Tessl • The management layer for AI agent skills

The body is an exceptionally dense, well-structured instruction set: concrete tool calls, explicit validation and error-recovery loops, and appropriate externalization of API detail to a single well-signaled reference. Its only minor gap is that common-case call signatures live in the reference file rather than inline.

Dimension	Reasoning	Score
Conciseness	Every sentence carries an environment-specific constraint or instruction ("The visible host has one page per session", "Observations emit directly; do not print them again", "The REPL supports standard JS, not Node modules, shell commands, fetch or timers") with no explanation of concepts Claude already knows and no padding. Not score 4 because no over-explanation was found that could be trimmed.	5 / 5
Actionability	Concrete, executable guidance is given throughout: a first-cell binding (`var tab = await browser.getTab()`), specific calls with arguments (`tab.getAXState()`, `tab.getScreenshot({ref})`, `fullPage: true` with `nextCursor` pagination). Not score 5 because full call signatures and inputs are deferred to the reference file rather than covered inline for the common cases.	4 / 5
Workflow Clarity	"Observe, act, verify" is a numbered five-step sequence with explicit validation ("Verify the outcome, URL and page identity before reporting success") and feedback loops for error recovery ("On stale_ref, observe again", "On stale_web_tool, list again", inspect whether a timed-out action's effect already happened). Not a destructive/batch skill, so no cap applies.	5 / 5
Progressive Disclosure	A single, clearly signaled one-level-deep reference ("Exact methods, inputs, outputs and platform limits are defined in [API reference](references/api.md)" — verified as a real 57-line file with no nested references) externalizes the bulk API detail, while the body stays a well-sectioned overview of behavior and constraints. Not score 4 because organization and signaling show no notable gaps.	5 / 5
	Total	19 / 20 Passed

Description

71%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 lean, concrete description that clearly states what the skill does and is well-scoped to a distinct niche. Its main weakness is the missing explicit "when to use" trigger clause, which caps completeness and limits trigger-term coverage.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to navigate, click, or type in the visible Agent browser pane, or to take a screenshot of a page."

Include natural synonyms users would say, such as "web page", "open a URL", or "scroll", to broaden trigger-term coverage.

Clarify what "debug" means as an action (e.g., "read page state and console output") so every listed capability is concrete.

Dimension	Reasoning	Score
Specificity	The description lists five concrete actions ("navigate, click, type, screenshot, debug") that comprehensively cover the browser-control domain, matching the top anchor's breadth (e.g., "Extract text and tables from PDF files, fill forms, merge documents"). Not score 4, as no meaningful coverage gaps exist within its domain.	5 / 5
Completeness	The "what" is clear and concrete (control the browser: navigate, click, type, screenshot, debug), but there is no "Use when..." clause or equivalent trigger guidance; "Needs the Browser setting" states a prerequisite, not a usage trigger, so completeness is capped at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	"browser", "navigate", "click", and "screenshot" are natural phrases users would say, giving good keyword coverage. Not score 5 because common variations and synonyms ("web page", "URL", "scroll", "browse", "open a site") are missing.	4 / 5
Distinctiveness Conflict Risk	"the user's visible Agent browser" carves a clear niche distinct from general web skills, with only minor overlap risk against standalone browser-automation tools. Not score 5 because "navigate, click, screenshot" could also match a generic browser-automation skill's triggers.	4 / 5
	Total	16 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: CherryHQ/cherry-studio
Path: resources/skills/cherry-browser/SKILL.md
Revision: 5078c86

Reviewed: 18 days ago

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.

cherry-browser

Table of Contents

Is this your skill?