, or editing screenshot files') and a fallback recommendation, making conflicts with other skills unlikely.

3 / 3

Total

12

/

12

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.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 missing, 2 deeper-than-1-level

Warning

Total

15

/

16

Passed

Repository
Chachamaru127/claude-code-harness
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.