{"id":"0b30a5d6-aa19-404c-b13c-1b3779bfdbc8","entityType":"agent","slug":"clawhub-clarezoe-image-prompt-engineer","name":"Image Prompt Engineer","canonicalUrl":"https://www.xpersona.co/agent/clawhub-clarezoe-image-prompt-engineer","canonicalPath":"/agent/clawhub-clarezoe-image-prompt-engineer","generatedAt":"2026-10-11T14:13:42.156Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T11:21:09.768Z","emptyReason":null},"description":"Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari... Skill: Image Prompt Engineer Owner: clarezoe Summary: Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari... Tags: generative-ai:1.0.2, image-prompts:1.0.2, latest:1.0.2, midjourney:1.0.2, prompt-engineering:1.0.2 Version history: v1.0.2 | 2026-06-09T13:21:12.924Z | user Add OpenCode, Kiro, Devin, Codex, Cursor,","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 s178sk484g47vwcbvvfcjpftfx8461fz:image-prompt-engineer","sourceUrl":"https://clawhub.ai/clarezoe/image-prompt-engineer","homepage":"https://clawhub.ai/clarezoe/skills/image-prompt-engineer","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/clarezoe/image-prompt-engineer","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/clarezoe/skills/image-prompt-engineer","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari..."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T11:21:09.768Z","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":"2026-10-11T11:21:09.768Z","emptyReason":null},"stars":null,"forks":null,"downloads":1078,"packageName":null,"latestVersion":"1.0.2","tractionLabel":"1.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T11:21:09.755Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T11:21:09.768Z","lastCrawledAt":"2026-10-11T11:21:09.755Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T11:21:09.755Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.2","createdAt":"2026-06-09T13:21:12.924Z","changelog":"Add OpenCode, Kiro, Devin, Codex, Cursor, and Windsurf skill compatibility","fileCount":5,"zipByteSize":7545},{"version":"1.0.1","createdAt":"2026-06-09T13:08:40.875Z","changelog":"Normalize SKILL.md frontmatter for OpenClaw, Hermes, and Claude compatibility","fileCount":5,"zipByteSize":7524},{"version":"1.0.0","createdAt":"2026-06-09T12:59:41.199Z","changelog":"Add Midjourney-first image prompt engineering skill","fileCount":5,"zipByteSize":7648}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s178sk484g47vwcbvvfcjpftfx8461fz:image-prompt-engineer","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-clarezoe-image-prompt-engineer/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T14:13:42.155Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-image-prompt-engineer/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":"2026-10-11T11:21:09.768Z","emptyReason":null},"readme":"Skill: Image Prompt Engineer\n\nOwner: clarezoe\n\nSummary: Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari...\n\nTags: generative-ai:1.0.2, image-prompts:1.0.2, latest:1.0.2, midjourney:1.0.2, prompt-engineering:1.0.2\n\nVersion history:\n\nv1.0.2 | 2026-06-09T13:21:12.924Z | user\n\nAdd OpenCode, Kiro, Devin, Codex, Cursor, and Windsurf skill compatibility\n\nv1.0.1 | 2026-06-09T13:08:40.875Z | user\n\nNormalize SKILL.md frontmatter for OpenClaw, Hermes, and Claude compatibility\n\nv1.0.0 | 2026-06-09T12:59:41.199Z | user\n\nAdd Midjourney-first image prompt engineering skill\n\nArchive index:\n\nArchive v1.0.2: 5 files, 7545 bytes\n\nFiles: agents/openai.yaml (214b), references/midjourney-prompt-guide.md (6647b), skill-card.md (1902b), SKILL.md (6161b), _meta.json (140b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: \"image-prompt-engineer\"\ndescription: \"Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce variants of image-generation prompts, AI art prompts, Midjourney /imagine prompts, product shots, logos, interiors, editorial images, style prompts, negative prompts, or prompt packs. Default to Midjourney when the user does not specify an image model, and remind the user once.\"\nlicense: \"MIT\"\nmetadata: {\"version\":\"1.0.2\",\"category\":\"creative-tools\",\"tags\":[\"image-prompts\",\"midjourney\",\"prompt-engineering\",\"generative-ai\"],\"license\":\"MIT\",\"hermes\":{\"tags\":[\"image-prompts\",\"midjourney\",\"prompt-engineering\",\"generative-ai\"]}}\n---\n\n# Image Prompt Engineer\n\n## Overview\n\nGenerate precise image prompts that a user can paste directly into an image model. Default to Midjourney when the user does not name a model, and briefly state that default before the prompt.\n\nUse English for the final prompt text unless the user explicitly requests another prompt language. Match explanations, labels, and notes to the user's language.\n\n## Workflow\n\n1. Identify the target model.\n   - No model specified: use Midjourney and say once: `Defaulting to Midjourney.`\n   - Model specified: adapt to that model and omit Midjourney-only parameters.\n2. Infer missing creative details from the user's goal. Ask only when the missing detail changes the fundamental output, such as product identity, brand constraints, or prohibited content.\n3. Produce prompt-ready output, not a tutorial. Include short rationale only when it helps the user choose between variants.\n4. For Midjourney, place all parameters at the absolute end of the prompt, separated by spaces, with no punctuation inside parameter syntax.\n\n## Midjourney Prompt Shape\n\nUse this order for most Midjourney prompts:\n\n```text\n[primary subject and visual objective], [composition and camera/view], [materials and physical properties], [environment or background], [lighting], [style constraints], [technical constraints] --parameters\n```\n\nFront-load the most important concept in the first 10 to 30 tokens. Keep standard prompts around 30 to 80 tokens. Use 80 to 150 tokens only for highly specific layouts, product constraints, or complex spatial relationships.\n\nPrefer natural, grammatical descriptive prose over comma-stuffed keyword lists. Use underscores to bind words that must stay together as one concept, such as `matte_polycarbonate_housing`.\n\n## Midjourney Defaults\n\nDefault to the current Midjourney default model for general prompts unless the user asks for a version-specific feature. As of the June 2026 Midjourney docs, the default model is V7, while V8.1 is available for faster generation and HD images.\n\nUse these defaults for controlled, production-oriented prompts:\n\n- General controlled prompt: `--raw --stylize 50 --chaos 0`\n- Minimal, flat, logo, or UI-like output: `--raw --stylize 20 --chaos 0`\n- High-detail interior or environmental image: `--raw --stylize 100` to `--stylize 150`\n- V8.1 HD output: append `--v 8.1 --hd` when native 2K output matters more than V7-only features.\n- V7 feature workflows: append `--v 7` when using Omni Reference, or when a requested workflow depends on features absent from V8.1.\n\nUse `--raw`, not `--style raw`, because current Midjourney docs name the parameter Raw as `--raw`.\n\n## Version And Feature Choices\n\n- Use V7 by default for most Midjourney prompt generation.\n- Use V8.1 for fast clean renders, native HD, image prompts, style references, and prompt adherence without V7-only references.\n- Use V7 for Omni Reference: `--oref URL --ow 100`.\n- Use V6.1 or another supported version only when the user specifically needs a feature that the current docs limit to that version.\n- Check official Midjourney docs before asserting latest compatibility or default version in a time-sensitive answer.\n\n## Quality Rules\n\n- Replace subjective terms with physical and visual specifics.\n  - Avoid: `beautiful`, `high quality`, `epic lighting`, `photorealistic`, `hyperrealistic`, `8K`.\n  - Prefer: material, finish, geometry, lens, lighting physics, color palette, surface texture, and background treatment.\n- For minimalist, Scandinavian, or \"Honest UI\" aesthetics, describe physical restraint instead of using only the word `minimalist`.\n  - Use terms like `vast negative space`, `centered composition`, `orthographic view`, `soft diffuse North-European daylight`, `matte injection-molded polycarbonate`, `brushed anodized aluminum`, `Helvetica typography`, `Akzidenz-Grotesk typography`.\n- For logos and flat graphics, ban text artifacts and depth effects.\n- For product design, prioritize material honesty, scale, seam lines, tactile controls, and real lighting.\n- For architecture, use camera and space terms such as `24mm wide angle lens`, `floor-to-ceiling windows`, `raw concrete`, `single low-profile wooden bench`, `expansive negative space`.\n\n## Negative Controls\n\nUse `--no` for Midjourney exclusions when compatible with the target version:\n\n```text\n--no text, letters, words, signature, watermark, clutter, noise, shadow, gradient, depth, 3d, overprocessed\n```\n\nDo not put compound phrases in `--no` when each word would be harmful independently. Midjourney may read words in the negative block separately.\n\nFor weighted prompts, keep the total concept weight greater than zero:\n\n```text\nminimalist mechanical dial::2 soft shadow::0.5 clutter::-0.5 --raw --v 7\n```\n\n## Output Formats\n\nFor a single prompt request, return:\n\n```markdown\nDefaulting to Midjourney.\n\nPrompt:\n[paste-ready prompt]\n```\n\nFor variants, return compact numbered prompts with meaningful variant names:\n\n```markdown\nDefaulting to Midjourney.\n\n1. Product Poster\n[prompt]\n\n2. Editorial Still Life\n[prompt]\n```\n\nFor prompt refinement, return the improved prompt first, then a short `Changed:` line naming the most important fixes.\n\n## Reference\n\nRead `references/midjourney-prompt-guide.md` when the request needs Midjourney parameter details, version tradeoffs, prompt weighting, image/style references, negative prompt strategy, or minimalist product/logo/interior blueprints.\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"image-prompt-engineer\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1781011272924\n}\n\nFile v1.0.2:references/midjourney-prompt-guide.md\n\n# Midjourney Prompt Guide\n\nThis reference condenses the user's `Midjourney Prompt Engineering Deep Dive.pdf` and the current official Midjourney docs checked on June 9, 2026.\n\n## Core Model Behavior\n\n- Midjourney gives the earliest prompt tokens the strongest influence. Put the subject, visual objective, and must-have constraint first.\n- The 30 to 80 token range usually balances control and coherence. Longer prompts can work for complex layouts, but late details are easier to ignore.\n- Modern Midjourney prefers grammatical, descriptive prose. Avoid old keyword-stuffing prompt style.\n- Standard punctuation helps readability, but parameters must be clean command syntax at the end.\n- Use underscores to bind critical compound descriptors: `matte_polycarbonate_housing`, `North_European_overcast_daylight`.\n\n## Parameter Syntax\n\nPlace parameters at the end:\n\n```text\n[prompt text] --ar 1:1 --raw --stylize 30 --chaos 0 --v 7\n```\n\nDo not use commas in parameter syntax. Do not put prompt text after parameters.\n\nCommon parameters:\n\n- `--ar W:H`: aspect ratio. Use integers, not decimals.\n- `--raw`: reduce Midjourney's default aesthetic styling.\n- `--stylize` or `--s`: default styling strength. Lower values give more literal control.\n- `--chaos` or `--c`: variation across initial grid. Use `0` for predictable iteration.\n- `--weird` or `--w`: eccentricity. Avoid for clean design unless the user asks for strangeness.\n- `--stop`: stop generation early for softer and less defined output.\n- `--iw`: image prompt weight, from 0 to 3, default 1.\n- `--sref URL`: style reference. Transfers palette, medium, lighting, and visual feel.\n- `--sw N`: style reference weight, from 0 to 1000, default 100.\n- `--oref URL`: Omni Reference for V7 object, character, vehicle, or asset consistency.\n- `--ow N`: Omni Reference weight, from 1 to 1000, default 100. Keep below 400 for predictable results.\n- `--no`: excludes concepts. Similar to a negative weight.\n- `--v N`: model version.\n- `--hd`: V8.1 native 2K HD generation.\n\n## Version Notes\n\n- Official docs checked June 9, 2026 list V7 as the current default Midjourney version.\n- V8.1 was released April 30, 2026. It is faster and supports HD images with `--hd`.\n- V8.1 is useful for clean high-resolution images when V7-only features are not needed.\n- V7 supports Omni Reference and strong reference-controlled workflows.\n- Official docs list Multi-Prompting support in older model versions and show compatibility differences by version. Verify official docs before making a time-sensitive compatibility claim.\n\n## Weighting\n\nUse `::` to split concepts and weight them. The double colon has no space on the left and one space on the right when more prompt text follows.\n\n```text\nminimalist desktop radio receiver::2 soft diffuse overcast light::1 clutter::-0.5 --raw --v 7\n```\n\nRules:\n\n- Unspecified weight defaults to 1.\n- Negative weights suppress concepts.\n- The sum of all weights must be greater than zero.\n- Use weights when a secondary concept dominates the image.\n\n## Minimalist And Honest UI Strategy\n\nMidjourney often adds visual complexity because training data rewards detailed, dramatic images. Counter that with literal design constraints.\n\nUse:\n\n- `vast negative space`\n- `centered composition`\n- `orthographic view`\n- `top-down knolling layout`\n- `90-degree aerial view`\n- `flat lay presentation`\n- `axonometric drawing`\n- `axially symmetric`\n- `perfect circular geometry`\n- `monoline art`\n- `deep depth of field`\n- `zero-contrast soft shadows`\n- `soft diffuse North-European daylight`\n- `subdued ambient studio lighting`\n\nAvoid:\n\n- `beautiful`\n- `high quality`\n- `premium`\n- `epic lighting`\n- `photorealistic`\n- `hyperrealistic`\n- `superdetailed`\n- `4K`, `8K`, `16K`\n\nReplace abstract praise with physical description:\n\n- `matte light grey polycarbonate chassis`\n- `brushed anodized aluminum`\n- `sandblasted silver steel`\n- `raw unfinished concrete`\n- `translucent violet-tinted plastic`\n- `flush-fitting seams`\n- `radial precision dial`\n- `perfectly debossed alignment markings`\n- `Helvetica typography`\n- `Akzidenz-Grotesk typography`\n\n## Negative Blocks\n\nFor flat, restrained, logo, or UI-like output:\n\n```text\n--no depth, 3d, shading, gradient, texture, shadow, noise, clutter, volumetric, realistic, hyperdetailed\n```\n\nFor logos:\n\n```text\n--no text, letters, words, signature, watermark, background, shadow, gradient, depth, 3d, shading, photorealistic, noise\n```\n\nFor photography where some shadow is desirable, do not ban all shadows. Use a softer phrase:\n\n```text\nzero-contrast soft shadows\n```\n\n## Blueprint Patterns\n\nMinimal product/object poster:\n\n```text\ndesign specification sheet for a restrained desktop radio receiver, matte light grey polycarbonate chassis, perfectly circular dial, Akzidenz-Grotesk typography labeling, flat top-down knolling layout, centered on a solid matte white surface, soft diffuse overcast light, zero-contrast shadows, architectural orthographic composition --raw --ar 707:500 --stylize 30 --chaos 0 --no depth, 3d, shading, gradient, texture, shadow, noise, clutter, volumetric, realistic, hyperdetailed --v 7\n```\n\nGraphic symbol or logo:\n\n```text\nabstract geometric hexagon frame symbol, monoline line art, single color, flat design, clean vector emblem, black on solid white background --raw --ar 1:1 --stylize 20 --chaos 0 --no text, letters, words, signature, watermark, background, shadow, gradient, depth, 3d, shading, photorealistic, noise --v 7\n```\n\nArchitectural concept space:\n\n```text\narchitectural wide shot of a minimalist Scandinavian concrete pavilion interior, floor-to-ceiling windows, soft morning light through glass, expansive smooth raw concrete walls, a single low-profile wooden bench, vast empty negative space, quiet contemplative atmosphere, 24mm wide angle lens --raw --ar 141:100 --stylize 120 --chaos 0 --hd --v 8.1\n```\n\nEditorial fine art:\n\n```text\nabstract composition of paper structures, delicate overlapping matte white sheets, geometric sharp folds, linear arrangement, subtle cast shadows, monochromatic grayscale palette, soft diffused North-European daylight, studio fine art photography, macro lens view, deep depth of field --raw --ar 100:141 --stylize 150 --chaos 0 --no color, vibrant, saturation, background noise, grain --v 7\n```\n\nFunctionalist control panel:\n\n```text\ndesign blueprint graphic of flat circular control dials on a modular matrix, sandblasted silver aluminum panel, flush-fitting toggle switches, clean Helvetica typography, orthographic projection view, perfectly balanced grid layout, neutral grey and silver color scheme --raw --ar 141:100 --stylize 40 --chaos 0 --v 8.1\n```\n\nFile v1.0.2:skill-card.md\n\n## Description:\n\nGenerate production-ready image prompts for Midjourney and other image models.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[clarezoe](https://clawhub.ai/user/clarezoe)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and creators use this skill to create, refine, critique, translate, and vary image-generation prompts for Midjourney and other image models.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill defaults to Midjourney and English prompt text when the request is underspecified.\n\nMitigation: Specify the target image model and prompt language when those choices matter.\n\nRisk: Built-in examples and defaults may influence the generated aesthetic.\n\nMitigation: Provide brand constraints, prohibited content, and desired style direction for production prompts.\n\nRisk: Midjourney model defaults and parameter compatibility can change over time.\n\nMitigation: Check official Midjourney documentation before relying on time-sensitive default-version or feature-compatibility claims.\n\n## Reference(s):\n\n- [Midjourney Prompt Guide](references/midjourney-prompt-guide.md)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, guidance]\n\n**Output Format:** [Markdown prompt text with optional numbered variants and short change notes]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Defaults to Midjourney when no target model is specified; final prompt text defaults to English unless the user requests another language.]\n\n## Skill Version(s):\n\n1.0.2 (source: server release metadata and skill 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\nFile v1.0.2:agents/openai.yaml\n\ninterface:\n  display_name: \"Image Prompt Engineer\"\n  short_description: \"Generate production-ready image prompts\"\n  default_prompt: \"Use $image-prompt-engineer to turn this concept into a Midjourney image prompt.\"\n\nArchive v1.0.1: 5 files, 7524 bytes\n\nFiles: agents/openai.yaml (214b), references/midjourney-prompt-guide.md (6647b), skill-card.md (1953b), SKILL.md (6161b), _meta.json (140b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: \"image-prompt-engineer\"\ndescription: \"Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce variants of image-generation prompts, AI art prompts, Midjourney /imagine prompts, product shots, logos, interiors, editorial images, style prompts, negative prompts, or prompt packs. Default to Midjourney when the user does not specify an image model, and remind the user once.\"\nlicense: \"MIT\"\nmetadata: {\"version\":\"1.0.1\",\"category\":\"creative-tools\",\"tags\":[\"image-prompts\",\"midjourney\",\"prompt-engineering\",\"generative-ai\"],\"license\":\"MIT\",\"hermes\":{\"tags\":[\"image-prompts\",\"midjourney\",\"prompt-engineering\",\"generative-ai\"]}}\n---\n\n# Image Prompt Engineer\n\n## Overview\n\nGenerate precise image prompts that a user can paste directly into an image model. Default to Midjourney when the user does not name a model, and briefly state that default before the prompt.\n\nUse English for the final prompt text unless the user explicitly requests another prompt language. Match explanations, labels, and notes to the user's language.\n\n## Workflow\n\n1. Identify the target model.\n   - No model specified: use Midjourney and say once: `Defaulting to Midjourney.`\n   - Model specified: adapt to that model and omit Midjourney-only parameters.\n2. Infer missing creative details from the user's goal. Ask only when the missing detail changes the fundamental output, such as product identity, brand constraints, or prohibited content.\n3. Produce prompt-ready output, not a tutorial. Include short rationale only when it helps the user choose between variants.\n4. For Midjourney, place all parameters at the absolute end of the prompt, separated by spaces, with no punctuation inside parameter syntax.\n\n## Midjourney Prompt Shape\n\nUse this order for most Midjourney prompts:\n\n```text\n[primary subject and visual objective], [composition and camera/view], [materials and physical properties], [environment or background], [lighting], [style constraints], [technical constraints] --parameters\n```\n\nFront-load the most important concept in the first 10 to 30 tokens. Keep standard prompts around 30 to 80 tokens. Use 80 to 150 tokens only for highly specific layouts, product constraints, or complex spatial relationships.\n\nPrefer natural, grammatical descriptive prose over comma-stuffed keyword lists. Use underscores to bind words that must stay together as one concept, such as `matte_polycarbonate_housing`.\n\n## Midjourney Defaults\n\nDefault to the current Midjourney default model for general prompts unless the user asks for a version-specific feature. As of the June 2026 Midjourney docs, the default model is V7, while V8.1 is available for faster generation and HD images.\n\nUse these defaults for controlled, production-oriented prompts:\n\n- General controlled prompt: `--raw --stylize 50 --chaos 0`\n- Minimal, flat, logo, or UI-like output: `--raw --stylize 20 --chaos 0`\n- High-detail interior or environmental image: `--raw --stylize 100` to `--stylize 150`\n- V8.1 HD output: append `--v 8.1 --hd` when native 2K output matters more than V7-only features.\n- V7 feature workflows: append `--v 7` when using Omni Reference, or when a requested workflow depends on features absent from V8.1.\n\nUse `--raw`, not `--style raw`, because current Midjourney docs name the parameter Raw as `--raw`.\n\n## Version And Feature Choices\n\n- Use V7 by default for most Midjourney prompt generation.\n- Use V8.1 for fast clean renders, native HD, image prompts, style references, and prompt adherence without V7-only references.\n- Use V7 for Omni Reference: `--oref URL --ow 100`.\n- Use V6.1 or another supported version only when the user specifically needs a feature that the current docs limit to that version.\n- Check official Midjourney docs before asserting latest compatibility or default version in a time-sensitive answer.\n\n## Quality Rules\n\n- Replace subjective terms with physical and visual specifics.\n  - Avoid: `beautiful`, `high quality`, `epic lighting`, `photorealistic`, `hyperrealistic`, `8K`.\n  - Prefer: material, finish, geometry, lens, lighting physics, color palette, surface texture, and background treatment.\n- For minimalist, Scandinavian, or \"Honest UI\" aesthetics, describe physical restraint instead of using only the word `minimalist`.\n  - Use terms like `vast negative space`, `centered composition`, `orthographic view`, `soft diffuse North-European daylight`, `matte injection-molded polycarbonate`, `brushed anodized aluminum`, `Helvetica typography`, `Akzidenz-Grotesk typography`.\n- For logos and flat graphics, ban text artifacts and depth effects.\n- For product design, prioritize material honesty, scale, seam lines, tactile controls, and real lighting.\n- For architecture, use camera and space terms such as `24mm wide angle lens`, `floor-to-ceiling windows`, `raw concrete`, `single low-profile wooden bench`, `expansive negative space`.\n\n## Negative Controls\n\nUse `--no` for Midjourney exclusions when compatible with the target version:\n\n```text\n--no text, letters, words, signature, watermark, clutter, noise, shadow, gradient, depth, 3d, overprocessed\n```\n\nDo not put compound phrases in `--no` when each word would be harmful independently. Midjourney may read words in the negative block separately.\n\nFor weighted prompts, keep the total concept weight greater than zero:\n\n```text\nminimalist mechanical dial::2 soft shadow::0.5 clutter::-0.5 --raw --v 7\n```\n\n## Output Formats\n\nFor a single prompt request, return:\n\n```markdown\nDefaulting to Midjourney.\n\nPrompt:\n[paste-ready prompt]\n```\n\nFor variants, return compact numbered prompts with meaningful variant names:\n\n```markdown\nDefaulting to Midjourney.\n\n1. Product Poster\n[prompt]\n\n2. Editorial Still Life\n[prompt]\n```\n\nFor prompt refinement, return the improved prompt first, then a short `Changed:` line naming the most important fixes.\n\n## Reference\n\nRead `references/midjourney-prompt-guide.md` when the request needs Midjourney parameter details, version tradeoffs, prompt weighting, image/style references, negative prompt strategy, or minimalist product/logo/interior blueprints.\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"image-prompt-engineer\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1781010520875\n}\n\nFile v1.0.1:references/midjourney-prompt-guide.md\n\n# Midjourney Prompt Guide\n\nThis reference condenses the user's `Midjourney Prompt Engineering Deep Dive.pdf` and the current official Midjourney docs checked on June 9, 2026.\n\n## Core Model Behavior\n\n- Midjourney gives the earliest prompt tokens the strongest influence. Put the subject, visual objective, and must-have constraint first.\n- The 30 to 80 token range usually balances control and coherence. Longer prompts can work for complex layouts, but late details are easier to ignore.\n- Modern Midjourney prefers grammatical, descriptive prose. Avoid old keyword-stuffing prompt style.\n- Standard punctuation helps readability, but parameters must be clean command syntax at the end.\n- Use underscores to bind critical compound descriptors: `matte_polycarbonate_housing`, `North_European_overcast_daylight`.\n\n## Parameter Syntax\n\nPlace parameters at the end:\n\n```text\n[prompt text] --ar 1:1 --raw --stylize 30 --chaos 0 --v 7\n```\n\nDo not use commas in parameter syntax. Do not put prompt text after parameters.\n\nCommon parameters:\n\n- `--ar W:H`: aspect ratio. Use integers, not decimals.\n- `--raw`: reduce Midjourney's default aesthetic styling.\n- `--stylize` or `--s`: default styling strength. Lower values give more literal control.\n- `--chaos` or `--c`: variation across initial grid. Use `0` for predictable iteration.\n- `--weird` or `--w`: eccentricity. Avoid for clean design unless the user asks for strangeness.\n- `--stop`: stop generation early for softer and less defined output.\n- `--iw`: image prompt weight, from 0 to 3, default 1.\n- `--sref URL`: style reference. Transfers palette, medium, lighting, and visual feel.\n- `--sw N`: style reference weight, from 0 to 1000, default 100.\n- `--oref URL`: Omni Reference for V7 object, character, vehicle, or asset consistency.\n- `--ow N`: Omni Reference weight, from 1 to 1000, default 100. Keep below 400 for predictable results.\n- `--no`: excludes concepts. Similar to a negative weight.\n- `--v N`: model version.\n- `--hd`: V8.1 native 2K HD generation.\n\n## Version Notes\n\n- Official docs checked June 9, 2026 list V7 as the current default Midjourney version.\n- V8.1 was released April 30, 2026. It is faster and supports HD images with `--hd`.\n- V8.1 is useful for clean high-resolution images when V7-only features are not needed.\n- V7 supports Omni Reference and strong reference-controlled workflows.\n- Official docs list Multi-Prompting support in older model versions and show compatibility differences by version. Verify official docs before making a time-sensitive compatibility claim.\n\n## Weighting\n\nUse `::` to split concepts and weight them. The double colon has no space on the left and one space on the right when more prompt text follows.\n\n```text\nminimalist desktop radio receiver::2 soft diffuse overcast light::1 clutter::-0.5 --raw --v 7\n```\n\nRules:\n\n- Unspecified weight defaults to 1.\n- Negative weights suppress concepts.\n- The sum of all weights must be greater than zero.\n- Use weights when a secondary concept dominates the image.\n\n## Minimalist And Honest UI Strategy\n\nMidjourney often adds visual complexity because training data rewards detailed, dramatic images. Counter that with literal design constraints.\n\nUse:\n\n- `vast negative space`\n- `centered composition`\n- `orthographic view`\n- `top-down knolling layout`\n- `90-degree aerial view`\n- `flat lay presentation`\n- `axonometric drawing`\n- `axially symmetric`\n- `perfect circular geometry`\n- `monoline art`\n- `deep depth of field`\n- `zero-contrast soft shadows`\n- `soft diffuse North-European daylight`\n- `subdued ambient studio lighting`\n\nAvoid:\n\n- `beautiful`\n- `high quality`\n- `premium`\n- `epic lighting`\n- `photorealistic`\n- `hyperrealistic`\n- `superdetailed`\n- `4K`, `8K`, `16K`\n\nReplace abstract praise with physical description:\n\n- `matte light grey polycarbonate chassis`\n- `brushed anodized aluminum`\n- `sandblasted silver steel`\n- `raw unfinished concrete`\n- `translucent violet-tinted plastic`\n- `flush-fitting seams`\n- `radial precision dial`\n- `perfectly debossed alignment markings`\n- `Helvetica typography`\n- `Akzidenz-Grotesk typography`\n\n## Negative Blocks\n\nFor flat, restrained, logo, or UI-like output:\n\n```text\n--no depth, 3d, shading, gradient, texture, shadow, noise, clutter, volumetric, realistic, hyperdetailed\n```\n\nFor logos:\n\n```text\n--no text, letters, words, signature, watermark, background, shadow, gradient, depth, 3d, shading, photorealistic, noise\n```\n\nFor photography where some shadow is desirable, do not ban all shadows. Use a softer phrase:\n\n```text\nzero-contrast soft shadows\n```\n\n## Blueprint Patterns\n\nMinimal product/object poster:\n\n```text\ndesign specification sheet for a restrained desktop radio receiver, matte light grey polycarbonate chassis, perfectly circular dial, Akzidenz-Grotesk typography labeling, flat top-down knolling layout, centered on a solid matte white surface, soft diffuse overcast light, zero-contrast shadows, architectural orthographic composition --raw --ar 707:500 --stylize 30 --chaos 0 --no depth, 3d, shading, gradient, texture, shadow, noise, clutter, volumetric, realistic, hyperdetailed --v 7\n```\n\nGraphic symbol or logo:\n\n```text\nabstract geometric hexagon frame symbol, monoline line art, single color, flat design, clean vector emblem, black on solid white background --raw --ar 1:1 --stylize 20 --chaos 0 --no text, letters, words, signature, watermark, background, shadow, gradient, depth, 3d, shading, photorealistic, noise --v 7\n```\n\nArchitectural concept space:\n\n```text\narchitectural wide shot of a minimalist Scandinavian concrete pavilion interior, floor-to-ceiling windows, soft morning light through glass, expansive smooth raw concrete walls, a single low-profile wooden bench, vast empty negative space, quiet contemplative atmosphere, 24mm wide angle lens --raw --ar 141:100 --stylize 120 --chaos 0 --hd --v 8.1\n```\n\nEditorial fine art:\n\n```text\nabstract composition of paper structures, delicate overlapping matte white sheets, geometric sharp folds, linear arrangement, subtle cast shadows, monochromatic grayscale palette, soft diffused North-European daylight, studio fine art photography, macro lens view, deep depth of field --raw --ar 100:141 --stylize 150 --chaos 0 --no color, vibrant, saturation, background noise, grain --v 7\n```\n\nFunctionalist control panel:\n\n```text\ndesign blueprint graphic of flat circular control dials on a modular matrix, sandblasted silver aluminum panel, flush-fitting toggle switches, clean Helvetica typography, orthographic projection view, perfectly balanced grid layout, neutral grey and silver color scheme --raw --ar 141:100 --stylize 40 --chaos 0 --v 8.1\n```\n\nFile v1.0.1:skill-card.md\n\n## Description: <br>\nGenerate production-ready image prompts for Midjourney and other image models. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[clarezoe](https://clawhub.ai/user/clarezoe) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nExternal users and creative teams use this skill to create, refine, translate, critique, and produce variants of prompt-ready image-generation prompts for Midjourney and other image models. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Midjourney model defaults and feature compatibility can change over time. <br>\nMitigation: Verify official Midjourney documentation before relying on latest version-specific parameters or compatibility. <br>\nRisk: The skill defaults unspecified image-generation requests to Midjourney, which may not match the user's intended model. <br>\nMitigation: Confirm the target model when model-specific syntax matters and omit Midjourney-only parameters for other models. <br>\n\n\n## Reference(s): <br>\n- [Midjourney Prompt Guide](references/midjourney-prompt-guide.md) <br>\n- [ClawHub Skill Page](https://clawhub.ai/clarezoe/image-prompt-engineer) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Text, Markdown, Guidance] <br>\n**Output Format:** [Markdown prompt text with optional numbered variants and short change notes] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [For Midjourney prompts, model parameters are placed at the end of the prompt when applicable.] <br>\n\n## Skill Version(s): <br>\n1.0.1 (source: server release metadata and skill metadata) <br>\n\n## Ethical Considerations: <br>\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. <br>\n\nFile v1.0.1:agents/openai.yaml\n\ninterface:\n  display_name: \"Image Prompt Engineer\"\n  short_description: \"Generate production-ready image prompts\"\n  default_prompt: \"Use $image-prompt-engineer to turn this concept into a Midjourney image prompt.\"\n\nArchive v1.0.0: 5 files, 7648 bytes\n\nFiles: agents/openai.yaml (214b), references/midjourney-prompt-guide.md (6647b), skill-card.md (2256b), SKILL.md (6064b), _meta.json (140b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: image-prompt-engineer\ndescription: Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce variants of image-generation prompts, AI art prompts, Midjourney /imagine prompts, product shots, logos, interiors, editorial images, style prompts, negative prompts, or prompt packs. Default to Midjourney when the user does not specify an image model, and remind the user once.\nlicense: MIT\nmetadata:\n  version: 1.0.0\n  category: creative-tools\n  tags:\n    - image-prompts\n    - midjourney\n    - prompt-engineering\n    - generative-ai\n---\n\n# Image Prompt Engineer\n\n## Overview\n\nGenerate precise image prompts that a user can paste directly into an image model. Default to Midjourney when the user does not name a model, and briefly state that default before the prompt.\n\nUse English for the final prompt text unless the user explicitly requests another prompt language. Match explanations, labels, and notes to the user's language.\n\n## Workflow\n\n1. Identify the target model.\n   - No model specified: use Midjourney and say once: `Defaulting to Midjourney.`\n   - Model specified: adapt to that model and omit Midjourney-only parameters.\n2. Infer missing creative details from the user's goal. Ask only when the missing detail changes the fundamental output, such as product identity, brand constraints, or prohibited content.\n3. Produce prompt-ready output, not a tutorial. Include short rationale only when it helps the user choose between variants.\n4. For Midjourney, place all parameters at the absolute end of the prompt, separated by spaces, with no punctuation inside parameter syntax.\n\n## Midjourney Prompt Shape\n\nUse this order for most Midjourney prompts:\n\n```text\n[primary subject and visual objective], [composition and camera/view], [materials and physical properties], [environment or background], [lighting], [style constraints], [technical constraints] --parameters\n```\n\nFront-load the most important concept in the first 10 to 30 tokens. Keep standard prompts around 30 to 80 tokens. Use 80 to 150 tokens only for highly specific layouts, product constraints, or complex spatial relationships.\n\nPrefer natural, grammatical descriptive prose over comma-stuffed keyword lists. Use underscores to bind words that must stay together as one concept, such as `matte_polycarbonate_housing`.\n\n## Midjourney Defaults\n\nDefault to the current Midjourney default model for general prompts unless the user asks for a version-specific feature. As of the June 2026 Midjourney docs, the default model is V7, while V8.1 is available for faster generation and HD images.\n\nUse these defaults for controlled, production-oriented prompts:\n\n- General controlled prompt: `--raw --stylize 50 --chaos 0`\n- Minimal, flat, logo, or UI-like output: `--raw --stylize 20 --chaos 0`\n- High-detail interior or environmental image: `--raw --stylize 100` to `--stylize 150`\n- V8.1 HD output: append `--v 8.1 --hd` when native 2K output matters more than V7-only features.\n- V7 feature workflows: append `--v 7` when using Omni Reference, or when a requested workflow depends on features absent from V8.1.\n\nUse `--raw`, not `--style raw`, because current Midjourney docs name the parameter Raw as `--raw`.\n\n## Version And Feature Choices\n\n- Use V7 by default for most Midjourney prompt generation.\n- Use V8.1 for fast clean renders, native HD, image prompts, style references, and prompt adherence without V7-only references.\n- Use V7 for Omni Reference: `--oref URL --ow 100`.\n- Use V6.1 or another supported version only when the user specifically needs a feature that the current docs limit to that version.\n- Check official Midjourney docs before asserting latest compatibility or default version in a time-sensitive answer.\n\n## Quality Rules\n\n- Replace subjective terms with physical and visual specifics.\n  - Avoid: `beautiful`, `high quality`, `epic lighting`, `photorealistic`, `hyperrealistic`, `8K`.\n  - Prefer: material, finish, geometry, lens, lighting physics, color palette, surface texture, and background treatment.\n- For minimalist, Scandinavian, or \"Honest UI\" aesthetics, describe physical restraint instead of using only the word `minimalist`.\n  - Use terms like `vast negative space`, `centered composition`, `orthographic view`, `soft diffuse North-European daylight`, `matte injection-molded polycarbonate`, `brushed anodized aluminum`, `Helvetica typography`, `Akzidenz-Grotesk typography`.\n- For logos and flat graphics, ban text artifacts and depth effects.\n- For product design, prioritize material honesty, scale, seam lines, tactile controls, and real lighting.\n- For architecture, use camera and space terms such as `24mm wide angle lens`, `floor-to-ceiling windows`, `raw concrete`, `single low-profile wooden bench`, `expansive negative space`.\n\n## Negative Controls\n\nUse `--no` for Midjourney exclusions when compatible with the target version:\n\n```text\n--no text, letters, words, signature, watermark, clutter, noise, shadow, gradient, depth, 3d, overprocessed\n```\n\nDo not put compound phrases in `--no` when each word would be harmful independently. Midjourney may read words in the negative block separately.\n\nFor weighted prompts, keep the total concept weight greater than zero:\n\n```text\nminimalist mechanical dial::2 soft shadow::0.5 clutter::-0.5 --raw --v 7\n```\n\n## Output Formats\n\nFor a single prompt request, return:\n\n```markdown\nDefaulting to Midjourney.\n\nPrompt:\n[paste-ready prompt]\n```\n\nFor variants, return compact numbered prompts with meaningful variant names:\n\n```markdown\nDefaulting to Midjourney.\n\n1. Product Poster\n[prompt]\n\n2. Editorial Still Life\n[prompt]\n```\n\nFor prompt refinement, return the improved prompt first, then a short `Changed:` line naming the most important fixes.\n\n## Reference\n\nRead `references/midjourney-prompt-guide.md` when the request needs Midjourney parameter details, version tradeoffs, prompt weighting, image/style references, negative prompt strategy, or minimalist product/logo/interior blueprints.\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"image-prompt-engineer\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1781009981199\n}\n\nFile v1.0.0:references/midjourney-prompt-guide.md\n\n# Midjourney Prompt Guide\n\nThis reference condenses the user's `Midjourney Prompt Engineering Deep Dive.pdf` and the current official Midjourney docs checked on June 9, 2026.\n\n## Core Model Behavior\n\n- Midjourney gives the earliest prompt tokens the strongest influence. Put the subject, visual objective, and must-have constraint first.\n- The 30 to 80 token range usually balances control and coherence. Longer prompts can work for complex layouts, but late details are easier to ignore.\n- Modern Midjourney prefers grammatical, descriptive prose. Avoid old keyword-stuffing prompt style.\n- Standard punctuation helps readability, but parameters must be clean command syntax at the end.\n- Use underscores to bind critical compound descriptors: `matte_polycarbonate_housing`, `North_European_overcast_daylight`.\n\n## Parameter Syntax\n\nPlace parameters at the end:\n\n```text\n[prompt text] --ar 1:1 --raw --stylize 30 --chaos 0 --v 7\n```\n\nDo not use commas in parameter syntax. Do not put prompt text after parameters.\n\nCommon parameters:\n\n- `--ar W:H`: aspect ratio. Use integers, not decimals.\n- `--raw`: reduce Midjourney's default aesthetic styling.\n- `--stylize` or `--s`: default styling strength. Lower values give more literal control.\n- `--chaos` or `--c`: variation across initial grid. Use `0` for predictable iteration.\n- `--weird` or `--w`: eccentricity. Avoid for clean design unless the user asks for strangeness.\n- `--stop`: stop generation early for softer and less defined output.\n- `--iw`: image prompt weight, from 0 to 3, default 1.\n- `--sref URL`: style reference. Transfers palette, medium, lighting, and visual feel.\n- `--sw N`: style reference weight, from 0 to 1000, default 100.\n- `--oref URL`: Omni Reference for V7 object, character, vehicle, or asset consistency.\n- `--ow N`: Omni Reference weight, from 1 to 1000, default 100. Keep below 400 for predictable results.\n- `--no`: excludes concepts. Similar to a negative weight.\n- `--v N`: model version.\n- `--hd`: V8.1 native 2K HD generation.\n\n## Version Notes\n\n- Official docs checked June 9, 2026 list V7 as the current default Midjourney version.\n- V8.1 was released April 30, 2026. It is faster and supports HD images with `--hd`.\n- V8.1 is useful for clean high-resolution images when V7-only features are not needed.\n- V7 supports Omni Reference and strong reference-controlled workflows.\n- Official docs list Multi-Prompting support in older model versions and show compatibility differences by version. Verify official docs before making a time-sensitive compatibility claim.\n\n## Weighting\n\nUse `::` to split concepts and weight them. The double colon has no space on the left and one space on the right when more prompt text follows.\n\n```text\nminimalist desktop radio receiver::2 soft diffuse overcast light::1 clutter::-0.5 --raw --v 7\n```\n\nRules:\n\n- Unspecified weight defaults to 1.\n- Negative weights suppress concepts.\n- The sum of all weights must be greater than zero.\n- Use weights when a secondary concept dominates the image.\n\n## Minimalist And Honest UI Strategy\n\nMidjourney often adds visual complexity because training data rewards detailed, dramatic images. Counter that with literal design constraints.\n\nUse:\n\n- `vast negative space`\n- `centered composition`\n- `orthographic view`\n- `top-down knolling layout`\n- `90-degree aerial view`\n- `flat lay presentation`\n- `axonometric drawing`\n- `axially symmetric`\n- `perfect circular geometry`\n- `monoline art`\n- `deep depth of field`\n- `zero-contrast soft shadows`\n- `soft diffuse North-European daylight`\n- `subdued ambient studio lighting`\n\nAvoid:\n\n- `beautiful`\n- `high quality`\n- `premium`\n- `epic lighting`\n- `photorealistic`\n- `hyperrealistic`\n- `superdetailed`\n- `4K`, `8K`, `16K`\n\nReplace abstract praise with physical description:\n\n- `matte light grey polycarbonate chassis`\n- `brushed anodized aluminum`\n- `sandblasted silver steel`\n- `raw unfinished concrete`\n- `translucent violet-tinted plastic`\n- `flush-fitting seams`\n- `radial precision dial`\n- `perfectly debossed alignment markings`\n- `Helvetica typography`\n- `Akzidenz-Grotesk typography`\n\n## Negative Blocks\n\nFor flat, restrained, logo, or UI-like output:\n\n```text\n--no depth, 3d, shading, gradient, texture, shadow, noise, clutter, volumetric, realistic, hyperdetailed\n```\n\nFor logos:\n\n```text\n--no text, letters, words, signature, watermark, background, shadow, gradient, depth, 3d, shading, photorealistic, noise\n```\n\nFor photography where some shadow is desirable, do not ban all shadows. Use a softer phrase:\n\n```text\nzero-contrast soft shadows\n```\n\n## Blueprint Patterns\n\nMinimal product/object poster:\n\n```text\ndesign specification sheet for a restrained desktop radio receiver, matte light grey polycarbonate chassis, perfectly circular dial, Akzidenz-Grotesk typography labeling, flat top-down knolling layout, centered on a solid matte white surface, soft diffuse overcast light, zero-contrast shadows, architectural orthographic composition --raw --ar 707:500 --stylize 30 --chaos 0 --no depth, 3d, shading, gradient, texture, shadow, noise, clutter, volumetric, realistic, hyperdetailed --v 7\n```\n\nGraphic symbol or logo:\n\n```text\nabstract geometric hexagon frame symbol, monoline line art, single color, flat design, clean vector emblem, black on solid white background --raw --ar 1:1 --stylize 20 --chaos 0 --no text, letters, words, signature, watermark, background, shadow, gradient, depth, 3d, shading, photorealistic, noise --v 7\n```\n\nArchitectural concept space:\n\n```text\narchitectural wide shot of a minimalist Scandinavian concrete pavilion interior, floor-to-ceiling windows, soft morning light through glass, expansive smooth raw concrete walls, a single low-profile wooden bench, vast empty negative space, quiet contemplative atmosphere, 24mm wide angle lens --raw --ar 141:100 --stylize 120 --chaos 0 --hd --v 8.1\n```\n\nEditorial fine art:\n\n```text\nabstract composition of paper structures, delicate overlapping matte white sheets, geometric sharp folds, linear arrangement, subtle cast shadows, monochromatic grayscale palette, soft diffused North-European daylight, studio fine art photography, macro lens view, deep depth of field --raw --ar 100:141 --stylize 150 --chaos 0 --no color, vibrant, saturation, background noise, grain --v 7\n```\n\nFunctionalist control panel:\n\n```text\ndesign blueprint graphic of flat circular control dials on a modular matrix, sandblasted silver aluminum panel, flush-fitting toggle switches, clean Helvetica typography, orthographic projection view, perfectly balanced grid layout, neutral grey and silver color scheme --raw --ar 141:100 --stylize 40 --chaos 0 --v 8.1\n```\n\nFile v1.0.0:skill-card.md\n\n## Description: <br>\nGenerate production-ready image prompts for Midjourney and other image models. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[clarezoe](https://clawhub.ai/user/clarezoe) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nCreative teams, designers, marketers, and developers use this skill to turn image concepts into paste-ready prompts for Midjourney and other image-generation models. It supports creating, refining, translating, critiquing, and producing prompt variants for product shots, logos, interiors, editorial imagery, style prompts, negative prompts, and prompt packs. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The skill may apply Midjourney defaults to broad image-prompt requests when the user does not specify a model. <br>\nMitigation: Confirm the target image model when model-specific behavior matters, and remove Midjourney-only parameters for other models. <br>\nRisk: Model defaults and parameter compatibility can change over time. <br>\nMitigation: Verify official model documentation before making time-sensitive claims about the latest Midjourney version, default model, or feature support. <br>\n\n\n## Reference(s): <br>\n- [Midjourney Prompt Guide](artifact/references/midjourney-prompt-guide.md) <br>\n- [ClawHub skill page](https://clawhub.ai/clarezoe/image-prompt-engineer) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown prompt text, often with numbered variants or a short refinement note.] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [Defaults to Midjourney when no model is specified, keeps image-model parameters at the end of the prompt, and may include concise rationale when it helps the user choose between variants.] <br>\n\n## Skill Version(s): <br>\n1.0.0 (source: ClawHub release metadata and skill metadata) <br>\n\n## Ethical Considerations: <br>\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. <br>\n\nFile v1.0.0:agents/openai.yaml\n\ninterface:\n  display_name: \"Image Prompt Engineer\"\n  short_description: \"Generate production-ready image prompts\"\n  default_prompt: \"Use $image-prompt-engineer to turn this concept into a Midjourney image prompt.\"","readmeExcerpt":"Skill: Image Prompt Engineer Owner: clarezoe Summary: Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari... Tags: generative-ai:1.0.2, image-prompts:1.0.2, latest:1.0.2, midjourney:1.0.2, prompt-engineering:1.0.2 Version history: v1.0.2 | 2026-06-09T13:21:12.924Z | user Add OpenCode, Kiro, Devin, Codex, Cursor, ","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"[primary subject and visual objective], [composition and camera/view], [materials and physical properties], [environment or background], [lighting], [style constraints], [technical constraints] --parameters"},{"language":"text","snippet":"--no text, letters, words, signature, watermark, clutter, noise, shadow, gradient, depth, 3d, overprocessed"},{"language":"text","snippet":"minimalist mechanical dial::2 soft shadow::0.5 clutter::-0.5 --raw --v 7"},{"language":"markdown","snippet":"Defaulting to Midjourney.\n\nPrompt:\n[paste-ready prompt]"},{"language":"markdown","snippet":"Defaulting to Midjourney.\n\n1. Product Poster\n[prompt]\n\n2. Editorial Still Life\n[prompt]"},{"language":"text","snippet":"[prompt text] --ar 1:1 --raw --stylize 30 --chaos 0 --v 7"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: \"image-prompt-engineer\"\ndescription: \"Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce variants of image-generation prompts, AI art prompts, Midjourney /imagine prompts, product shots, logos, interiors, editorial images, style prompts, negative prompts, or prompt packs. Default to Midjourney when the user does not specify an image model, and remind the user once.\"\nlicense: \"MIT\"\nmetadata: {\"version\":\"1.0.2\",\"category\":\"creative-tools\",\"tags\":[\"image-prompts\",\"midjourney\",\"prompt-engineering\",\"generative-ai\"],\"license\":\"MIT\",\"hermes\":{\"tags\":[\"image-prompts\",\"midjourney\",\"prompt-engineering\",\"generative-ai\"]}}\n---\n\n# Image Prompt Engineer\n\n## Overview\n\nGenerate precise image prompts that a user can paste directly into an image model. Default to Midjourney when the user does not name a model, and briefly state that default before the prompt.\n\nUse English for the final prompt text unless the user explicitly requests another prompt language. Match explanations, labels, and notes to the user's language.\n\n## Workflow\n\n1. Identify the target model.\n   - No model specified: use Midjourney and say once: `Defaulting to Midjourney.`\n   - Model specified: adapt to that model and omit Midjourney-only parameters.\n2. Infer missing creative details from the user's goal. Ask only when the missing detail changes the fundamental output, such as product identity, brand constraints, or prohibited content.\n3. Produce prompt-ready output, not a tutorial. Include short rationale only when it helps the user choose between variants.\n4. For Midjourney, place all parameters at the absolute end of the prompt, separated by spaces, with no punctuation inside parameter syntax.\n\n## Midjourney Prompt Shape\n\nUse this order for most Midjourney prompts:\n\n```text\n[primary subject and visual objective], [composition and camera/view], [materials and physical properties], [environment or background], [lighting], [style constraints], [technical constraints] --parameters\n```\n\nFront-load the most important concept in the first 10 to 30 tokens. Keep standard prompts around 30 to 80 tokens. Use 80 to 150 tokens only for highly specific layouts, product constraints, or complex spatial relationships.\n\nPrefer natural, grammatical descriptive prose over comma-stuffed keyword lists. Use underscores to bind words that must stay together as one concept, such as `matte_polycarbonate_housing`.\n\n## Midjourney Defaults\n\nDefault to the current Midjourney default model for general prompts unless the user asks for a version-specific feature. As of the June 2026 Midjourney docs, the default model is V7, while V8.1 is available for faster generation and HD images.\n\nUse these defaults for controlled, production-oriented prompts:\n\n- General controlled prompt: `--raw --stylize 50 --chaos 0`\n- Minimal, flat, logo, or UI-like output: `--raw --stylize 20 --chaos 0`\n- High-detail interior or environmen"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"image-prompt-engineer\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1781011272924\n}"},{"path":"references/midjourney-prompt-guide.md","content":"# Midjourney Prompt Guide\n\nThis reference condenses the user's `Midjourney Prompt Engineering Deep Dive.pdf` and the current official Midjourney docs checked on June 9, 2026.\n\n## Core Model Behavior\n\n- Midjourney gives the earliest prompt tokens the strongest influence. Put the subject, visual objective, and must-have constraint first.\n- The 30 to 80 token range usually balances control and coherence. Longer prompts can work for complex layouts, but late details are easier to ignore.\n- Modern Midjourney prefers grammatical, descriptive prose. Avoid old keyword-stuffing prompt style.\n- Standard punctuation helps readability, but parameters must be clean command syntax at the end.\n- Use underscores to bind critical compound descriptors: `matte_polycarbonate_housing`, `North_European_overcast_daylight`.\n\n## Parameter Syntax\n\nPlace parameters at the end:\n\n```text\n[prompt text] --ar 1:1 --raw --stylize 30 --chaos 0 --v 7\n```\n\nDo not use commas in parameter syntax. Do not put prompt text after parameters.\n\nCommon parameters:\n\n- `--ar W:H`: aspect ratio. Use integers, not decimals.\n- `--raw`: reduce Midjourney's default aesthetic styling.\n- `--stylize` or `--s`: default styling strength. Lower values give more literal control.\n- `--chaos` or `--c`: variation across initial grid. Use `0` for predictable iteration.\n- `--weird` or `--w`: eccentricity. Avoid for clean design unless the user asks for strangeness.\n- `--stop`: stop generation early for softer and less defined output.\n- `--iw`: image prompt weight, from 0 to 3, default 1.\n- `--sref URL`: style reference. Transfers palette, medium, lighting, and visual feel.\n- `--sw N`: style reference weight, from 0 to 1000, default 100.\n- `--oref URL`: Omni Reference for V7 object, character, vehicle, or asset consistency.\n- `--ow N`: Omni Reference weight, from 1 to 1000, default 100. Keep below 400 for predictable results.\n- `--no`: excludes concepts. Similar to a negative weight.\n- `--v N`: model version.\n- `--hd`: V8.1 native 2K HD generation.\n\n## Version Notes\n\n- Official docs checked June 9, 2026 list V7 as the current default Midjourney version.\n- V8.1 was released April 30, 2026. It is faster and supports HD images with `--hd`.\n- V8.1 is useful for clean high-resolution images when V7-only features are not needed.\n- V7 supports Omni Reference and strong reference-controlled workflows.\n- Official docs list Multi-Prompting support in older model versions and show compatibility differences by version. Verify official docs before making a time-sensitive compatibility claim.\n\n## Weighting\n\nUse `::` to split concepts and weight them. The double colon has no space on the left and one space on the right when more prompt text follows.\n\n```text\nminimalist desktop radio receiver::2 soft diffuse overcast light::1 clutter::-0.5 --raw --v 7\n```\n\nRules:\n\n- Unspecified weight defaults to 1.\n- Negative weights suppress concepts.\n- The sum of all weights must be greater than zero.\n- Use weights when a secondary concept "},{"path":"skill-card.md","content":"## Description:\n\nGenerate production-ready image prompts for Midjourney and other image models.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[clarezoe](https://clawhub.ai/user/clarezoe)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and creators use this skill to create, refine, critique, translate, and vary image-generation prompts for Midjourney and other image models.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill defaults to Midjourney and English prompt text when the request is underspecified.\n\nMitigation: Specify the target image model and prompt language when those choices matter.\n\nRisk: Built-in examples and defaults may influence the generated aesthetic.\n\nMitigation: Provide brand constraints, prohibited content, and desired style direction for production prompts.\n\nRisk: Midjourney model defaults and parameter compatibility can change over time.\n\nMitigation: Check official Midjourney documentation before relying on time-sensitive default-version or feature-compatibility claims.\n\n## Reference(s):\n\n- [Midjourney Prompt Guide](references/midjourney-prompt-guide.md)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, guidance]\n\n**Output Format:** [Markdown prompt text with optional numbered variants and short change notes]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Defaults to Midjourney when no target model is specified; final prompt text defaults to English unless the user requests another language.]\n\n## Skill Version(s):\n\n1.0.2 (source: server release metadata and skill 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."},{"path":"agents/openai.yaml","content":"interface:\n  display_name: \"Image Prompt Engineer\"\n  short_description: \"Generate production-ready image prompts\"\n  default_prompt: \"Use $image-prompt-engineer to turn this concept into a Midjourney image prompt.\""}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari... Skill: Image Prompt Engineer Owner: clarezoe Summary: Generate production-ready image prompts for Midjourney and other image models. Use when the user asks to create, refine, translate, critique, or produce vari... Tags: generative-ai:1.0.2, image-prompts:1.0.2, latest:1.0.2, midjourney:1.0.2, prompt-engineering:1.0.2 Version history: v1.0.2 | 2026-06-09T13:21:12.924Z | user Add OpenCode, Kiro, Devin, Codex, Cursor,","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1480,"uniquenessScore":46,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T11:21:09.768Z","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":"2026-10-11T11:21:09.768Z","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":"2026-10-11T14:13:42.156Z","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"}]}}}