core/capabilities/review/visual-review/SKILL.md
Evaluates UI implementation by analyzing screenshots at multiple breakpoints against the spec, design conventions, and web standards. Checks layout, hierarchy, responsiveness, spacing, and accessibility. Use after implementing UI components, after a redesign, when the user says "how does it look", "check the design", "visual review", or "review the UI".
npx skillsauth add xoai/sage visual-reviewInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Evaluate the visual implementation using real screenshots — not code inspection, not assumptions. Screenshots are evidence. Code can look correct and render wrong.
Core Principle: Visual quality is verified visually. Reading JSX tells you what the DOM structure is. Screenshots tell you what the USER sees. This skill bridges the gap between "the code is correct" and "the page looks right."
After implementing UI components in BUILD or ARCHITECT mode. Specifically:
Do NOT use for:
Run the screenshot tool at the standard breakpoints:
bash .sage/tools/sage-screenshot.sh http://localhost:3000 \
--output .sage/work/<feature>/screenshots \
--label <feature-name> \
--full-page \
--after
If "before" screenshots exist from before the implementation, the tool will
have captured them with --before. Both sets enable comparison.
If the dev server isn't running, start it first:
npm run dev &
sleep 5 # wait for server
# capture screenshots
kill %1 # stop server
For each screenshot (mobile 375px, tablet 768px, desktop 1440px), evaluate:
If both "before" and "after" screenshots exist:
Output a structured visual review to .sage/work/<feature>/visual-review.md:
# Visual Review: [feature name]
**Date:** [timestamp]
**Breakpoints reviewed:** 375px, 768px, 1440px
## Summary
[1-2 sentence overall assessment]
## Mobile (375px)
**Status:** PASS / ISSUES FOUND
[Specific findings with reference to what's visible in screenshot]
## Tablet (768px)
**Status:** PASS / ISSUES FOUND
[Specific findings]
## Desktop (1440px)
**Status:** PASS / ISSUES FOUND
[Specific findings]
## Before/After Comparison
[If available — specific improvements and regressions]
## Issues Found
1. [Issue]: [description] — [severity: critical/major/minor]
2. ...
## Recommendation
PASS — ready to merge
PASS WITH NOTES — merge but address [minor issues] in follow-up
FAIL — [critical issues] must be fixed before merge
Show the report to the user. If issues are found, present the screenshot alongside the issue description so the user can see what you're referring to.
MUST (violation = missed visual bugs):
SHOULD (violation = incomplete review):
MAY (context-dependent):
npm run dev and tell me when it's ready."tools
Captures agent mistakes, corrections, and discovered gotchas so they are not repeated. Use when: (1) a command or operation fails unexpectedly, (2) the user corrects the agent, (3) the agent discovers non-obvious behavior through debugging, (4) an API or tool behaves differently than expected, (5) a better approach is found for a recurring task. Also searches past learnings before starting tasks to avoid known pitfalls. Activate alongside the sage-memory skill — they share the same MCP backend but serve different purposes (sage-memory = codebase knowledge, sage-self-learning = agent mistakes and gotchas).
development
Typed knowledge graph stored in sage-memory. Use when creating or querying structured entities (Person, Project, Task, Event, Document), linking related objects, checking dependencies, planning multi-step actions as graph transformations, or when skills need to share structured state. Trigger on "remember that X is Y", "what do I know about", "link X to Y", "show dependencies", "what blocks X", entity CRUD, cross-skill data access, or any request involving structured relationships between things.
tools
Integrates sage-memory into Sage workflows. Teaches the agent when to remember (store findings during work), when to recall (search memory at session start and task start), and how to learn (structured knowledge capture via sage learn). Use when the user mentions memory, remember, recall, learn, capture knowledge, onboard to codebase, or when starting any session where sage-memory MCP tools are available.
tools
Captures agent mistakes, corrections, and discovered gotchas so they are not repeated. Use when: (1) a command or operation fails unexpectedly, (2) the user corrects the agent, (3) the agent discovers non-obvious behavior through debugging, (4) an API or tool behaves differently than expected, (5) a better approach is found for a recurring task. Also searches past learnings before starting tasks to avoid known pitfalls. Activate alongside the sage-memory skill — they share the same MCP backend but serve different purposes (sage-memory = codebase knowledge, sage-self-learning = agent mistakes and gotchas).