Use this when you need a design-system overview of one page without creating a replication folder. Avoid it when you need a complete full-page build package; use create_replication_kit once instead. Extract design tokens, typography, CSS variables, primitives, and reusable component families. Cost: standard. Session: uses one browser page for one URL. Does not write files. Possible follow-ups if evidence is still missing: inspect_component, clone_section, extract_responsive.

Use this when you need to inspect generated, offline, or test HTML that you already have. Avoid it when the source is a live URL; use analyze_site, inspect_component, or create_replication_kit. Render supplied markup and extract the same design-system data as a remote page analysis. Cost: standard. Session: renders supplied markup without navigating to a remote URL. Does not write files.

Use this when you need an exact style and state X-ray of a known element, role, or text target. Avoid it when you need portable HTML, scoped CSS, assets, and a screenshot; use clone_section. Return targeted computed styles, cascade rules, variables, box geometry, fonts, and interaction states. Cost: light. Session: uses one browser page for one URL. Does not write files. Possible follow-ups if evidence is still missing: clone_section, capture_motion, compare_replica.

Use this when you only need visual evidence or a reference image of a page or element. Avoid it when you need the full implementation package; create_replication_kit already includes screenshots. Capture a viewport, full page, or selected element as an image the agent can inspect. Cost: light. Session: uses one browser page for one URL. May write files when an output path is provided. Possible follow-ups if evidence is still missing: inspect_component, compare_replica.

Use this when you specifically need the original asset files from one page. Avoid it when you are starting a full-page replica; create_replication_kit already downloads assets. Download original images, SVGs, fonts, videos, favicons, and backgrounds with a usage manifest. Cost: standard. Session: uses one browser page for one URL. May write files when an output path is provided. Possible follow-ups if evidence is still missing: clone_section, create_replication_kit.

Use this when you need static motion definitions and replication-ready CSS. Avoid it when you need to observe JS-driven movement over time; use capture_motion. Extract declared keyframes, transitions, timings, scroll effects, WAAPI records, and generated CSS. Cost: standard. Session: uses one browser page for one URL. Does not write files. Possible follow-ups if evidence is still missing: capture_motion, compare_replica.

Use this when you specifically need structural geometry for a page or subtree. Avoid it when you need a full replication package or multiple viewport widths. Return a bounded layout tree with sections, scroll containers, positioning, stacking, grid, and flex detail. Cost: standard. Session: uses one browser page for one URL. Does not write files. Possible follow-ups if evidence is still missing: inspect_component, extract_responsive.

Use this when you need a portable implementation package for one known section. Avoid it when you need an entire page; use create_replication_kit. Return cleaned HTML, scoped CSS, dependencies, assets, and an optional screenshot for one section. Cost: standard. Session: uses one browser page for one URL. May write files when an output path is provided. Possible follow-ups if evidence is still missing: capture_motion, compare_replica.

Use this when important UI is hidden behind interactions and you do not yet know its selectors. Avoid it when you already know the target component or need continuous motion sampling. Operate common triggers and capture menus, dialogs, tabs, accordions, and other revealed UI. Cost: heavy. Session: uses one browser page for one URL. May write files when an output path is provided. Possible follow-ups if evidence is still missing: inspect_component, clone_section, capture_motion.

Use this when you need breakpoints and concrete layout changes across screen sizes. Avoid it when one viewport is enough or a complete kit with desktop and mobile captures is the goal. Compare media rules and layout trees across requested viewport widths, with optional screenshots. Cost: heavy. Session: captures one URL at multiple viewport states. May write files when an output path is provided. Possible follow-ups if evidence is still missing: inspect_component, compare_replica.

Use this when a replica exists and you need evidence-driven visual verification or a same-page noise-floor check. Avoid it when you have not implemented a replica yet. Run a deterministic perceptual diff and attribute the worst visual mismatches to elements and styles. Cost: heavy. Session: loads the original and comparison target. Does not write files. Possible follow-ups if evidence is still missing: inspect_component.

Use this when you need a broad first-pass package for reproducing a complete page. Avoid it when you only need one component, one screenshot, or a quick design overview. Write screenshots, assets, design data, layout trees, animations, and a replication guide in one pass. Cost: heavy. Session: captures one URL at multiple viewport states. May write files when an output path is provided. Replaces these separate first-pass calls: analyze_site, capture_screenshot, extract_assets, extract_layout. Possible follow-ups if evidence is still missing: extract_responsive, extract_animations, capture_motion, compare_replica.

Use this when you need to observe actual motion, especially JS-driven animation. Avoid it when declared CSS and WAAPI definitions are sufficient; use extract_animations. Sample frames, live animations, and numeric style curves after load, scroll, hover, or click. Cost: heavy. Session: uses one browser page for one URL. May write files when an output path is provided. Possible follow-ups if evidence is still missing: inspect_component, compare_replica.

Use this when the goal spans several pages and you need shared versus page-specific design patterns. Avoid it when one page is the target; analyze or create a kit for that page instead. Merge design evidence from bounded same-origin pages into a site-wide system and page inventory. Cost: heavy. Session: loads several bounded same-origin pages. May write files when an output path is provided. Possible follow-ups if evidence is still missing: create_replication_kit.