{"id":"b0ee1dd7-abec-4f1d-9521-2c26ba3a38ed","entityType":"agent","slug":"clawhub-gpt-img-2-gpt-image-2-prompt-architect","name":"GPT Image 2 Prompt Architect","canonicalUrl":"https://www.xpersona.co/agent/clawhub-gpt-img-2-gpt-image-2-prompt-architect","canonicalPath":"/agent/clawhub-gpt-img-2-gpt-image-2-prompt-architect","generatedAt":"2026-10-11T10:52:52.360Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T07:42:44.410Z","emptyReason":null},"description":"Research real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops. Use for prompt search, prompt rewriting, ecommerce images, readable text, posters, social creatives, character sheets, storyboards, or image-to-video source frames; do not use for unrelated models or account support.","descriptionLabel":"Source description","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 s170hngh0a5eh4edhc6xmvg3vs84gc12:gpt-image-2-prompt-architect","sourceUrl":"https://clawhub.ai/gpt-img-2/gpt-image-2-prompt-architect","homepage":"https://clawhub.ai/gpt-img-2/skills/gpt-image-2-prompt-architect","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/gpt-img-2/gpt-image-2-prompt-architect","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/gpt-img-2/skills/gpt-image-2-prompt-architect","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"GPT Image 2 Prompt Architect technical dossier on Xpersona with agent coverage, OPENCLEW support, and live trust metadata."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T07:42:44.410Z","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":"2026-10-11T07:42:44.410Z","emptyReason":null},"stars":null,"forks":null,"downloads":1121,"packageName":null,"latestVersion":"1.1.0","tractionLabel":"1.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T07:42:44.308Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T07:42:44.410Z","lastCrawledAt":"2026-10-11T07:42:44.308Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T07:42:44.308Z","lastVerifiedAt":null,"highlights":[{"version":"1.1.0","createdAt":"2026-08-26T12:02:40.214Z","changelog":"Add optional read-only Image3 Prompt MCP search, canonical source attribution, and OpenClaw MCP setup.","fileCount":3,"zipByteSize":4704},{"version":"1.0.0","createdAt":"2026-05-19T09:16:37.977Z","changelog":"Initial release of gpt-image-2-prompt-architect: - Transforms rough AI image ideas into structured GPT Image 2 prompt packs. - Supports workflows for product photos, reference-image edits, UI/poster mockups, storyboards, and image-to-video frames. - Provides clear classification, essential info extraction, and revision guidance for stronger, more controlled prompt outputs. - Includes output templates for different prompt types and practical debugging advice. - References public GPTImg2.art documentation, demos, and galleries for further guidance.","fileCount":5,"zipByteSize":5190}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s170hngh0a5eh4edhc6xmvg3vs84gc12:gpt-image-2-prompt-architect","setupComplexity":"low","setupSteps":["Install using `clawhub skill install s170hngh0a5eh4edhc6xmvg3vs84gc12:gpt-image-2-prompt-architect` in an isolated environment before connecting it to live workloads.","No published capability contract is available yet, so validate auth and request/response behavior manually.","Review the upstream CLAWHUB listing at https://clawhub.ai/gpt-img-2/gpt-image-2-prompt-architect before using production credentials."],"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-gpt-img-2-gpt-image-2-prompt-architect/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T10:52:52.359Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gpt-img-2-gpt-image-2-prompt-architect/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":"medium","updatedAt":"2026-10-11T07:42:44.410Z","emptyReason":null},"readme":"Skill: GPT Image 2 Prompt Architect\n\nOwner: gpt-img-2\n\nSummary: Research real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops. Use for prompt search, prompt rewriting, ecommerce images, readable text, posters, social creatives, character sheets, storyboards, or image-to-video source frames; do not use for unrelated models or account support.\n\nTags: creative-workflow:1.0.0, design:1.0.0, ecommerce:1.0.0, gpt-image-2:1.1.0, image-generation:1.0.0, latest:1.1.0, mcp:1.1.0, openclaw:1.1.0, prompt-engineering:1.1.0\n\nVersion history:\n\nv1.1.0 | 2026-08-26T12:02:40.214Z | user\n\nAdd optional read-only Image3 Prompt MCP search, canonical source attribution, and OpenClaw MCP setup.\n\nv1.0.0 | 2026-05-19T09:16:37.977Z | user\n\nInitial release of gpt-image-2-prompt-architect:\n\n- Transforms rough AI image ideas into structured GPT Image 2 prompt packs.\n- Supports workflows for product photos, reference-image edits, UI/poster mockups, storyboards, and image-to-video frames.\n- Provides clear classification, essential info extraction, and revision guidance for stronger, more controlled prompt outputs.\n- Includes output templates for different prompt types and practical debugging advice.\n- References public GPTImg2.art documentation, demos, and galleries for further guidance.\n\nArchive index:\n\nArchive v1.1.0: 3 files, 4704 bytes\n\nFiles: skill-card.md (2992b), SKILL.md (7350b), _meta.json (147b)\n\nFile v1.1.0:SKILL.md\n\n---\nname: gpt-image-2-prompt-architect\ndescription: Research real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops. Use for prompt search, prompt rewriting, ecommerce images, readable text, posters, social creatives, character sheets, storyboards, or image-to-video source frames; do not use for unrelated models or account support.\nmetadata:\n openclaw:\n homepage: https://image3.org/docs/gpt-image-2-prompt-architect\n version: 1.1.0\n---\n\n# GPT Image 2 Prompt Architect\n\nThis skill turns loose creative ideas into cleaner GPT Image 2 prompt packs with stronger subject control, composition, text rendering, reference-image handling, and revision loops.\n\n## Canonical links\n\n- Docs: https://image3.org/docs/gpt-image-2-prompt-architect\n- Demo: https://image3.org/models/gpt-image-2\n- Create: https://image3.org/ai-image\n- Prompt gallery: https://image3.org/prompts/gpt-image-2\n- Raw SKILL.md: https://image3.org/skills/gpt-image-2-prompt-architect/SKILL.md\n- Prompt guide: https://image3.org/blog/gpt-image-2-prompt-guide\n- Product photo prompts: https://image3.org/blog/gpt-image-2-product-photo-prompts\n- Image-to-video workflow: https://image3.org/blog/gpt-image-2-image-to-video-workflow\n- Image3 Prompt MCP: https://github.com/gpt-img-2/image3-prompt-mcp\n- OpenClaw listing: https://clawhub.ai/gpt-img-2/skills/gpt-image-2-prompt-architect\n\n## Provenance and safety\n\n- Maintained around the public Image3 prompt workflow, prompt gallery, and documentation on `image3.org`.\n- The skill works as a text-only prompt workflow without any external tool.\n- The optional Image3 Prompt MCP is read-only, needs no API key, and never generates images or spends credits.\n- Keep the canonical Image3 source URL when sharing an example returned by the MCP.\n\n## When to use\n\n- The user has a rough AI image idea and wants a stronger GPT Image 2 prompt\n- The user wants product photos, ecommerce listing images, lifestyle ads, packaging mockups, or detail shots\n- The user needs UI mockups, posters, infographics, social media creatives, readable text, or branded layouts\n- The user is editing from reference images and needs identity, product, composition, or style preservation\n- The user wants source frames, character sheets, product references, or storyboard frames for image-to-video workflows\n- The user has unstable image outputs and needs diagnosis plus a cleaner second-pass prompt\n\n## When not to use\n\n- The request is mainly about a different model or non-image workflow\n- The user only wants final image generation, API integration, payment help, or account support\n- The user asks for unsupported model settings, hidden system behavior, or official provider claims\n\n## Workflow\n\n1. Classify the request:\n - text-to-image\n - reference-image edit\n - product photo or ecommerce visual\n - UI, poster, infographic, or readable-text layout\n - image-to-video source frame or storyboard\n2. Extract or ask for only the missing essentials:\n - subject or product\n - intended use\n - composition and camera/framing\n - environment or background\n - visual style and lighting\n - text that must appear exactly\n - reference-image constraints\n - aspect ratio or output format\n - hard negatives and brand safety constraints\n3. Keep the first draft focused:\n - one primary subject or product\n - one clear composition rule\n - one lighting or style direction\n - one concise constraint block\n4. Return a prompt pack with:\n - a brief diagnosis or strategy note\n - one primary prompt\n - 2 or 3 focused variants\n - a short avoid list\n - 3 concrete revision moves for the next round\n\n## Optional Image3 Prompt MCP\n\nWhen the Image3 MCP tools are available, use them only when real examples or gallery research would materially improve the answer:\n\n1. Call `search_prompts` with the user's task, language, and a focused result limit.\n2. Present concise candidates with title, preview, and canonical Image3 source URL.\n3. Call `get_prompt` only for the selected candidate; do not bulk-fetch the library.\n4. Use `build_prompt_brief` when the user's idea is rough, then adapt the result with the prompt construction rules below.\n5. Use `list_categories` only when the user is exploring rather than asking for a specific result.\n\nInstall the read-only MCP in OpenClaw:\n\n```bash\nopenclaw mcp add image3 \\\n --command npx \\\n --arg -y \\\n --arg github:gpt-img-2/image3-prompt-mcp \\\n --include 'search_prompts,get_prompt,list_categories,build_prompt_brief'\n\nopenclaw mcp doctor image3 --probe\n```\n\nThe MCP does not provide image generation. For a final image, use a generation capability already available in the user's host or direct the user to the Image3 generator; do not claim an image was generated when only a prompt was produced.\n\n## Prompt construction rules\n\n- Prefer concrete visual language over broad style adjectives.\n- Name the subject, product, materials, scale, framing, and lighting before adding mood.\n- For product photos, preserve label readability, product geometry, material texture, and commercial usability.\n- For reference-image edits, state what must remain unchanged before describing what may change.\n- For readable text, quote the exact text and keep the layout simple.\n- For UI mockups, describe the device, screen type, layout hierarchy, content density, and visual system.\n- For image-to-video source frames, prioritize stable identity, clear silhouette, coherent lighting, and simple motion-ready composition.\n- Avoid stacking many subjects, styles, camera angles, and layout goals into one prompt.\n- Do not invent unsupported model settings.\n\n## Output formats\n\n### Text-to-image\n\n```md\nGoal:\nSubject:\nComposition:\nEnvironment:\nStyle and lighting:\nText requirements:\nConstraints:\nPrompt:\n```\n\n### Reference-image edit\n\n```md\nReference anchor:\nWhat must stay stable:\nWhat may change:\nEdit direction:\nStyle and lighting:\nConstraints:\nPrompt:\n```\n\n### Product photo\n\n```md\nCommercial goal:\nProduct anchor:\nHero angle:\nBackground or scene:\nLighting:\nLabel and material rules:\nConstraints:\nPrompt:\n```\n\n### UI, poster, or readable-text layout\n\n```md\nFormat:\nAudience:\nLayout hierarchy:\nExact text:\nVisual system:\nConstraints:\nPrompt:\n```\n\n### Image-to-video source frame\n\n```md\nVideo goal:\nSource frame subject:\nMotion-ready composition:\nWhat must remain stable:\nLighting and style:\nConstraints:\nPrompt:\n```\n\n## Debugging heuristics\n\n- If the image is visually attractive but off-brief, rewrite around the intended use first.\n- If product geometry drifts, reduce scene complexity and strengthen product anchor language.\n- If text is wrong, shorten the text, quote it exactly, and simplify surrounding design.\n- If the subject changes identity, state preservation rules before the edit request.\n- If the composition is cluttered, reduce secondary objects and specify one dominant framing.\n- If the result cannot become a good video source frame, simplify pose, background, and lighting.\n\n## Response style\n\n- Be structured and concise.\n- Prefer prompt packs over long theory.\n- Offer practical variants that test one axis at a time: subject, composition, lighting, style, or constraints.\n- When external examples are useful, point the user to the canonical Image3 pages listed above.\n\nFile v1.1.0:_meta.json\n\n{\n \"ownerId\": \"kn7b3brv49z9vayew7kt8a5s8s84ghz7\",\n \"slug\": \"gpt-image-2-prompt-architect\",\n \"version\": \"1.1.0\",\n \"publishedAt\": 1787745760214\n}\n\nFile v1.1.0:skill-card.md\n\n## Description:\n\nResearch real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gpt-img-2](https://clawhub.ai/user/gpt-img-2)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, designers, marketers, and developers use this skill to turn rough GPT Image 2 image ideas into structured prompt packs for product visuals, reference-image edits, readable layouts, social creatives, storyboards, and image-to-video source frames. It also supports prompt example research and concise debugging loops for unstable image outputs.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The optional Image3 MCP setup runs an unpinned GitHub package.\n\nMitigation: Review or pin the GitHub package before installation, and use the text-only workflow when external tooling is not required.\n\nRisk: Optional prompt-search tools can send user prompt details to an external service.\n\nMitigation: Avoid sending confidential briefs, personal details, unreleased product plans, private URLs, customer names, or proprietary campaign text through the external prompt-search tools.\n\nRisk: Users may expect final image generation when the skill only produces prompt guidance.\n\nMitigation: Make clear that the skill returns prompt packs and example research; use a separate approved image-generation capability for final images.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gpt-img-2/skills/gpt-image-2-prompt-architect)\n- [Image3 skill documentation](https://image3.org/docs/gpt-image-2-prompt-architect)\n- [GPT Image 2 prompt gallery](https://image3.org/prompts/gpt-image-2)\n- [GPT Image 2 prompt guide](https://image3.org/blog/gpt-image-2-prompt-guide)\n- [Product photo prompts](https://image3.org/blog/gpt-image-2-product-photo-prompts)\n- [Image-to-video workflow](https://image3.org/blog/gpt-image-2-image-to-video-workflow)\n- [Image3 Prompt MCP](https://github.com/gpt-img-2/image3-prompt-mcp)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown prompt packs with structured fields, concise strategy notes, focused variants, avoid lists, revision moves, and optional MCP setup commands.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [The skill is text-only by default; optional prompt-search tooling can return candidate examples with canonical Image3 source URLs but does not generate images or spend credits.]\n\n## Skill Version(s):\n\n1.1.0 (source: server release evidence and OpenClaw metadata)\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\nArchive v1.0.0: 5 files, 5190 bytes\n\nFiles: agents/openai.yaml (248b), README.md (1433b), skill-card.md (3178b), SKILL.md (6000b), _meta.json (147b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: gpt-image-2-prompt-architect\ndescription: Turn rough AI image ideas into structured GPT Image 2 prompt packs, reference-image edit instructions, product photo prompts, UI mockup prompts, and debugging loops. Use when the user wants better AI image prompts for ecommerce visuals, readable text, posters, social media creatives, product mockups, character sheets, storyboards, headshots, image-to-video source frames, or GPT Image 2 prompt rewriting.\nmetadata:\n openclaw:\n homepage: https://gptimg2.art/docs/gpt-image-2-prompt-architect\n version: 1.0.0\n---\n\n# GPT Image 2 Prompt Architect\n\nThis skill turns loose creative ideas into cleaner GPT Image 2 prompt packs with stronger subject control, composition, text rendering, reference-image handling, and revision loops.\n\n## Canonical links\n\n- Docs: https://gptimg2.art/docs/gpt-image-2-prompt-architect\n- Demo: https://gptimg2.art/models/gpt-image-2\n- Create: https://gptimg2.art/ai-image\n- Prompt gallery: https://gptimg2.art/prompts/gpt-image-2\n- Raw SKILL.md: https://gptimg2.art/skills/gpt-image-2-prompt-architect/SKILL.md\n- Prompt guide: https://gptimg2.art/blog/gpt-image-2-prompt-guide\n- Product photo prompts: https://gptimg2.art/blog/gpt-image-2-product-photo-prompts\n- Image-to-video workflow: https://gptimg2.art/blog/gpt-image-2-image-to-video-workflow\n\n## Provenance and safety\n\n- Maintained around the public GPTImg2.art prompt workflow, prompt gallery, and documentation on `gptimg2.art`.\n- Text-only skill pack.\n- No helper scripts, no local binaries, no required environment variables, and no autonomous network calls.\n- It guides prompt design and references public pages only.\n\n## When to use\n\n- The user has a rough AI image idea and wants a stronger GPT Image 2 prompt\n- The user wants product photos, ecommerce listing images, lifestyle ads, packaging mockups, or detail shots\n- The user needs UI mockups, posters, infographics, social media creatives, readable text, or branded layouts\n- The user is editing from reference images and needs identity, product, composition, or style preservation\n- The user wants source frames, character sheets, product references, or storyboard frames for image-to-video workflows\n- The user has unstable image outputs and needs diagnosis plus a cleaner second-pass prompt\n\n## When not to use\n\n- The request is mainly about a different model or non-image workflow\n- The user only wants final image generation, API integration, payment help, or account support\n- The user asks for unsupported model settings, hidden system behavior, or official provider claims\n\n## Workflow\n\n1. Classify the request:\n - text-to-image\n - reference-image edit\n - product photo or ecommerce visual\n - UI, poster, infographic, or readable-text layout\n - image-to-video source frame or storyboard\n2. Extract or ask for only the missing essentials:\n - subject or product\n - intended use\n - composition and camera/framing\n - environment or background\n - visual style and lighting\n - text that must appear exactly\n - reference-image constraints\n - aspect ratio or output format\n - hard negatives and brand safety constraints\n3. Keep the first draft focused:\n - one primary subject or product\n - one clear composition rule\n - one lighting or style direction\n - one concise constraint block\n4. Return a prompt pack with:\n - a brief diagnosis or strategy note\n - one primary prompt\n - 2 or 3 focused variants\n - a short avoid list\n - 3 concrete revision moves for the next round\n\n## Prompt construction rules\n\n- Prefer concrete visual language over broad style adjectives.\n- Name the subject, product, materials, scale, framing, and lighting before adding mood.\n- For product photos, preserve label readability, product geometry, material texture, and commercial usability.\n- For reference-image edits, state what must remain unchanged before describing what may change.\n- For readable text, quote the exact text and keep the layout simple.\n- For UI mockups, describe the device, screen type, layout hierarchy, content density, and visual system.\n- For image-to-video source frames, prioritize stable identity, clear silhouette, coherent lighting, and simple motion-ready composition.\n- Avoid stacking many subjects, styles, camera angles, and layout goals into one prompt.\n- Do not invent unsupported model settings.\n\n## Output formats\n\n### Text-to-image\n\n```md\nGoal:\nSubject:\nComposition:\nEnvironment:\nStyle and lighting:\nText requirements:\nConstraints:\nPrompt:\n```\n\n### Reference-image edit\n\n```md\nReference anchor:\nWhat must stay stable:\nWhat may change:\nEdit direction:\nStyle and lighting:\nConstraints:\nPrompt:\n```\n\n### Product photo\n\n```md\nCommercial goal:\nProduct anchor:\nHero angle:\nBackground or scene:\nLighting:\nLabel and material rules:\nConstraints:\nPrompt:\n```\n\n### UI, poster, or readable-text layout\n\n```md\nFormat:\nAudience:\nLayout hierarchy:\nExact text:\nVisual system:\nConstraints:\nPrompt:\n```\n\n### Image-to-video source frame\n\n```md\nVideo goal:\nSource frame subject:\nMotion-ready composition:\nWhat must remain stable:\nLighting and style:\nConstraints:\nPrompt:\n```\n\n## Debugging heuristics\n\n- If the image is visually attractive but off-brief, rewrite around the intended use first.\n- If product geometry drifts, reduce scene complexity and strengthen product anchor language.\n- If text is wrong, shorten the text, quote it exactly, and simplify surrounding design.\n- If the subject changes identity, state preservation rules before the edit request.\n- If the composition is cluttered, reduce secondary objects and specify one dominant framing.\n- If the result cannot become a good video source frame, simplify pose, background, and lighting.\n\n## Response style\n\n- Be structured and concise.\n- Prefer prompt packs over long theory.\n- Offer practical variants that test one axis at a time: subject, composition, lighting, style, or constraints.\n- When external examples are useful, point the user to the canonical GPTImg2.art pages listed above.\n\nFile v1.0.0:README.md\n\n# GPT Image 2 Prompt Architect\n\n`gpt-image-2-prompt-architect` is a public prompt-design skill for turning rough AI image ideas into structured GPT Image 2 prompt packs, variants, and debugging loops.\n\n## Links\n\n- Docs: https://gptimg2.art/docs/gpt-image-2-prompt-architect\n- Demo: https://gptimg2.art/models/gpt-image-2\n- Create: https://gptimg2.art/ai-image\n- Prompt gallery: https://gptimg2.art/prompts/gpt-image-2\n- Raw SKILL.md: https://gptimg2.art/skills/gpt-image-2-prompt-architect/SKILL.md\n- Prompt guide: https://gptimg2.art/blog/gpt-image-2-prompt-guide\n- Product photo prompts: https://gptimg2.art/blog/gpt-image-2-product-photo-prompts\n- Image-to-video workflow: https://gptimg2.art/blog/gpt-image-2-image-to-video-workflow\n\n## What it does\n\n- Converts rough creative briefs into structured GPT Image 2 prompts\n- Handles text-to-image, reference-image edits, product photos, UI mockups, posters, and video source frames\n- Produces a clean base prompt plus focused variants\n- Adds short debugging and revision loops for unstable outputs\n\n## Safety profile\n\n- Text-only skill pack\n- No scripts\n- No binaries\n- No required environment variables\n- No autonomous network calls\n\n## Best fit\n\nUse this skill for ecommerce visuals, product photos, readable text layouts, social media creatives, UI mockups, posters, character sheets, storyboards, headshots, and image-to-video source frames that need stronger prompt structure.\n\nFile v1.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7b3brv49z9vayew7kt8a5s8s84ghz7\",\n \"slug\": \"gpt-image-2-prompt-architect\",\n \"version\": \"1.0.0\",\n \"publishedAt\": 1779182197977\n}\n\nFile v1.0.0:skill-card.md\n\n## Description:
\nTurn rough AI image ideas into structured GPT Image 2 prompt packs, reference-image edit instructions, product photo prompts, UI mockup prompts, and debugging loops.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[aitools101](https://clawhub.ai/user/aitools101)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nExternal creators, marketers, designers, and developers use this skill to turn rough image briefs into structured GPT Image 2 prompt packs, variants, and revision guidance for ecommerce, UI, poster, social media, storyboard, and image-to-video source-frame workflows.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Users may mistake the third-party GPTImg2.art prompt workflow for an official OpenAI tool or for provider account, payment, API, or support guidance.
\nMitigation: Present it as third-party prompt-writing guidance only, and verify model capabilities, account actions, payments, and API integration details against authoritative provider documentation before use.
\nRisk: Marketplace capability tags include purchase and crypto labels even though the artifact is a text-only prompt-writing skill.
\nMitigation: Do not use this skill to make purchases, conduct crypto activity, or authorize financial actions; limit use to drafting and revising image prompts.
\nRisk: Prompt packs can still produce off-brief images, poor text rendering, brand issues, or unsupported model-setting claims.
\nMitigation: Human-review generated prompts before production use, keep exact text and brand constraints explicit, and remove unsupported model settings or official-provider claims.
\n\n\n## Reference(s):
\n- [ClawHub Skill Page](https://clawhub.ai/aitools101/gpt-image-2-prompt-architect)
\n- [GPT Image 2 Prompt Architect Documentation](https://gptimg2.art/docs/gpt-image-2-prompt-architect)
\n- [GPT Image 2 Demo](https://gptimg2.art/models/gpt-image-2)
\n- [GPT Image 2 Prompt Gallery](https://gptimg2.art/prompts/gpt-image-2)
\n- [GPT Image 2 Prompt Guide](https://gptimg2.art/blog/gpt-image-2-prompt-guide)
\n- [GPT Image 2 Product Photo Prompts](https://gptimg2.art/blog/gpt-image-2-product-photo-prompts)
\n- [GPT Image 2 Image-to-Video Workflow](https://gptimg2.art/blog/gpt-image-2-image-to-video-workflow)
\n\n\n## Skill Output:
\n**Output Type(s):** [text, markdown, guidance]
\n**Output Format:** [Markdown prompt packs with structured fields, variants, avoid lists, and revision moves]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Text-only guidance; no scripts, binaries, required environment variables, or autonomous network calls are included in the artifact.]
\n\n## Skill Version(s):
\n1.0.0 (source: server release metadata and OpenClaw metadata)
\n\n## Ethical Considerations:
\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 v1.0.0:agents/openai.yaml\n\ninterface:\n display_name: 'GPT Image 2 Prompt Architect'\n short_description: 'Build structured GPT Image 2 prompt packs'\n default_prompt: 'Use $gpt-image-2-prompt-architect to turn my rough image idea into a structured GPT Image 2 prompt pack.'","readmeExcerpt":"Skill: GPT Image 2 Prompt Architect Owner: gpt-img-2 Summary: Research real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops. Use for prompt search, prompt rewriting, ecommerce images, readable text, posters, social creatives, character sheets, storyboards, or image-to-video source frames; do not use f","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"openclaw mcp add image3 \\\n --command npx \\\n --arg -y \\\n --arg github:gpt-img-2/image3-prompt-mcp \\\n --include 'search_prompts,get_prompt,list_categories,build_prompt_brief'\n\nopenclaw mcp doctor image3 --probe"},{"language":"md","snippet":"Goal:\nSubject:\nComposition:\nEnvironment:\nStyle and lighting:\nText requirements:\nConstraints:\nPrompt:"},{"language":"md","snippet":"Reference anchor:\nWhat must stay stable:\nWhat may change:\nEdit direction:\nStyle and lighting:\nConstraints:\nPrompt:"},{"language":"md","snippet":"Commercial goal:\nProduct anchor:\nHero angle:\nBackground or scene:\nLighting:\nLabel and material rules:\nConstraints:\nPrompt:"},{"language":"md","snippet":"Format:\nAudience:\nLayout hierarchy:\nExact text:\nVisual system:\nConstraints:\nPrompt:"},{"language":"md","snippet":"Video goal:\nSource frame subject:\nMotion-ready composition:\nWhat must remain stable:\nLighting and style:\nConstraints:\nPrompt:"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: gpt-image-2-prompt-architect\ndescription: Research real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops. Use for prompt search, prompt rewriting, ecommerce images, readable text, posters, social creatives, character sheets, storyboards, or image-to-video source frames; do not use for unrelated models or account support.\nmetadata:\n openclaw:\n homepage: https://image3.org/docs/gpt-image-2-prompt-architect\n version: 1.1.0\n---\n\n# GPT Image 2 Prompt Architect\n\nThis skill turns loose creative ideas into cleaner GPT Image 2 prompt packs with stronger subject control, composition, text rendering, reference-image handling, and revision loops.\n\n## Canonical links\n\n- Docs: https://image3.org/docs/gpt-image-2-prompt-architect\n- Demo: https://image3.org/models/gpt-image-2\n- Create: https://image3.org/ai-image\n- Prompt gallery: https://image3.org/prompts/gpt-image-2\n- Raw SKILL.md: https://image3.org/skills/gpt-image-2-prompt-architect/SKILL.md\n- Prompt guide: https://image3.org/blog/gpt-image-2-prompt-guide\n- Product photo prompts: https://image3.org/blog/gpt-image-2-product-photo-prompts\n- Image-to-video workflow: https://image3.org/blog/gpt-image-2-image-to-video-workflow\n- Image3 Prompt MCP: https://github.com/gpt-img-2/image3-prompt-mcp\n- OpenClaw listing: https://clawhub.ai/gpt-img-2/skills/gpt-image-2-prompt-architect\n\n## Provenance and safety\n\n- Maintained around the public Image3 prompt workflow, prompt gallery, and documentation on `image3.org`.\n- The skill works as a text-only prompt workflow without any external tool.\n- The optional Image3 Prompt MCP is read-only, needs no API key, and never generates images or spends credits.\n- Keep the canonical Image3 source URL when sharing an example returned by the MCP.\n\n## When to use\n\n- The user has a rough AI image idea and wants a stronger GPT Image 2 prompt\n- The user wants product photos, ecommerce listing images, lifestyle ads, packaging mockups, or detail shots\n- The user needs UI mockups, posters, infographics, social media creatives, readable text, or branded layouts\n- The user is editing from reference images and needs identity, product, composition, or style preservation\n- The user wants source frames, character sheets, product references, or storyboard frames for image-to-video workflows\n- The user has unstable image outputs and needs diagnosis plus a cleaner second-pass prompt\n\n## When not to use\n\n- The request is mainly about a different model or non-image workflow\n- The user only wants final image generation, API integration, payment help, or account support\n- The user asks for unsupported model settings, hidden system behavior, or official provider claims\n\n## Workflow\n\n1. Classify the request:\n - text-to-image\n - reference-image edit\n - product photo or ecommerce visual\n - UI, poster, infographic, or readable-text layout\n - image-to-video source fram"},{"path":"_meta.json","content":"{\n \"ownerId\": \"kn7b3brv49z9vayew7kt8a5s8s84ghz7\",\n \"slug\": \"gpt-image-2-prompt-architect\",\n \"version\": \"1.1.0\",\n \"publishedAt\": 1787745760214\n}"},{"path":"skill-card.md","content":"## Description:\n\nResearch real GPT Image 2 prompt examples and turn rough image ideas into structured prompt packs, reference-image edit instructions, product visuals, layouts, and debugging loops.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gpt-img-2](https://clawhub.ai/user/gpt-img-2)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, designers, marketers, and developers use this skill to turn rough GPT Image 2 image ideas into structured prompt packs for product visuals, reference-image edits, readable layouts, social creatives, storyboards, and image-to-video source frames. It also supports prompt example research and concise debugging loops for unstable image outputs.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The optional Image3 MCP setup runs an unpinned GitHub package.\n\nMitigation: Review or pin the GitHub package before installation, and use the text-only workflow when external tooling is not required.\n\nRisk: Optional prompt-search tools can send user prompt details to an external service.\n\nMitigation: Avoid sending confidential briefs, personal details, unreleased product plans, private URLs, customer names, or proprietary campaign text through the external prompt-search tools.\n\nRisk: Users may expect final image generation when the skill only produces prompt guidance.\n\nMitigation: Make clear that the skill returns prompt packs and example research; use a separate approved image-generation capability for final images.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gpt-img-2/skills/gpt-image-2-prompt-architect)\n- [Image3 skill documentation](https://image3.org/docs/gpt-image-2-prompt-architect)\n- [GPT Image 2 prompt gallery](https://image3.org/prompts/gpt-image-2)\n- [GPT Image 2 prompt guide](https://image3.org/blog/gpt-image-2-prompt-guide)\n- [Product photo prompts](https://image3.org/blog/gpt-image-2-product-photo-prompts)\n- [Image-to-video workflow](https://image3.org/blog/gpt-image-2-image-to-video-workflow)\n- [Image3 Prompt MCP](https://github.com/gpt-img-2/image3-prompt-mcp)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown prompt packs with structured fields, concise strategy notes, focused variants, avoid lists, revision moves, and optional MCP setup commands.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [The skill is text-only by default; optional prompt-search tooling can return candidate examples with canonical Image3 source URLs but does not generate images or spend credits.]\n\n## Skill Version(s):\n\n1.1.0 (source: server release evidence and OpenClaw metadata)\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."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":null,"editorialQuality":{"score":100,"threshold":65,"status":"thin","wordCount":1315,"uniquenessScore":41,"reasons":["uniqueness-below-45"]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T07:42:44.410Z","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":"2026-10-11T07:42:44.410Z","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":"2026-10-11T10:52:52.360Z","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"}]}}}