{"id":"84f90a5d-59e6-4dec-bd6c-f4acc3e509d6","entityType":"agent","slug":"clawhub-xrowgmbh-xrowgmbh-image-generation","name":"Image Generation","canonicalUrl":"https://www.xpersona.co/agent/clawhub-xrowgmbh-xrowgmbh-image-generation","canonicalPath":"/agent/clawhub-xrowgmbh-xrowgmbh-image-generation","generatedAt":"2026-10-10T01:28:35.789Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T03:37:45.280Z","emptyReason":null},"description":"Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation. Skill: Image Generation Owner: xrowgmbh Summary: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation. Tags: latest:1.108.1 Version history: v1.108.1 | 2026-10-08T22:08:17.551Z | auto - Removed the file: skill-card.md - No changes to the core skill logic or documentation content. v1.108.0 | 2026-10-08T20:07:45.484Z | auto - Removed the skill car","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 6K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s17aekq8k9ea9rzxqpwjjz8ea987ze4c:xrowgmbh-image-generation","sourceUrl":"https://clawhub.ai/xrowgmbh/xrowgmbh-image-generation","homepage":"https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/xrowgmbh/xrowgmbh-image-generation","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":76,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation. Skill: Image Generation Owner: xr"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T03:37:45.280Z","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":"2026-10-09T03:37:45.280Z","emptyReason":null},"stars":null,"forks":null,"downloads":5962,"packageName":null,"latestVersion":"1.108.1","tractionLabel":"6K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T03:37:45.280Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T03:37:45.280Z","lastCrawledAt":"2026-10-09T03:37:45.280Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T03:37:45.280Z","lastVerifiedAt":null,"highlights":[{"version":"1.108.1","createdAt":"2026-10-08T22:08:17.551Z","changelog":"- Removed the file: skill-card.md - No changes to the core skill logic or documentation content.","fileCount":4,"zipByteSize":4525},{"version":"1.108.0","createdAt":"2026-10-08T20:07:45.484Z","changelog":"- Removed the skill card documentation file (skill-card.md). - No changes to behavior, rules, or workflow in the main skill documentation. - This release is documentation-only with file cleanup.","fileCount":4,"zipByteSize":4693},{"version":"1.107.1","createdAt":"2026-10-08T18:09:20.102Z","changelog":"- Removed the file: skill-card.md - No other changes were made in this version.","fileCount":4,"zipByteSize":4699},{"version":"1.107.0","createdAt":"2026-10-08T17:05:16.384Z","changelog":"- Removed the skill-card.md file. - No user-facing changes to skill behavior or documentation.","fileCount":4,"zipByteSize":4645},{"version":"1.106.1","createdAt":"2026-10-08T10:18:12.688Z","changelog":"- Removed the skill-card.md file. - No other functional or documentation changes in this version.","fileCount":4,"zipByteSize":4630},{"version":"1.106.0","createdAt":"2026-10-08T08:39:23.243Z","changelog":"- Removed the skill documentation file skill-card.md. - No changes to skill behavior or implementation. - Documentation and workflow in SKILL.md remain unchanged.","fileCount":4,"zipByteSize":4567},{"version":"1.105.1","createdAt":"2026-10-06T18:34:15.728Z","changelog":"- Removed the `skill-card.md` file from the repository. - No changes to the main skill workflow, rules, or documentation.","fileCount":4,"zipByteSize":4556},{"version":"1.105.0","createdAt":"2026-10-06T16:30:00.585Z","changelog":"- Removed the file skill-card.md. - No functional or workflow changes to the skill itself. - Documentation and rules in SKILL.md remain unchanged.","fileCount":4,"zipByteSize":4592}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17aekq8k9ea9rzxqpwjjz8ea987ze4c:xrowgmbh-image-generation","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","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-xrowgmbh-xrowgmbh-image-generation/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T01:28:35.787Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xrowgmbh-xrowgmbh-image-generation/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":"2026-10-09T03:37:45.280Z","emptyReason":null},"readme":"Skill: Image Generation\n\nOwner: xrowgmbh\n\nSummary: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nTags: latest:1.108.1\n\nVersion history:\n\nv1.108.1 | 2026-10-08T22:08:17.551Z | auto\n\n- Removed the file: skill-card.md\n- No changes to the core skill logic or documentation content.\n\nv1.108.0 | 2026-10-08T20:07:45.484Z | auto\n\n- Removed the skill card documentation file (skill-card.md).\n- No changes to behavior, rules, or workflow in the main skill documentation.\n- This release is documentation-only with file cleanup.\n\nv1.107.1 | 2026-10-08T18:09:20.102Z | auto\n\n- Removed the file: skill-card.md\n- No other changes were made in this version.\n\nv1.107.0 | 2026-10-08T17:05:16.384Z | auto\n\n- Removed the skill-card.md file.\n- No user-facing changes to skill behavior or documentation.\n\nv1.106.1 | 2026-10-08T10:18:12.688Z | auto\n\n- Removed the skill-card.md file.\n- No other functional or documentation changes in this version.\n\nv1.106.0 | 2026-10-08T08:39:23.243Z | auto\n\n- Removed the skill documentation file skill-card.md.\n- No changes to skill behavior or implementation.\n- Documentation and workflow in SKILL.md remain unchanged.\n\nv1.105.1 | 2026-10-06T18:34:15.728Z | auto\n\n- Removed the `skill-card.md` file from the repository.\n- No changes to the main skill workflow, rules, or documentation.\n\nv1.105.0 | 2026-10-06T16:30:00.585Z | auto\n\n- Removed the file skill-card.md.\n- No functional or workflow changes to the skill itself.\n- Documentation and rules in SKILL.md remain unchanged.\n\nv1.104.0 | 2026-10-06T15:27:57.565Z | auto\n\n- Documentation file skill-card.md was removed.\n- No functional skill changes; primary effect is on project documentation footprint.\n\nv1.103.0 | 2026-10-06T14:54:54.449Z | auto\n\n- Removed the skill-card.md file.\n- No changes were made to functionality or documentation of the skill itself.\n\nv1.102.0 | 2026-10-06T13:17:05.287Z | auto\n\n- Removed the documentation file skill-card.md.\n- No changes to core functionality or skill behavior.\n- Documentation and workflow guidance remain in SKILL.md.\n\nv1.101.0 | 2026-10-06T08:26:52.200Z | auto\n\n- Removed the skill-card.md file.\n- No changes were made to the core skill description or documentation.\n- Skill workflow and conventions remain unchanged.\n\nv1.100.0 | 2026-10-05T21:33:33.034Z | auto\n\n- Removed the file: skill-card.md\n- No other changes to workflow, conventions, or validation documented.\n\nv1.99.1 | 2026-10-05T20:47:19.357Z | auto\n\n- Removed the skill card file (skill-card.md) from the repository.\n- No changes to functionality, workflow, or rules; only documentation change.\n\nv1.99.0 | 2026-10-05T17:51:20.493Z | auto\n\n- Removed the file skill-card.md from the repository.\n- No changes made to the core functionality or rules of the skill.\n- Documentation and usage guidelines in SKILL.md remain unchanged.\n\nv1.98.1 | 2026-10-05T14:46:54.025Z | auto\n\n- Removed the file skill-card.md.\n- No other changes to the skill logic or instructions.\n\nv1.98.0 | 2026-10-05T11:10:43.671Z | auto\n\n- Removed the file: skill-card.md.\n- No changes to skill logic or workflow; documentation otherwise unchanged.\n\nv1.97.1 | 2026-10-05T01:03:46.319Z | auto\n\n- Removed the skill-card.md file from the repository.\n- No functional or behavioral changes to the image-generation skill itself.\n- Documentation, workflow, rules, and validation remain unchanged.\n\nv1.97.0 | 2026-10-03T11:54:16.947Z | auto\n\n- Removed the skill-card.md file.\n- No changes to core functionality or workflow. \n- The SKILL.md documentation remains unchanged.\n\nv1.96.0 | 2026-10-01T15:07:14.991Z | auto\n\n- Removed the file: skill-card.md.\n- No other functional or documentation changes detected.\n\nv1.95.1 | 2026-10-01T14:01:01.287Z | auto\n\n- Removed the file: skill-card.md\n- No changes to behavior or workflow; documentation and conventions remain as previously described.\n\nv1.95.0 | 2026-09-30T18:10:38.653Z | auto\n\n- Removed the skill-card.md file.\n- No other functional or documentation changes detected in this version.\n\nv1.94.0 | 2026-09-24T05:09:09.293Z | auto\n\n- Removed the file: skill-card.md\n- No changes to skill logic, usage, or documentation content.\n\nv1.93.1 | 2026-09-23T21:54:13.773Z | auto\n\n- Removed the file: skill-card.md\n- No changes to implementation or skill rules—this release only removes an unused documentation file.\n\nv1.93.0 | 2026-09-20T05:54:27.799Z | auto\n\n- Removed the skill-card.md file.\n- No other functional or documentation changes.\n\nv1.92.0 | 2026-09-19T19:16:28.549Z | auto\n\nVersion 1.92.0 of xrowgmbh-image-generation\n\n- No file changes detected in this release.\n- No updates to functionality or documentation.\n\nv1.91.8 | 2026-09-19T01:50:18.951Z | auto\n\nNo user-visible changes in this release.\n\n- Version bump with no detected file or documentation changes.\n\nv1.91.7 | 2026-09-18T18:25:21.742Z | auto\n\n- Removed the skill-card.md file.\n- No functional or behavioral changes to the skill itself.\n- Documentation and workflow remain unchanged.\n\nv1.91.6 | 2026-09-17T04:05:33.352Z | auto\n\nVersion 1.91.6 of xrowgmbh-image-generation\n\n- No file changes detected in this release.\n- Documentation, skill description, rules, and workflow remain unchanged.\n\nv1.91.5 | 2026-09-16T18:29:43.786Z | auto\n\n- Removed the skill-card.md file.\n- No changes to skill functionality or usage instructions.\n\nv1.91.4 | 2026-09-16T07:21:14.438Z | auto\n\n- Removed the file skill-card.md from the repository.\n- No changes were made to functionality, documentation, or workflow instructions.\n\nv1.91.3 | 2026-09-13T00:56:30.244Z | auto\n\n- Removed the skill-card.md file.\n- No changes to functionality or documentation of the image-generation skill itself.\n\nv1.91.2 | 2026-09-11T06:40:44.171Z | auto\n\n- Removed the skill-card.md file. \n- No other changes to skill behavior or documentation.\n\nv1.91.1 | 2026-09-09T16:56:35.271Z | auto\n\n- Removed the skill-card.md file.\n- No functional or workflow changes were made.\n- All documentation and main skill instructions remain unchanged.\n\nv1.91.0 | 2026-09-09T13:53:56.752Z | auto\n\n- Removed the file: skill-card.md\n- No other functional or documentation changes in this version.\n\nv1.90.0 | 2026-09-08T06:42:20.394Z | auto\n\n- Removed the sample file: skill-card.md.\n- No changes to core logic, rules, or workflow in SKILL.md.\n- Documentation remains unchanged except for file removal.\n\nv1.89.0 | 2026-09-08T06:23:47.777Z | auto\n\n- Removed the file: skill-card.md\n- No changes made to documentation, rules, workflow, or conventions in SKILL.md\n- No new features or breaking changes introduced in this version\n\nv1.88.0 | 2026-09-07T23:34:20.780Z | auto\n\n- Removed the skill card file (`skill-card.md`).\n- No changes were made to the main skill logic or documentation.\n\nv1.87.0 | 2026-09-07T17:19:11.321Z | auto\n\n- Removed the `skill-card.md` file from the repository.\n- No other changes to skill logic, documentation, or workflow.\n\nv1.86.0 | 2026-09-07T08:24:16.047Z | auto\n\n- Removed the file skill-card.md.\n- No other functional or documentation changes.\n\nv1.85.1 | 2026-09-07T06:29:44.503Z | auto\n\n- Removed the skill-card.md file.\n- No changes to the core skill logic or documentation.\n\nv1.85.0 | 2026-09-07T05:31:09.493Z | auto\n\n- Removed the skill-card.md file.\n- Skill documentation and rules remain unchanged.\n\nv1.84.7 | 2026-09-05T00:43:26.316Z | auto\n\n- Removed the file: skill-card.md.\n- No changes to features or workflow; documentation and rules remain unchanged.\n\nv1.84.6 | 2026-09-03T00:15:31.327Z | auto\n\n- Removed the file: skill-card.md.\n- No changes to skill behavior or documentation content.\n- This update is a minor cleanup to remove redundant or unnecessary files.\n\nv1.84.5 | 2026-09-02T00:33:40.142Z | auto\n\n- Removed the file `skill-card.md`.\n- No changes to skill behavior or documentation beyond file removal.\n\nv1.84.4 | 2026-09-01T10:42:49.920Z | auto\n\n- Removed the skill-card.md file.\n- No changes were made to skill behavior or documentation content.\n\nv1.84.3 | 2026-09-01T08:28:28.781Z | auto\n\n- Removed the file: skill-card.md\n- No other functional changes to the skill or documentation.\n\nv1.84.2 | 2026-08-25T00:38:33.857Z | auto\n\n- Removed the skill-card.md file.\n- No changes to core skill functionality or rules.\n- Documentation and workflow guidance remain unchanged.\n\nv1.84.1 | 2026-08-24T20:11:25.989Z | auto\n\n- Removed the skill-card.md file.\n- No changes to skill logic, rules, or workflow.  \n- Documentation and operational details remain unchanged.\n\nv1.84.0 | 2026-08-16T12:29:14.013Z | auto\n\n- Removed the file: skill-card.md.\n- No other changes to workflow, rules, or documentation.\n\nArchive index:\n\nArchive v1.108.1: 4 files, 4525 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1492b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.108.1:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.108.1:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.108.1\",\n  \"publishedAt\": 1791497297551\n}\n\nFile v1.108.1:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content teams use this skill to create or revise images for documentation, PDFs, web pages, and release materials while checking the requested format and final artifacts.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Generated images or rebuilt PDFs and release artifacts may not match the requested format or quality.\n\nMitigation: Review generated images and rebuilt documents before merging; check format, resolution, and how images appear in the final artifact.\n\n## Reference(s):\n\n\n## Skill Output:\n\n**Output Type(s):** [Image files, Guidance, Shell commands]\n\n**Output Format:** [SVG, PNG, WebP, and Markdown guidance]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Image briefs may be kept alongside generated files; validate format, resolution, and embedded output.]\n\n## Skill Version(s):\n\n1.108.1 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.108.0: 4 files, 4693 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1811b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.108.0:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.108.0:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.108.0\",\n  \"publishedAt\": 1791490065484\n}\n\nFile v1.108.0:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content creators use this skill to create, revise, and review images for documents, PDFs, web pages, and release artifacts while preserving requested formats and checking final outputs.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: A broad scan root can expose filenames and modification times from unrelated private directories.\n\nMitigation: Run image validation against the intended repository or workspace only.\n\nRisk: Generated image formats or dimensions may not match the requested deliverable or remain readable when embedded.\n\nMitigation: Preserve the requested format and inspect the rebuilt PDF, document, or web page at its final display size.\n\n## Reference(s):\n\n- [Image Generation on ClawHub](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Shell commands, Image files]\n\n**Output Format:** [Markdown guidance with validation commands; SVG, PNG, and WebP images when generated]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Check image format, dimensions, freshness, and readability in the final document or page.]\n\n## Skill Version(s):\n\n1.108.0 (source: ClawHub release 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 v1.107.1: 4 files, 4699 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1829b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.107.1:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.107.1:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.107.1\",\n  \"publishedAt\": 1791482960102\n}\n\nFile v1.107.1:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and reviewers use this skill to create and revise images for documentation, PDFs, web pages, and release artifacts while checking format, readability, and generated assets.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Generated image variants can drift from the requested format or become outdated after the image brief changes.\n\nMitigation: Keep variants alongside the image brief, preserve the requested format, and run the included checker before review.\n\nRisk: An image may appear unclear or be omitted in the final PDF or document despite looking correct in its source file.\n\nMitigation: Build downstream documents and inspect the embedded image at its final display size.\n\n## Reference(s):\n\n- [Image Generation on ClawHub](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Image files, Shell commands]\n\n**Output Format:** [Markdown guidance and generated SVG, PNG, or WebP image assets]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Image dimensions and format follow the request; the local checker reports missing or outdated image variants.]\n\n## Skill Version(s):\n\n1.107.1 (source: ClawHub release 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 v1.107.0: 4 files, 4645 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1735b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.107.0:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.107.0:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.107.0\",\n  \"publishedAt\": 1791479116384\n}\n\nFile v1.107.0:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content creators use this skill to produce or revise images for documents, web pages, and reviews while checking the requested format and final artifact quality.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Image work may change assets and downstream documents beyond the intended deliverable.\n\nMitigation: Confirm the requested format and affected documents before making changes, then review rebuilt outputs.\n\nRisk: The validator may enumerate image-prompt files throughout the supplied workspace root.\n\nMitigation: Choose the intended workspace root and review reported files before regenerating images.\n\n## Reference(s):\n\n- [Image Generation on ClawHub](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Code, Shell commands, Image files]\n\n**Output Format:** [Markdown guidance and generated SVG, PNG, or WebP images]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Checks image formats, resolution, and freshness against the requested deliverable.]\n\n## Skill Version(s):\n\n1.107.0 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.106.1: 4 files, 4630 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1728b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.106.1:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.106.1:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.106.1\",\n  \"publishedAt\": 1791454692688\n}\n\nFile v1.106.1:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content creators use this skill to create or revise images for documents, PDFs, web pages, and release artifacts while preserving requested formats and checking final outputs.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Generated images or rebuilt documents may change beyond the requested edit.\n\nMitigation: Review image, document, and release-artifact diffs before accepting changes.\n\nRisk: Validation may scan the wrong repository or miss the intended generated images.\n\nMitigation: Run the image checker against the intended repository and inspect its reported missing or stale images.\n\n## Reference(s):\n\n- [Image Generation on ClawHub](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Images, Image-generation prompts, Guidance, Shell commands]\n\n**Output Format:** [PNG, SVG, WebP, and Markdown]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Generated image siblings are checked against their prompt files; requested output formats take priority.]\n\n## Skill Version(s):\n\n1.106.1 (source: ClawHub release 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 v1.106.0: 4 files, 4567 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1556b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.106.0:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.106.0:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.106.0\",\n  \"publishedAt\": 1791448763243\n}\n\nFile v1.106.0:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and documentation teams use this skill to create or revise images for documents, web pages, and reviews while preserving requested formats and checking generated artifacts.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Creating or rebuilding images and documents can change deliverables unexpectedly.\n\nMitigation: Review changed files and final rendered artifacts before committing.\n\n## Reference(s):\n\n- [Image Generation skill release](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n- [xrowgmbh publisher profile](https://clawhub.ai/user/xrowgmbh)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Image files, Shell commands]\n\n**Output Format:** [Markdown guidance and SVG, PNG, or WebP image files]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Can include image briefs and checks for missing or outdated image variants.]\n\n## Skill Version(s):\n\n1.106.0 (source: ClawHub release 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 v1.105.1: 4 files, 4556 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1514b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.105.1:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.105.1:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.105.1\",\n  \"publishedAt\": 1791311655728\n}\n\nFile v1.105.1:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content creators use this skill to create and review images for documents, websites, and releases while preserving requested formats and checking final artifacts.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Image edits may create or rebuild sibling files and documents that embed them.\n\nMitigation: Review generated files and rebuilt documents before committing changes.\n\n## Reference(s):\n\n- [Image Generation on ClawHub](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Shell commands, Image files]\n\n**Output Format:** [Markdown guidance and SVG, PNG, or WebP image files]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Image prompts may be kept in adjacent .image.genai files; final artifacts require visual and format checks.]\n\n## Skill Version(s):\n\n1.105.1 (source: ClawHub release 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 v1.105.0: 4 files, 4592 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1600b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.105.0:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.105.0:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.105.0\",\n  \"publishedAt\": 1791304200585\n}\n\nFile v1.105.0:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content creators use this skill to create or revise images for documents, websites, and reviews, then check that the delivered assets and embedded results match the requested format and quality.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Using the skill may change generated image files and documents in the selected workspace.\n\nMitigation: Point it at the intended workspace and review the resulting files and rebuilt documents before release.\n\n## Reference(s):\n\n- [ClawHub skill listing](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Image files, Guidance, Shell commands]\n\n**Output Format:** [SVG, PNG, and WebP assets with Markdown guidance and validation output]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Generated image formats may be narrowed by the task; images should be checked in their final destination.]\n\n## Skill Version(s):\n\n1.105.0 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.104.0: 4 files, 4653 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1776b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.104.0:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.104.0:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.104.0\",\n  \"publishedAt\": 1791300477565\n}\n\nFile v1.104.0:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content creators use this skill to create and review image assets for documentation, PDFs, websites, and releases while preserving requested formats and checking final outputs.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Regenerating images can change the requested format or leave embedded documents with outdated assets.\n\nMitigation: Confirm the requested format, inspect the resulting images, and rebuild and review downstream documents.\n\nRisk: Checking the wrong repository root can produce misleading image-validation results.\n\nMitigation: Run the local checker against the intended workspace and review reported missing or stale assets.\n\n## Reference(s):\n\n- [Image Generation on ClawHub](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Image files, Validation results]\n\n**Output Format:** [Markdown guidance and image assets (SVG, PNG, WebP)]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Image formats follow the request; generated assets are checked for missing or stale siblings.]\n\n## Skill Version(s):\n\n1.104.0 (source: ClawHub release 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 v1.103.0: 4 files, 4620 bytes\n\nFiles: scripts/check-image-genai.py (2384b), skill-card.md (1698b), SKILL.md (5011b), _meta.json (146b)\n\nFile v1.103.0:SKILL.md\n\n---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constraints, text that must appear in the image, and downstream consumers in the prompt file.\n- Generate all sibling formats from the same brief: `docs/architecture.svg`, `docs/architecture.png`, and `docs/architecture.webp`.\n- Prefer the SVG sibling for diagrams and scalable documentation, the PNG sibling for PDFs or places that need stable raster rendering, and the WebP sibling for web delivery.\n- Regenerate all siblings together after prompt changes so the formats do not drift.\n- Before final review, run `{baseDir}/scripts/check-image-genai.py --root <workspace>` to find prompt files without generated image siblings and generated images that are older than their prompt.\n- Mention the prompt file and the generated sibling dimensions in MR notes.\n\n## Validation Checklist\n\n- Image files have the requested extension and MIME format.\n- Every changed `*.image.genai` file has matching `.svg`, `.png`, and `.webp` siblings unless the MR explains why a format is intentionally omitted.\n- The generated image checker passes; add `All Images generated` to the final Definition of Done when it does.\n- PNG files are generated at source size, not post-upscaled.\n- PDF-bound PNG files have sufficient dimensions for their displayed size and about 300 DPI metadata when print quality is requested.\n- Text and fine lines are readable after the image is embedded in the final PDF or web page.\n- The PDF or DOCX contains the intended image objects and does not silently drop unsupported formats.\n- CI checks enforce the important invariants when the repository already has artifact validation.\n- MR notes state the exact format, dimensions, validation commands, and any remaining limitation.\n\n## GitLab Review Handling\n\nFor image-related MR feedback, answer the latest reviewer request first. If new feedback contradicts an earlier fix, stop extending the previous approach and explicitly revert or narrow it to the latest requested format. Resolve discussions only after the final artifact has been rebuilt and checked.\n\nFile v1.103.0:_meta.json\n\n{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.103.0\",\n  \"publishedAt\": 1791298494449\n}\n\nFile v1.103.0:skill-card.md\n\n## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content teams use this skill to create or revise images for documents and web pages, then check the requested format and quality in the finished artifacts.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Generated images may have the wrong format, insufficient resolution, or unreadable detail in the final document.\n\nMitigation: Check the requested format, dimensions, and legibility in the rebuilt document before accepting the result.\n\nRisk: Regenerating images may change multiple sibling formats and downstream documents.\n\nMitigation: Review generated images and affected documents before committing changes.\n\n## Reference(s):\n\n- [Image Generation skill release](https://clawhub.ai/xrowgmbh/skills/xrowgmbh-image-generation)\n\n## Skill Output:\n\n**Output Type(s):** [Images, Files, Guidance]\n\n**Output Format:** [PNG, SVG, and WebP image files; Markdown guidance]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May update documents and release artifacts that embed the images.]\n\n## Skill Version(s):\n\n1.103.0 (source: ClawHub release 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.","readmeExcerpt":"Skill: Image Generation Owner: xrowgmbh Summary: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation. Tags: latest:1.108.1 Version history: v1.108.1 | 2026-10-08T22:08:17.551Z | auto - Removed the file: skill-card.md - No changes to the core skill logic or documentation content. v1.108.0 | 2026-10-08T20:07:45.484Z | auto - Removed the skill car","codeSnippets":[],"executableExamples":[],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: image-generation\ndescription: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n---\n\n# Image Generation Skill\n\nUse this skill when creating, regenerating, or reviewing images, diagrams, screenshots, or generated graphics for Markdown, PDF, DOCX, web pages, merge requests, or release artifacts.\n\n## What Went Wrong In claw-support !1\n\nThe image review loop failed because the work drifted between SVG and PNG, optimized technical metadata before preserving the requested format, and claimed image quality based on checks that did not match the maintainer's desired output. Treat this as the default failure mode to prevent.\n\n## Rules\n\n- Preserve the requested output format. If the reviewer asks for PNG, do not switch to SVG unless they explicitly accept that change.\n- Preserve the requested style or earlier format when asked. Do not redesign a diagram as a workaround for readability feedback.\n- Render raster images directly at the final source resolution. Never upscale a smaller bitmap and call it high resolution.\n- For PDF-bound PNGs, use at least 300 DPI metadata and enough pixels for the printed/displayed size.\n- Scale text, line widths, arrows, borders, spacing, and icons together. Large canvases with tiny text still fail review.\n- Prefer vector-like drawing primitives rendered into the final PNG canvas for diagrams. Avoid screenshots or nested raster snippets unless the screenshot is the actual subject.\n- Keep source-of-truth files clear: if the deliverable is PNG, make the build pipeline consume PNGs directly and avoid hidden SVG dependencies unless SVG is explicitly the editable source.\n- Use `*.image.genai` prompt files as the source-of-intent convention for generated repository images. For `example.image.genai`, generate sibling `example.svg`, `example.png`, and `example.webp` files unless the task explicitly narrows the output formats.\n- Do not add broad image-generation rules to repository `AGENTS.md` unless asked. Prefer a reusable skill or a task-local note.\n\n## Workflow\n\n1. Read the requested artifact path, target consumer, and exact format from the issue or MR discussion.\n2. Inspect existing artifacts before editing: dimensions, format, DPI metadata, source generator, and where each image is embedded.\n3. If a `*.image.genai` file exists or a new generated image is needed, treat that file as the editable prompt/brief and create the matching `.svg`, `.png`, and `.webp` siblings next to it.\n4. Make the smallest change that satisfies the reviewer: format, resolution, clarity, or embedding.\n5. Rebuild every downstream artifact that embeds the image, such as Markdown previews, PDFs, DOCX files, and release bundles.\n6. Validate the actual output, not only the source file.\n\n## `*.image.genai` Convention\n\n- Keep the file next to the intended generated assets, for example `docs/architecture.image.genai`.\n- Store the image brief, required dimensions, style constrai"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn7frykp2tnj00b0czk5f4r505809an4\",\n  \"slug\": \"xrowgmbh-image-generation\",\n  \"version\": \"1.108.1\",\n  \"publishedAt\": 1791497297551\n}"},{"path":"skill-card.md","content":"## Description:\n\nCreate or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xrowgmbh](https://clawhub.ai/user/xrowgmbh)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and content teams use this skill to create or revise images for documentation, PDFs, web pages, and release materials while checking the requested format and final artifacts.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Generated images or rebuilt PDFs and release artifacts may not match the requested format or quality.\n\nMitigation: Review generated images and rebuilt documents before merging; check format, resolution, and how images appear in the final artifact.\n\n## Reference(s):\n\n\n## Skill Output:\n\n**Output Type(s):** [Image files, Guidance, Shell commands]\n\n**Output Format:** [SVG, PNG, WebP, and Markdown guidance]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Image briefs may be kept alongside generated files; validate format, resolution, and embedded output.]\n\n## Skill Version(s):\n\n1.108.1 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation. Skill: Image Generation Owner: xrowgmbh Summary: Create or revise document, PDF, web, or review images with the requested format, sharp raster output, and artifact validation. Tags: latest:1.108.1 Version history: v1.108.1 | 2026-10-08T22:08:17.551Z | auto - Removed the file: skill-card.md - No changes to the core skill logic or documentation content. v1.108.0 | 2026-10-08T20:07:45.484Z | auto - Removed the skill car","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1205,"uniquenessScore":47,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-09T03:37:45.280Z","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":"2026-10-09T03:37:45.280Z","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":"2026-10-10T01:28:35.789Z","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"}]}}}