{"id":"efd02f32-d365-4665-ad04-063ebc4d4cf6","entityType":"agent","slug":"clawhub-tobewin-four-layer-photo-poster","name":"four-layer-photo-poster","canonicalUrl":"https://www.xpersona.co/agent/clawhub-tobewin-four-layer-photo-poster","canonicalPath":"/agent/clawhub-tobewin-four-layer-photo-poster","generatedAt":"2026-10-11T16:03:09.958Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T12:58:43.720Z","emptyReason":null},"description":"Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene. Skill: four-layer-photo-poster Owner: tobewin Summary: Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene. Tags: latest:0.1.0 Version history: v0.1.0 | 2026-08-26T03:05:50.131Z | auto - Initial release of four-layer-photo-poster skill. - Creates one high-design 3:4 vertical poster per input photo, structured as four equal horizontal","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s174rz3f0862tcfw7pzfh5w8kn83hv2z:four-layer-photo-poster","sourceUrl":"https://clawhub.ai/tobewin/four-layer-photo-poster","homepage":"https://clawhub.ai/tobewin/skills/four-layer-photo-poster","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/tobewin/four-layer-photo-poster","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/tobewin/skills/four-layer-photo-poster","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene. Skill: four-layer-phot"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T12:58:43.720Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T12:58:43.720Z","emptyReason":null},"stars":null,"forks":null,"downloads":1061,"packageName":null,"latestVersion":"0.1.0","tractionLabel":"1.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T12:58:43.707Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T12:58:43.720Z","lastCrawledAt":"2026-10-11T12:58:43.707Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T12:58:43.707Z","lastVerifiedAt":null,"highlights":[{"version":"0.1.0","createdAt":"2026-08-26T03:05:50.131Z","changelog":"- Initial release of four-layer-photo-poster skill. - Creates one high-design 3:4 vertical poster per input photo, structured as four equal horizontal bands: original photo, selective-memory stickers, watercolor ticket, and gilded enamel magnet. - Requires a two-stage workflow: generate four panoramic layer assets per photo and assemble with an included script for precise alignment. - Ensures subject integrity and visual consistency across all bands; no collaging or mixing of multiple photos. - Text is only added if supplied or unambiguously visible—never inferred or generated by default. - Assembled output is always an exact 3:4 canvas with four seamless, contiguous bands.","fileCount":11,"zipByteSize":9220}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s174rz3f0862tcfw7pzfh5w8kn83hv2z:four-layer-photo-poster","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T16:03:09.957Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-tobewin-four-layer-photo-poster/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-11T12:58:43.720Z","emptyReason":null},"readme":"Skill: four-layer-photo-poster\n\nOwner: tobewin\n\nSummary: Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene.\n\nTags: latest:0.1.0\n\nVersion history:\n\nv0.1.0 | 2026-08-26T03:05:50.131Z | auto\n\n- Initial release of four-layer-photo-poster skill.\n- Creates one high-design 3:4 vertical poster per input photo, structured as four equal horizontal bands: original photo, selective-memory stickers, watercolor ticket, and gilded enamel magnet.\n- Requires a two-stage workflow: generate four panoramic layer assets per photo and assemble with an included script for precise alignment.\n- Ensures subject integrity and visual consistency across all bands; no collaging or mixing of multiple photos.\n- Text is only added if supplied or unambiguously visible—never inferred or generated by default.\n- Assembled output is always an exact 3:4 canvas with four seamless, contiguous bands.\n\nArchive index:\n\nArchive v0.1.0: 11 files, 9220 bytes\n\nFiles: .gitignore (168b), agents (0b), agents/openai.yaml (197b), LICENSE (1064b), references (0b), references/production-prompt.md (6174b), scripts (0b), scripts/compose_poster.py (2130b), skill-card.md (2073b), SKILL.md (4199b), _meta.json (142b)\n\nFile v0.1.0:SKILL.md\n\n---\nname: four-layer-photo-poster\ndescription: Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene.\nmetadata:\n  short-description: Four-layer photo design posters\n---\n\n# Four-layer photo poster\n\nUse this skill for one independent high-design 3:4 poster per supplied photograph. The poster transitions through four contiguous horizontal bands: original photography, selective-memory stickers, watercolor ticket, and gilded enamel magnet. It is never a multi-photo collage.\n\n## Required two-stage workflow\n\nFor each source photo, first create four separate panoramic layer assets, then assemble them deterministically. Do not ask the image model to draw the final four-band layout in one pass: it cannot reliably enforce equal bands or a seamless join.\n\n1. Inspect one source image and derive only source-evident variables. Repeat the full workflow independently for every supplied photo.\n2. Read [layer prompts](references/production-prompt.md). Use the same source image for every generated layer, requesting each layer as a single approximately 3:1 panorama.\n3. Make the photo layer from the original pixels wherever the crop permits. If horizontal background expansion is necessary, use a photo-faithful, background-only outpaint; never regenerate or retouch the subject.\n4. Generate the three stylized layers independently using their corresponding prompt. Preserve the same subject anchors (center, scale, silhouette, direction, viewpoint and core spatial relationship) set by the photo layer.\n5. Assemble the four prepared layer files with [compose_poster.py](scripts/compose_poster.py). It creates an exact 3:4 canvas, four equal 25%-height bands, and zero inter-band pixels of space.\n6. Inspect the assembled poster against the acceptance checklist. Fix the failed layer and reassemble; do not solve a stylization failure by altering the original photo band.\n\n## Invariants\n\n- Canvas is exactly 3:4 portrait. Bands are exactly `1:1:1:1`, from top to bottom, with no separator, border, gutter, blank seam, rounded panel, logo, watermark, or layer number.\n- The source photo is the only visual source. All bands retain the same identity, subject count, structure, proportions, pose, gaze, direction, camera viewpoint, perspective, occlusion, left/right placement, foreground/background order, lighting logic, and narrative.\n- Each band is a shallow landscape crop. Horizontally extend background or lightly crop only when it does not remove a head, limbs, key object, or essential relationship. Never vertically stretch the subject.\n- Align subject anchors across bands: the principal silhouette, visual center, direction of motion/gaze, and key contour positions should land in comparable horizontal positions.\n\n## Text policy\n\nDefault to no generated text. Add text only when the user supplies the copy or when a short source-supported label is safely legible. Never infer a place, date, identity, or factual identifier. Decorative ticket rules, linework and a blank stamp are preferable to garbled microtext.\n\n## Assembly\n\nRun the helper with four image paths in visual order:\n\n```bash\npython3 scripts/compose_poster.py --photo /path/photo.png --stickers /path/stickers.png --ticket /path/ticket.png --magnet /path/magnet.png --output /path/poster.png\n```\n\nIts default output is 1200×1600. Use `--width` with a positive integer divisible by 3 to select another 3:4 output size; each band will be exactly one quarter of the resulting height.\n\n## Acceptance checklist\n\n- Exactly one output per uploaded source photo, with no shared visual facts across sources.\n- Final image is 3:4; its four bands are identical pixel heights and meet with no gap.\n- Top band is an unillustrated faithful photo, with its subject preserved from original pixels.\n- All four bands show the same moment and scene relation; no added subject, altered face, changed pose, reversed placement, missing key element, or invented setting.\n- Sticker band has one main visual and six unequal, non-grid memory stickers.\n- Ticket and magnet bands are restrained, legible when text exists, and contain no invented or garbled prominent copy.\n\nFile v0.1.0:_meta.json\n\n{\n  \"ownerId\": \"kn75z6gevjsyrznm7dg2ez6sen82h8sz\",\n  \"slug\": \"four-layer-photo-poster\",\n  \"version\": \"0.1.0\",\n  \"publishedAt\": 1787713550131\n}\n\nFile v0.1.0:references/production-prompt.md\n\n# Layer prompts\n\nUse one prompt per layer and one source photo per run. Derive variables only from what is visible. Every generated layer should be a clean horizontal panorama near 3:1, with no exterior frame, panel border, layer label, watermark, or logo; the assembly script supplies the final geometry.\n\n## Shared source lock\n\nPrepend this to every generated-layer prompt:\n\n```text\nUse this one reference photograph as the sole visual source. Preserve the same subject identity, count, physical structure, proportions, pose, action direction, gaze, camera viewpoint, perspective, occlusion, left/right placement, foreground/background order and core narrative. Maintain the same approximate horizontal anchor points: [SUBJECT CENTER], [PRINCIPAL SILHOUETTE], [KEY RELATIONSHIP]. Create one full-bleed, approximately 3:1 horizontal panorama. If adaptation is needed, extend only scene-consistent background sideways or lightly crop it sideways; never vertically stretch, crop key anatomy/objects, change the scene, or invent an unrelated subject. No text unless the layer explicitly asks for user-provided copy.\n```\n\n## Layer 1 — original photography\n\nPrefer the source image directly, cropped or background-outpainted only as necessary to form the panorama. When a generation tool is required for outpainting, append:\n\n```text\nKeep the source photograph completely photographic and unretouched. Preserve the subject pixels exactly; expand only the background at the left and/or right edge with the same environment, perspective, natural illumination, shadow direction, detail level and core colors. No face/body regeneration, illustration, beautification, material conversion, color grade, invented scenery, or decorative element.\n```\n\n## Layer 2 — selective-memory sticker reconstruction\n\n```text\nTranslate the same source scene into a premium independent-publication selective-memory sticker page. Compress the scene into 3–6 large recognizable gouache, paper-cut, Risograph and screen-print flat shapes rather than tracing it. Make one dominant main visual and exactly six smaller memory stickers selected from these source-true details: [SIX SOURCE-TRUE DETAILS]. The six stickers must vary substantially in scale and importance: integrate two or three partly overlapping, clipped at an edge, or attached near the main visual; avoid a six-grid or product-catalog arrangement. Give them thick pale cut edges, relaxed asymmetric placement, gentle overlap/rotation, generous quiet negative space, and clear hierarchy.\n\nUse opaque matte gouache color blocks, paper grain, hand-cut edges and slight print-registration offsets. Palette: airy muted powder blue, mist blue, sky blue, ivory, cream, pale beige, soft sage and architectural neutrals, with only tiny dusty-rose accents. Avoid literal photorealism, watercolor washes, dense linework, smooth vector art, 3D plastic, anime, ecommerce styling and excess detail. Do not generate text.\n```\n\n## Layer 3 — watercolor travel-ticket postcard\n\n```text\nCreate a centered horizontal watercolor ticket on a spacious ivory background. Use thick matte watercolor paper, delicately serrated edges and an extremely subtle natural shadow. The ticket is structured precisely: its left 74% is a watercolor scene; its right 26% is a calm ticket-information stub; a vertical dotted/perforation/tear line separates them. Use aligned but blank information rules, short lines, dot columns and at most one unlabeled circular stamp; do not generate words, dates, place names, numbers or pseudo-text unless the user supplied exact copy.\n\nIn the 74% watercolor scene, retain the source’s recognizable subject and necessary environment relation. Paint with bright clean low-to-medium saturation watercolor, using 2–4 related tones from sage, mist blue, pale blue-gray, cream yellow, warm sand and light terracotta, with minimal warm highlights. Show paper white, thin translucent washes, wet/dry variation and soft granulation. The subject is slightly clearer than the naturally subdued environment. Avoid dirty vintage casts, heavy brown, clutter, floating information, a commercial template and photo-real rendering.\n```\n\n## Layer 4 — gilded enamel travel magnet\n\n```text\nReduce the same source scene to one refined premium collectible enamel travel magnet/badge. Preserve the distinctive [SUBJECT SILHOUETTE / KEY STRUCTURE] and essential spatial relation while simplifying all nonessential detail. Use clean vivid harmonious colors sampled from the source; smooth real enamel fills; thin flowing gold-metal internal dividers; crisp white-metal outer contour; rounded beveled thickness; restrained metal glints; and a short soft cast shadow on a broad uncluttered cream, pale gray, pastel or source-derived light background. The badge may be mildly off-center or tilted but must feel stable and collectible.\n\nDo not generate wording by default. If the user provided exact short copy, integrate only that copy into the badge edge or enamel nameplate. Avoid a flat sticker, cartoon, plastic 3D render, thick harsh shadow, cheap gold trim, busy decoration, ecommerce product photograph and generic template appearance.\n```\n\n## Variables and repair instructions\n\n- **[SUBJECT CENTER]**: a source-observed horizontal anchor, such as `the cyclist centered slightly left`.\n- **[PRINCIPAL SILHOUETTE]**: the dominant source-observed outline.\n- **[KEY RELATIONSHIP]**: a source-observed relation, such as `the dog in front of the seated person`.\n- **[SIX SOURCE-TRUE DETAILS]**: six visible, unequal-priority details: gesture, prop, material, plant, vehicle, pattern, food, light, etc.\n- **[SUBJECT SILHOUETTE / KEY STRUCTURE]**: one or two distinctive source-observed outline features.\n\nWhen retrying, target one defect only:\n\n- `Restore the source subject’s identity, count, pose, direction, viewpoint and placement. Change nothing else.`\n- `Preserve the existing style but realign the principal silhouette and visual center with the source-photo strip.`\n- `Remove all words, numbers and pseudo-text while keeping the ticket structure.`\n- `Keep exactly six varied memory stickers, with a clearly dominant main visual and no grid.`\n\nFile v0.1.0:skill-card.md\n\n## Description:\n\nTurn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[tobewin](https://clawhub.ai/user/tobewin)\n\n### License/Terms of Use:\n\nMIT\n\n## Use Case:\n\nExternal users and creative agents use this skill to transform each supplied photograph into a separate four-band design poster while preserving the source scene, subject identity, and exact 3:4 geometry.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Users provide photos to an image-generation workflow, and generated stylized layers may add unwanted invented details.\n\nMitigation: Use the source photo as the sole visual source, derive variables only from visible evidence, and review outputs against the acceptance checklist before use.\n\nRisk: A one-pass generated poster may produce uneven bands, gaps, or misaligned subject anchors.\n\nMitigation: Generate four panoramic layer assets independently, assemble them with the included script, and verify the final 3:4 canvas and equal-height bands.\n\n## Reference(s):\n\n- [Layer prompts](references/production-prompt.md)\n- [ClawHub skill page](https://clawhub.ai/tobewin/skills/four-layer-photo-poster)\n- [Server-resolved GitHub provenance](https://github.com/ToBeWin/four-layer-photo-poster)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, guidance]\n\n**Output Format:** [Markdown guidance with prompt text and shell commands for assembling image files]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Produces one assembled 3:4 poster per source photo after four layer assets are generated.]\n\n## Skill Version(s):\n\n0.1.0 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v0.1.0:agents/openai.yaml\n\ndisplay_name: \"Four-Layer Photo Poster\"\nshort_description: \"Create separate four-stage 3:4 design posters from photos\"\ndefault_prompt: \"Create a four-layer photo poster from each image I provide.\"\n\nFile v0.1.0:LICENSE\n\nMIT License\n\nCopyright (c) 2026 ToBeWin\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.","readmeExcerpt":"Skill: four-layer-photo-poster Owner: tobewin Summary: Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene. Tags: latest:0.1.0 Version history: v0.1.0 | 2026-08-26T03:05:50.131Z | auto - Initial release of four-layer-photo-poster skill. - Creates one high-design 3:4 vertical poster per input photo, structured as four equal horizontal","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"python3 scripts/compose_poster.py --photo /path/photo.png --stickers /path/stickers.png --ticket /path/ticket.png --magnet /path/magnet.png --output /path/poster.png"},{"language":"text","snippet":"Use this one reference photograph as the sole visual source. Preserve the same subject identity, count, physical structure, proportions, pose, action direction, gaze, camera viewpoint, perspective, occlusion, left/right placement, foreground/background order and core narrative. Maintain the same approximate horizontal anchor points: [SUBJECT CENTER], [PRINCIPAL SILHOUETTE], [KEY RELATIONSHIP]. Create one full-bleed, approximately 3:1 horizontal panorama. If adaptation is needed, extend only scene-consistent background sideways or lightly crop it sideways; never vertically stretch, crop key anatomy/objects, change the scene, or invent an unrelated subject. No text unless the layer explicitly asks for user-provided copy."},{"language":"text","snippet":"Keep the source photograph completely photographic and unretouched. Preserve the subject pixels exactly; expand only the background at the left and/or right edge with the same environment, perspective, natural illumination, shadow direction, detail level and core colors. No face/body regeneration, illustration, beautification, material conversion, color grade, invented scenery, or decorative element."},{"language":"text","snippet":"Translate the same source scene into a premium independent-publication selective-memory sticker page. Compress the scene into 3–6 large recognizable gouache, paper-cut, Risograph and screen-print flat shapes rather than tracing it. Make one dominant main visual and exactly six smaller memory stickers selected from these source-true details: [SIX SOURCE-TRUE DETAILS]. The six stickers must vary substantially in scale and importance: integrate two or three partly overlapping, clipped at an edge, or attached near the main visual; avoid a six-grid or product-catalog arrangement. Give them thick pale cut edges, relaxed asymmetric placement, gentle overlap/rotation, generous quiet negative space, and clear hierarchy.\n\nUse opaque matte gouache color blocks, paper grain, hand-cut edges and slight print-registration offsets. Palette: airy muted powder blue, mist blue, sky blue, ivory, cream, pale beige, soft sage and architectural neutrals, with only tiny dusty-rose accents. Avoid literal photorealism, watercolor washes, dense linework, smooth vector art, 3D plastic, anime, ecommerce styling and excess detail. Do not generate text."},{"language":"text","snippet":"Create a centered horizontal watercolor ticket on a spacious ivory background. Use thick matte watercolor paper, delicately serrated edges and an extremely subtle natural shadow. The ticket is structured precisely: its left 74% is a watercolor scene; its right 26% is a calm ticket-information stub; a vertical dotted/perforation/tear line separates them. Use aligned but blank information rules, short lines, dot columns and at most one unlabeled circular stamp; do not generate words, dates, place names, numbers or pseudo-text unless the user supplied exact copy.\n\nIn the 74% watercolor scene, retain the source’s recognizable subject and necessary environment relation. Paint with bright clean low-to-medium saturation watercolor, using 2–4 related tones from sage, mist blue, pale blue-gray, cream yellow, warm sand and light terracotta, with minimal warm highlights. Show paper white, thin translucent washes, wet/dry variation and soft granulation. The subject is slightly clearer than the naturally subdued environment. Avoid dirty vintage casts, heavy brown, clutter, floating information, a commercial template and photo-real rendering."},{"language":"text","snippet":"Reduce the same source scene to one refined premium collectible enamel travel magnet/badge. Preserve the distinctive [SUBJECT SILHOUETTE / KEY STRUCTURE] and essential spatial relation while simplifying all nonessential detail. Use clean vivid harmonious colors sampled from the source; smooth real enamel fills; thin flowing gold-metal internal dividers; crisp white-metal outer contour; rounded beveled thickness; restrained metal glints; and a short soft cast shadow on a broad uncluttered cream, pale gray, pastel or source-derived light background. The badge may be mildly off-center or tilted but must feel stable and collectible.\n\nDo not generate wording by default. If the user provided exact short copy, integrate only that copy into the badge edge or enamel nameplate. Avoid a flat sticker, cartoon, plastic 3D render, thick harsh shadow, cheap gold trim, busy decoration, ecommerce product photograph and generic template appearance."}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: four-layer-photo-poster\ndescription: Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene.\nmetadata:\n  short-description: Four-layer photo design posters\n---\n\n# Four-layer photo poster\n\nUse this skill for one independent high-design 3:4 poster per supplied photograph. The poster transitions through four contiguous horizontal bands: original photography, selective-memory stickers, watercolor ticket, and gilded enamel magnet. It is never a multi-photo collage.\n\n## Required two-stage workflow\n\nFor each source photo, first create four separate panoramic layer assets, then assemble them deterministically. Do not ask the image model to draw the final four-band layout in one pass: it cannot reliably enforce equal bands or a seamless join.\n\n1. Inspect one source image and derive only source-evident variables. Repeat the full workflow independently for every supplied photo.\n2. Read [layer prompts](references/production-prompt.md). Use the same source image for every generated layer, requesting each layer as a single approximately 3:1 panorama.\n3. Make the photo layer from the original pixels wherever the crop permits. If horizontal background expansion is necessary, use a photo-faithful, background-only outpaint; never regenerate or retouch the subject.\n4. Generate the three stylized layers independently using their corresponding prompt. Preserve the same subject anchors (center, scale, silhouette, direction, viewpoint and core spatial relationship) set by the photo layer.\n5. Assemble the four prepared layer files with [compose_poster.py](scripts/compose_poster.py). It creates an exact 3:4 canvas, four equal 25%-height bands, and zero inter-band pixels of space.\n6. Inspect the assembled poster against the acceptance checklist. Fix the failed layer and reassemble; do not solve a stylization failure by altering the original photo band.\n\n## Invariants\n\n- Canvas is exactly 3:4 portrait. Bands are exactly `1:1:1:1`, from top to bottom, with no separator, border, gutter, blank seam, rounded panel, logo, watermark, or layer number.\n- The source photo is the only visual source. All bands retain the same identity, subject count, structure, proportions, pose, gaze, direction, camera viewpoint, perspective, occlusion, left/right placement, foreground/background order, lighting logic, and narrative.\n- Each band is a shallow landscape crop. Horizontally extend background or lightly crop only when it does not remove a head, limbs, key object, or essential relationship. Never vertically stretch the subject.\n- Align subject anchors across bands: the principal silhouette, visual center, direction of motion/gaze, and key contour positions should land in comparable horizontal positions.\n\n## Text policy\n\nDefault to no generated text. Add text only when the user supplies the copy or when a short source-supported label is safely legible. Never infer a place, date, identity, or factual"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn75z6gevjsyrznm7dg2ez6sen82h8sz\",\n  \"slug\": \"four-layer-photo-poster\",\n  \"version\": \"0.1.0\",\n  \"publishedAt\": 1787713550131\n}"},{"path":"references/production-prompt.md","content":"# Layer prompts\n\nUse one prompt per layer and one source photo per run. Derive variables only from what is visible. Every generated layer should be a clean horizontal panorama near 3:1, with no exterior frame, panel border, layer label, watermark, or logo; the assembly script supplies the final geometry.\n\n## Shared source lock\n\nPrepend this to every generated-layer prompt:\n\n```text\nUse this one reference photograph as the sole visual source. Preserve the same subject identity, count, physical structure, proportions, pose, action direction, gaze, camera viewpoint, perspective, occlusion, left/right placement, foreground/background order and core narrative. Maintain the same approximate horizontal anchor points: [SUBJECT CENTER], [PRINCIPAL SILHOUETTE], [KEY RELATIONSHIP]. Create one full-bleed, approximately 3:1 horizontal panorama. If adaptation is needed, extend only scene-consistent background sideways or lightly crop it sideways; never vertically stretch, crop key anatomy/objects, change the scene, or invent an unrelated subject. No text unless the layer explicitly asks for user-provided copy.\n```\n\n## Layer 1 — original photography\n\nPrefer the source image directly, cropped or background-outpainted only as necessary to form the panorama. When a generation tool is required for outpainting, append:\n\n```text\nKeep the source photograph completely photographic and unretouched. Preserve the subject pixels exactly; expand only the background at the left and/or right edge with the same environment, perspective, natural illumination, shadow direction, detail level and core colors. No face/body regeneration, illustration, beautification, material conversion, color grade, invented scenery, or decorative element.\n```\n\n## Layer 2 — selective-memory sticker reconstruction\n\n```text\nTranslate the same source scene into a premium independent-publication selective-memory sticker page. Compress the scene into 3–6 large recognizable gouache, paper-cut, Risograph and screen-print flat shapes rather than tracing it. Make one dominant main visual and exactly six smaller memory stickers selected from these source-true details: [SIX SOURCE-TRUE DETAILS]. The six stickers must vary substantially in scale and importance: integrate two or three partly overlapping, clipped at an edge, or attached near the main visual; avoid a six-grid or product-catalog arrangement. Give them thick pale cut edges, relaxed asymmetric placement, gentle overlap/rotation, generous quiet negative space, and clear hierarchy.\n\nUse opaque matte gouache color blocks, paper grain, hand-cut edges and slight print-registration offsets. Palette: airy muted powder blue, mist blue, sky blue, ivory, cream, pale beige, soft sage and architectural neutrals, with only tiny dusty-rose accents. Avoid literal photorealism, watercolor washes, dense linework, smooth vector art, 3D plastic, anime, ecommerce styling and excess detail. Do not generate text.\n```\n\n## Layer 3 — watercolor travel-ticket postcard\n\n```te"},{"path":"skill-card.md","content":"## Description:\n\nTurn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[tobewin](https://clawhub.ai/user/tobewin)\n\n### License/Terms of Use:\n\nMIT\n\n## Use Case:\n\nExternal users and creative agents use this skill to transform each supplied photograph into a separate four-band design poster while preserving the source scene, subject identity, and exact 3:4 geometry.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Users provide photos to an image-generation workflow, and generated stylized layers may add unwanted invented details.\n\nMitigation: Use the source photo as the sole visual source, derive variables only from visible evidence, and review outputs against the acceptance checklist before use.\n\nRisk: A one-pass generated poster may produce uneven bands, gaps, or misaligned subject anchors.\n\nMitigation: Generate four panoramic layer assets independently, assemble them with the included script, and verify the final 3:4 canvas and equal-height bands.\n\n## Reference(s):\n\n- [Layer prompts](references/production-prompt.md)\n- [ClawHub skill page](https://clawhub.ai/tobewin/skills/four-layer-photo-poster)\n- [Server-resolved GitHub provenance](https://github.com/ToBeWin/four-layer-photo-poster)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, guidance]\n\n**Output Format:** [Markdown guidance with prompt text and shell commands for assembling image files]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Produces one assembled 3:4 poster per source photo after four layer assets are generated.]\n\n## Skill Version(s):\n\n0.1.0 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment."},{"path":"agents/openai.yaml","content":"display_name: \"Four-Layer Photo Poster\"\nshort_description: \"Create separate four-stage 3:4 design posters from photos\"\ndefault_prompt: \"Create a four-layer photo poster from each image I provide.\""}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene. Skill: four-layer-photo-poster Owner: tobewin Summary: Turn each supplied photo into its own 3:4 vertical, four-stage design poster with exact four-band assembly and a continuous source scene. Tags: latest:0.1.0 Version history: v0.1.0 | 2026-08-26T03:05:50.131Z | auto - Initial release of four-layer-photo-poster skill. - Creates one high-design 3:4 vertical poster per input photo, structured as four equal horizontal","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1306,"uniquenessScore":47,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T12:58:43.720Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T12:58:43.720Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-11T16:03:09.958Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}