photo-slideshow
Skill1 starsChanged 10 days agoMITView on GitHub
Build a photo promo or slideshow in a scene — probe the photos, sequence them as a track of clips with transitions, fill off-aspect photos with a blurred copy, add an animated title that stays readable over any photo, choose a font, and verify with a sweep of full-resolution frames. Use when asked for a photo slideshow, photo montage, photo reel, promo video from photos, Reels/Shorts/TikTok from pictures, or a titled sequence of images.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- PicsArt/gen-ai-skillsword for word, first seen Nov 1, 2026
- willlee1995/skillstheir own version, first seen Oct 13, 2026
- iart-ai/ecommerce-video-skillstheir own version, first seen Jun 22, 2026
Other versions3 projects use a version of this.
Word for word · 1
- PicsArt/gen-ai-skills★ 7100% same
Edited · 0
None found.
Wrote their own version · 2
- iart-ai/ecommerce-video-skills★ 17same name
- willlee1995/skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository48 other files it gives its agents.
- animate-designSkill
- app-ui-animationSkill
- asset-auditSkill
- caption-decorationsSkill
- character-riggingSkill
- custom-transitionsSkill
- design-layer-importSkill
- edit-footageSkill
- edit-on-your-computerSkill
- edit-pipelineSkill
- face-hidingSkill
- figma-storyboard-animationSkill
- film-cuttingSkill
- film-finishingSkill
- footage-analysisSkill
- footage-reframingSkill
- get-startedSkill
- large-scene-assemblySkill
- linked-element-motionSkill
- make-cartoonSkill
- make-screen-videosSkill
- motion-design-craftSkill
- motion-intentSkill
- motion-pattern-librarySkill
- motion-reviewSkill
- motion-scene-continuitySkill
- motion-tracking-from-videoSkill
- picsart-video-rulesSkill
- quality-reviewSkill
- quick-merge-and-exportSkill
- realistic-color-gradeSkill
- recreate-videoSkill
- reel-structureSkill
- reference-fragment-matchingSkill
- resting-frame-rulesSkill
- retention-hooksSkill
- scene-3d-authoringSkill
- scene-animation-referenceSkill
- scene-building-basicsSkill
- scene-template-remixSkill
- screen-recording-analysisSkill
- screen-recording-layoutsSkill
- screen-recording-zoomSkill
- screen-ui-redesignSkill
- social-postingSkill
- speech-aware-cuttingSkill
- spoken-captionsSkill
- visual-styleSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
When you read picsart_scene_capabilities or the recipe index (picsart_scene_recipe without name), pass useCase: "photo-slideshow": the …
- Distinctive: line 18
Every value below was validated and rendered at 1080x1920.
- Very specific: line 26
Call picsart_media_probe on every photo URL before authoring.
- Very specific: line 68
Choosing a transition: Read picsart_scene_capabilities({ sections: ["transitions"] }).
- Very specific: line 106
An entrance starts at its layer's local time 0: No text-animation preset takes a delay.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Photo promo / slideshow
When you read picsart_scene_capabilities or the recipe index (picsart_scene_recipe without name), pass useCase: "photo-slideshow": the answer leads with what this job reads and names the rest in one line.
Assumes the font asset from picsart_scene_fonts (a live.fonts entry), and that picsart_scene_patch returns the patched scene you pass to the next tool.
The most common scene job: N photos, one after another, a transition at each seam, a title. Every value below was validated and rendered at 1080x1920; the whole scene is in the worked example.
Related recipes: picsart_scene_recipe scene-video-edits owns the track, transition-window and
blur_fill detail; picsart_scene_recipe motion-intent chooses durations and easing.
1. Probe the photos
Call picsart_media_probe on every photo URL before authoring. Declare its width and height on the
asset. Those numbers decide which fit each photo gets (section 3), and they let blur_fill skip
its fill on a photo whose aspect already matches the composition.
"p2": { "type": "image", "uri": "https://…/507090463024201.jpg", "width": 2494, "height": 2560 }Compare each photo's aspect (w/h) with the composition's (1080/1920 = 0.5625). cover crops
the difference away: 2296x4080 (0.563) loses nothing visible, and a 2:3 portrait such as
2308x3462 (0.667) loses 16% of its width. A square or landscape photo such as 2494x2560 (0.974)
would lose 42% of its width, so it gets a blurred fill instead.
2. A track of clips with transitions
One track layer holds the photos as clips, back to back. Each seam that should not be a hard
cut gets one transitions[] entry keyed by between, the index of the gap (0 = between clips 0
and 1):
{ "id": "reel", "start": 0, "duration": 12, "content": { "kind": "track",
"clips": [
{ "id": "c1", "duration": 3, "content": { "kind": "media", "assetId": "p1", "fit": "cover" } },
{ "id": "c2", "duration": 3, "content": { "kind": "media", "assetId": "p2" },
"looks": [{ "look": "blur_fill", "params": { "blur": 8, "resFactor": 10, "dim": 0.25 } }] },
{ "id": "c3", "duration": 3, "content": { "kind": "media", "assetId": "p3", "fit": "cover" } },
{ "id": "c4", "duration": 3, "content": { "kind": "media", "assetId": "p4", "fit": "cover" } }
],
"transitions": [
{ "between": 0, "id": "crossfade", "duration": 0.6 },
{ "between": 1, "id": "slide", "duration": 0.6, "params": { "direction": "left" } },
{ "between": 2, "id": "crossfade", "duration": 0.6 }
] } }- Timing. The track's
durationis the sum of its clip durations (4 x 3 = 12). A transition does not shorten the reel. Its window is centred on the seam, so a 0.6s crossfade at the 3.0s seam runs from 2.7 to 3.3. - Duration rule. Each window takes half its duration out of each clip beside it. The halves a
clip gives up to both of its seams must fit inside that clip, or validation refuses the scene
with
transition_window_exceeds_clip. With 3s photos and 0.6s transitions, an inner photo still has 2.4s on its own. - Choosing a transition. Read
picsart_scene_capabilities({ sections: ["transitions"] }). A closed-set param lists its allowedenumvalues, and the validator refuses anything else. Keep one calm transition,crossfade, as the default, and spend a directional one (slide) on the seam that should read as "next". The rendered example'sslidedirection: "left"visibly moves through its window. - All seams the same?
scene://montage(picsart_scene_capabilities, sectiontemplate:scene://montage) binds a clip list plus onetransitionid andtransitionDurationfor every seam. Author the track by hand as above when seams differ or a transition takes params.
3. Cover, or contain with a blurred fill
| Photo aspect vs composition | Clip content |
| Portrait, cover loses under about a fifth of the photo | "fit": "cover". It fills the frame and crops the overflow. Add align ("top", …) if the subject sits off-centre. |
| Clearly different (square, landscape) | No fit, plus looks: [{ "look": "blur_fill", … }]. The whole photo is contained, and a blurred, cover-fitted copy of it fills the margins. |
blur_fill with blur: 8, resFactor: 10, dim: 0.25 renders the fill as an out-of-focus copy of
the photo under a light scrim that separates it from the contained photo. Fine detail softens,
but large shapes stay recognisable: the example's tree trunks still read in the margins. Raising
blur to its maximum of 32 halves the fill's edge detail, and the trunks still show. The fill is
a soft backdrop, not an abstract wash. Ranges and defaults are in the looks
capability section. A look on a clip is written on the clip itself, as above, and the clip keeps
its transitions.
4. The title and its timing
A text layer above the track. Keep the entrance editable as a preset reference in
content.animations, or bake it with picsart_scene_apply_preset (textAnimation:<name>). Both render identically.
{ "id": "title", "start": 0.5, "duration": 2.1, "transform": { "position": [540, 1400] },
"animations": { "opacity": { "keyframes": [ { "time": 2.3, "value": 1 }, { "time": 2.6, "value": 0 } ] } },
"content": { "kind": "text", "text": "OUR WINTER",
"font": { "family": "gilroy-bold", "size": 132 }, "color": "#ffffff", "alignment": "center",
"animations": [ { "preset": "fade_in_chars", "params": { "duration": 0.8, "stagger": 0.4 } } ] },
"effects": [ { "id": "drop_shadow", "params": { "offset": [0, 6], "blur": 24, "color": "#000000b3" } } ] }- An entrance starts at its layer's local time 0. No text-animation preset takes a delay.
To reveal later, move the layer:
start: 0.5begins the reveal at 0.5s. Changestartanddurationtogether. Rendered: no ink at 0.5, partial at 0.9, the whole title by 1.3. - Layer keyframes are composition seconds. A layer's own
animationskeyframes (the opacity exit2.3 → 2.6) are composition time by default, unlike the text preset's clock. The validator warns (keyframe_time_outside_window) when one falls outside the layer's window. - Gone before the first seam, by choice. The example's title is an opening card over the first photo, so it leaves before anything moves under it. The layer ends at 0.5 + 2.1 = 2.6s, and the first seam's window opens at 2.7s (0.6s centred on 3.0). Rendered at 2.65: no title and no scrim. To keep a title up for the whole promo, give it and its scrim the reel's window instead. The scrim then has to hold over every photo, including the brightest (section 5).
drop_shadowandstrokeare layer styles, drawn last. Wherever they sit ineffects[], they are drawn from the layer after its other effects and its mask, as After Effects orders them. So a shadow follows what the effects left. In"effects": [ { "id": "drop_shadow", "params": { "offset": [0, 6], "blur": 24, "color": "#000000b3" } }, { "id": "gaussian_blur", "params": { "blur": 6 } } ]the title is blurred first, then casts a soft shadow from the blurred shape: listing the shadow first changes nothing. Likewise a stroke traces the edge the effects left, and a word a mask cuts casts no shadow where it is cut.- Wrap at a column you declare.
wrapWidth: 900wraps a long title inside the 900px between 90px side gutters. WithoutwrapWidtha line wraps at the composition's full width, so it can run edge to edge. "OUR WINTER" at 132pxgilroy-bolddraws 776px of ink, one line inside 900px.
5. Readability over photos
White text over a photo is only as readable as the brightest photo it can sit on. Measured over
the brightest photo in the example (snow; background median luminance 0.59), with white
gilroy-bold 132px, against the median background pixel under the title (brighter patches
fare worse). The scrim and plate rows keep the shadow on:
| Treatment | Contrast vs background |
| none | 1.6 : 1 |
drop_shadow offset [0,6], blur 24, #000000b3 | 2.0 : 1 |
| gradient scrim, alpha 0 → 0.6 at the text line → 0.6 at the bottom | 4.0 : 1 |
backing plate: 860x190 rect, roundness 28, black fill opacity 0.55 | 4.0 : 1 |
Rules that follow:
- A shadow defines edges; it does not make text readable on a bright photo. Keep the shadow for crispness, and put a scrim or a plate under any title that can land on a light photo. The large-text floor is 3:1, and both treatments clear it.
- Scrim. A full-width shape layer beneath the title, with exactly the title's window
(
start0.5,duration2.1) and its own opacity fades at both ends:
{ "id": "scrim", "start": 0.5, "duration": 2.1, "transform": { "position": [540, 1420] },
"animations": { "opacity": { "keyframes": [
{ "time": 0.5, "value": 0 }, { "time": 0.8, "value": 1 }, { "time": 2.3, "value": 1 }, { "time": 2.6, "value": 0 } ] } },
"content": { "kind": "shape", "items": [
{ "type": "rect", "position": [0, 0], "size": [1080, 1000] },
{ "type": "gradientFill", "gradientType": "linear", "startPoint": [0, -500], "endPoint": [0, 500],
"colorStops": [ { "offset": 0, "color": "#000000" }, { "offset": 1, "color": "#000000" } ],
"alphaStops": [ { "offset": 0, "alpha": 0 }, { "offset": 0.5, "alpha": 0.6 }, { "offset": 1, "alpha": 0.6 } ] } ] } }- Placement. A common convention for 9:16 social output is to keep text inside x 90–990 and y 250–1540. The example's title ink spans x 153–928 and y 1351–1447.
6. Fonts
font.family is a NAME. Search the catalog with picsart_scene_fonts({ query: "gilroy" }) and put a
returned key in font.family. The example uses gilroy-bold, which that query returns and
which is also in the list picsart_scene_fonts() gives with no query. Pass licenseType: "commercial" to list only faces cleared
for commercial use; personal-licence faces are listed too otherwise.
The renderer has no system fonts, so a CSS name such as Inter is refused with
unknown_font_family, and the refusal lists the closest keys.
For a face the catalog lacks, the same picsart_scene_fonts({ query }) call returns it in live.fonts. Each entry carries asset, a
ready-made font asset: add it to assets under an id, and name that id:
"assets": { "display": { "type": "font", "name": "Casual Fonts", "uri": "https://cdn111.picsart.com/230454663055900.ttf" } },
… "font": { "family": "display", "size": 64 }A bare .ttf URL in font.family also renders, but it loses the font's name.
7. Keep the context lean
Author the scene once. Then edit it with picsart_scene_patch ops anchored on ids
({ op: "set", layer: "title", path: "content/text", value: "…" }) rather than re-sending the
document. A patch returns the patched scene as scene: pass it as the next tool's
scene (patch, validate, contact sheet, export). A successful patch has already been validated, so don't validate again until you
hand-edit.
8. Verify with one sweep
Ask for the decisive moments in as few calls as possible: picsart_scene_contact_sheet with times: [t1, t2, …] and resolution set to the scene's composition.width and composition.height. Without resolution the sheet returns downscaled thumbnails, which are not for judging position. The sheet image scales every frame to fit one picture; to judge full-resolution detail, open a frame's url.
The example has twelve moments, so it takes two calls, both with resolution: { width: 1080, height: 1920 }:
times: [0.5, 0.9, 1.3, 1.5, 2.65, 3.0]
times: [4.5, 6.0, 7.5, 9.0, 10.5, 11.967]| Times | What they prove |
| 0.5, 0.9, 1.3 | The title's reveal: nothing, partial, whole |
| 1.5, 4.5, 7.5, 10.5 | Clip midpoints: the fit / blur_fill framing of each photo |
| 2.65 | Title and scrim are gone before the first seam's window (2.7) opens |
| 3.0, 6.0, 9.0 | Seam centres: each transition is mid-flight, not a hard cut |
| 11.967 | The last frame (duration − 1/fps): the reel ends on a photo, not on black |
A seam centre that shows only one photo is a hard cut. Before you re-author anything, check the
transition's values.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

