Skip to main content

social-pack

Reframe one approved hero image into platform-specific social formats while preserving the subject, palette, and visual identity.

Source facts

Repository
suphiozdedee/projenicdesign.com
Last source activity
September 27, 2026 at 06:46
Detected SKILL.md language
English
Stars
0
Forks
0

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

Showing SKILL.md

SKILL.md
Source instructions · Read-only preview
name
Social Pack
slug
social-pack
version
1.0.0
category
image
description
Reframe one approved hero image into platform-specific social formats while preserving the subject, palette, and visual identity.
status
ready
muapi_capabilities
["media.edit_image","media.upload_file","media.check_result"]
required_connections
["muapi"]
permissions
["draft-only","workspace-write"]
# Social Pack ## Mission Turn a selected source image into a bounded set of channel crops or reframes. Use this skill when the user wants one visual adapted for multiple platforms, not a new concept for each platform. ## Required inputs - Approved source image URL or local file. - Requested formats or platforms. - Elements that must remain visible and any safe area for copy or UI overlays. - Optional short caption/overlay direction and channel brand rules. Common starting formats are `1:1`, `9:16`, `4:5`, and `16:9`; confirm current platform requirements before delivery rather than treating this list as policy. ## Model selection Use the editing/reframe entries in [MODELS.md](../../MODELS.md) and the current schema in [the MuAPI tool reference](../../references/muapi-image-tools.md). Do not use text-to-image for a simple crop unless the user asks for a creative recomposition and accepts subject drift. ## Workflow 1. Read the source asset manifest and confirm that it is the approved parent image. If local, upload it once and reuse the hosted URL for all formats. 2. Normalize each requested format to a platform label, dimensions, focal subject, crop-safe region, text-safe region, and delivery filename. 3. Skip a generation/edit call when the source already matches a requested format; return the original asset ID for that format. 4. For each remaining format, call `media.edit_image` with a reframe prompt: keep the subject uncropped, match the original lighting/palette, preserve identity/product details, and leave the requested safe area. 5. Run independent format branches in parallel when the host supports it. State the number of calls and cost signal before execution. 6. Poll and preserve each result. Review subject visibility, hands/faces, product geometry, text-safe space, edge artifacts, aspect ratio, and platform dimensions. 7. Return the complete pack with the parent-child relationship and any format that still needs manual typography or layout work. ## Decision rules - Reframing is not permission to invent missing sides of a product or crop out a required disclosure. - Keep generated text minimal; add exact copy in post-production when possible. - Preserve the selected source as the canonical parent. Do not overwrite it. - A social crop that changes the subject materially should be reported as a new creative direction, not as a faithful adaptation. ## Output format Return one row per format: platform label, ratio/dimensions, asset ID, parent asset, model/tool, output URL, status, crop/text-safe notes, and receipt link. ## Failure and missing-data behavior If a format is unsupported or a crop loses a required element, preserve the source, mark that branch failed or unsuitable, and recommend a supported ratio or manual layout. Do not silently replace it with a fresh unrelated image.
View on GitHub