CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-first-screenshots

Agent-first screenshots — an agent drives the real app via CDP and produces clean, defect-free product screenshots (newsletters, landing pages, social, decks, PR). Dual-channel verification (DOM + pixels + vision) in a capture loop. Use for any "take/redo screenshots of the app" task.

75

Quality

92%

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

SKILL.md
Quality
Evals
Security

Quality

Content

85%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 tight, expert-level skill body that pairs concrete thresholds and a copy-paste vision rubric with a clearly sequenced, validation-gated capture loop and well-organized references to real bundle scripts. Minor gains available from trimming manifesto framing and showing one inline script invocation example.

Suggestions

Trim the rhetorical framing in 'The Zero-Defect Bar' and 'The One Method' (e.g. 'Full stop.', 'the bar is not "the content is present"') to tighten conciseness without losing the rule.

Add one short inline example showing how to invoke capture-verify.mjs with a populated spec JSON, so the actionability is fully copy-paste ready end-to-end.

Consider moving the detailed vision-rubric JSON schema into a short reference file and linking to it, keeping the overview even leaner.

DimensionReasoningScore

Conciseness

Mostly lean and information-dense, assuming Claude knows CDP/sharp; minor manifesto framing ('the bar is not "the content is present"', 'Full stop.') and the 'Zero-Defect Bar'/'One Method' prose could be trimmed slightly without losing signal.

4 / 5

Actionability

Highly concrete — exact CDP metrics (1440x900 @ deviceScaleFactor:2), variance thresholds (>200, <50, bgRatio >0.97), and a copy-paste vision JSON schema — plus two real referenced scripts; the minor gap is no inline example of invoking the scripts with a populated spec.

4 / 5

Workflow Clarity

The capture loop is clearly sequenced (clean reload → navigate UI → leaf cleanup → three-channel verify → capture) with explicit validation checkpoints and a feedback loop ('If it fails: reload and redo', reject on any_obvious_defect, diagnose root cause, recapture), plus a failure-mode recovery table.

5 / 5

Progressive Disclosure

Single overview file with well-signaled one-level-deep references to two real bundle scripts (screenshot-verify.mjs, capture-verify.mjs); executable verification logic lives in the scripts while the overview stays navigable with clear section headers.

5 / 5

Total

18

/

20

Passed

Description

100%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, third-person description that crisply states what the skill does, how it does it (CDP + dual-channel verification), and exactly when to invoke it. Concrete artifact types and natural trigger phrasing make it both specific and easily triggered.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'drives the real app via CDP', 'produces clean, defect-free product screenshots', 'Dual-channel verification (DOM + pixels + vision) in a capture loop' — with comprehensive coverage of the artifact types (newsletters, landing pages, social, decks, PR).

5 / 5

Completeness

Explicitly answers both: what ('an agent drives the real app via CDP and produces clean, defect-free product screenshots ... Dual-channel verification') and when ('Use for any "take/redo screenshots of the app" task') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive natural terms users would actually say — 'screenshots', 'take/redo screenshots of the app', 'newsletters', 'landing pages', 'social', 'decks', 'PR' — with synonyms and concrete artifact categories.

5 / 5

Distinctiveness Conflict Risk

Clear niche (agent-driven product screenshots via CDP with three-channel verification) with minimal conflict risk; the body even distinguishes it from a sibling evidence skill ('fraimz').

5 / 5

Total

20

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
Devin-AXIS/iPolloWork
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.