[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M718VB3T401AH24WJA443DB0":3,"skill-same:01M718VB3T401AH24WJA443DB0":93},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":6,"ownerHtmlUrl":54,"ownerBlogUrl":55,"ownerDescription":6,"updatedAt":56,"install":57,"resources":70},"01M718VB3T401AH24WJA443DB0","oma-image",null,"Multi-vendor AI image generation with authentication-aware parallel dispatch. Routes to Codex (gpt-image-2 via ChatGPT OAuth), Antigravity (Gemini-family \"nano-banana\" image models via `agy` CLI + Gemini Code Assist; exact model chosen internally by agy), and Pollinations (flux\u002Fzimage, free with signup). Use for image generation, image creation, visual asset generation, and AI art.","vasceannie\u002Fslopgate","vasceannie","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F20271133?v=4","01M718RGA8VV2EF22XJEMXVQ0Q","Travis Vasceannie","01M718RGB0GWSXBBZBN67Q6NHQ",1,"MIT","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"oma image","image","for","gemini","parallel","asset generation","asset","flux","chatgpt","signup","ai image generation","ai image","banana","cli","code","free","Art","authentication","model","creation","2026-11-10 10:12:30",false,"---\nname: oma-image\ndescription: Multi-vendor AI image generation with authentication-aware parallel\n  dispatch. Routes to Codex (gpt-image-2 via ChatGPT OAuth), Antigravity\n  (Gemini-family \"nano-banana\" image models via `agy` CLI + Gemini Code Assist;\n  exact model chosen internally by agy), and Pollinations (flux\u002Fzimage, free\n  with signup). Use for image generation, image creation, visual asset\n  generation, and AI art.\n---\n\n# Image Agent - Multi-Vendor Image Router\n\n## Scheduling\n\n### Goal\nGenerate images and visual assets through authenticated multi-vendor routing while preserving prompt clarity, reference-guided regeneration, cost controls, and reproducible output manifests.\n\n### Intent signature\n- User asks to generate images, visual assets, illustrations, product photos, concept art, mockups, or AI art.\n- Another skill needs shared image-generation infrastructure.\n- User provides reference images for a new generated variation or asks for vendor comparison.\n\n### When to use\n\n- Generating images, visual assets, illustrations, product photos, concept art\n- Comparing output between multiple image models for the same prompt\n- Producing images from prompts within editor workflows (Claude Code, Codex, Gemini CLI)\n- Other skills needing image generation infrastructure (shared invocation)\n- Regenerating a new image that follows a supplied reference's subject, style, lighting, or composition\n\n### When NOT to use\n\n- Pixel-level editing, masking, inpainting, object removal, compositing, cropping, resizing, or format conversion -> out of scope; use an image editor or a tool that explicitly supports that operation.\n- Generating videos or audio -> out of scope\n- Inline vector art \u002F SVG composition from structured data -> use a templating skill\n- Simple asset resizing or format conversion -> use a dedicated image library\n\n### Expected inputs\n- Image prompt or creative brief\n- Optional vendor, size, quality, count, output directory, and reference images\n- Authentication\u002Fenvironment state for Codex, Pollinations, or Gemini\n\n### Expected outputs\n- Generated image files under `.agents\u002Fresults\u002Fimages\u002F` or requested output directory\n- `manifest.json` with prompt, vendor, model, and reproducibility metadata\n- Vendor comparison outputs when `--vendor all` is used\n\n### Dependencies\n- `oma image generate` CLI and vendor authentication\n- Codex image generation, Pollinations API, or Gemini API\u002FCLI strategy\n- `resources\u002Fvendor-matrix.md`, `resources\u002Fprompt-tips.md`, and the `image:` section of `.agents\u002Foma-config.yaml`\n\n### Control-flow features\n- Branches by prompt ambiguity, vendor auth, cost threshold, reference-guided regeneration support, path safety, and safety\u002Ftimeout exit codes\n- Calls external vendor APIs\u002FCLIs\n- Reads reference images and writes generated images plus manifests\n\n## Structural Flow\n\n### Entry\n1. Validate that the request contains enough subject, setting, style, usage, and aspect-ratio signal.\n2. Classify a supplied image as a reference for a newly generated variation, or as a request for unsupported pixel\u002Fmask editing.\n3. Check authentication, cost guardrails, output path, and count limits.\n\n### Scenes\n1. **PREPARE**: Clarify or amplify prompt and choose vendor strategy.\n2. **ACQUIRE**: Validate auth, references, output path, and provider availability.\n3. **ACT**: Invoke `oma image generate` with selected vendor(s), prompt, references, and options.\n4. **VERIFY**: Check manifest, output files, exit code, and provider result.\n5. **FINALIZE**: Return output paths and relevant warnings.\n\n### Transitions\n- If prompt lacks required signal, clarify or show amplified prompt before generation.\n- If `--vendor all` is requested, require every requested vendor to be available.\n- If the request is reference-guided regeneration and the selected vendor supports references, pass it automatically.\n- If the request requires pixel, mask, crop, resize, or format editing, stop before generation and name the unsupported operation and an appropriate editor route.\n- If estimated cost exceeds guardrail, require confirmation unless bypassed.\n\n### Failure and recovery\n- If auth is missing, report vendor-specific authentication requirement.\n- If reference support is unavailable for the selected vendor, reject with actionable guidance.\n- If local CLI is outdated, ask user to run `oma update`.\n- If generation times out or is blocked, surface exit code and provider status.\n\n### Exit\n- Success: images and manifest exist in the output directory.\n- Partial success: some vendors fail in comparison mode and failures are reported.\n- Failure: no image is produced and the route\u002Fcost\u002Fauth\u002Fsafety blocker is explicit.\n\n## Logical Operations\n\n### Actions\n| Action | SSL primitive | Evidence |\n|--------|---------------|----------|\n| Validate prompt completeness | `VALIDATE` | Clarification protocol |\n| Select vendor strategy | `SELECT` | Vendor matrix and auth state |\n| Read reference images | `READ` | `--reference` paths |\n| Call generation CLI\u002FAPI | `CALL_TOOL` | `oma image generate` |\n| Write image outputs | `WRITE` | Image files and manifest |\n| Validate result | `VALIDATE` | Exit code, manifest, files |\n| Report output | `NOTIFY` | Final path summary |\n\n### Tools and instruments\n- `oma image generate`, `oma image doctor`, `oma image vendor list`\n- Codex, Pollinations, and Gemini provider paths\n- Prompt tips, vendor matrix, and image config\n\n### Canonical command path\n```bash\noma image doctor\noma image generate \"\u003Cprompt>\" --vendor auto --size auto --quality auto --output json\n```\n\nWith reference images:\n```bash\noma image generate --reference \"\u003Cabsolute-path>\" --vendor codex \"\u003Cprompt>\"\n```\n\n### Resource scope\n| Scope | Resource target |\n|-------|-----------------|\n| `LOCAL_FS` | Reference images, generated images, manifests |\n| `PROCESS` | Provider CLIs and image router commands |\n| `NETWORK` | Pollinations\u002FGemini or provider APIs |\n| `CREDENTIALS` | Provider auth and API keys |\n\n### Preconditions\n- Prompt is sufficiently specified or user approves amplification.\n- Required vendor auth and output permissions exist.\n- Reference paths are accessible when used.\n\n### Effects and side effects\n- Creates image files and manifests.\n- May call paid or rate-limited provider APIs.\n- May read attached\u002Freference images.\n\n### Guardrails\n\n1. **Clarify before invoking**: if the user's request is ambiguous about subject, style, composition, or usage context, **ask the user first** or **amplify the prompt explicitly** (showing the user the expanded version for approval). Do NOT silently generate from a vague prompt. See `Clarification Protocol` below.\n2. **Authentication-aware dispatch**: detect which vendor CLIs are available and run only those; with `--vendor all`, every requested vendor must be available (strict). Caveat: the `antigravity` health check verifies installation only (`agy --version`) — a signed-out agy passes health and fails at generate time with agy's own error.\n3. **Cost guardrail**: confirm before executing runs whose estimated cost is ≥ `$0.20` (configurable). `--yes` \u002F `OMA_IMAGE_YES=1` bypass. Default vendors `pollinations` (flux\u002Fzimage) and `antigravity` (nano-banana via Gemini Code Assist) are free, so auto-triggering on keywords is safe. **Non-interactive contexts** (agents, CI — no TTY on stdin): the CLI cannot prompt, so a run at\u002Fover the threshold exits 1 with a message naming `--yes`. Calling agents must confirm the estimated cost with the user in-conversation (use `--dry-run` to get the estimate), then re-run with `-y`.\n4. **Path safety**: output paths outside `$PWD` require `--allow-external-out`.\n5. **Cancellable**: SIGINT\u002FSIGTERM aborts in-flight provider calls and the orchestrator.\n6. **Deterministic outputs**: every run writes `manifest.json` next to the images for reproducibility.\n7. **Max `n` = 5**: wall-time bound.\n8. **Exit codes align with `oma search fetch`** (0, 1, 2=safety, 3=not-found, 4=invalid-input, 5=auth-required, 6=timeout).\n\n### Clarification Protocol\n\nBefore invoking `oma image generate`, the calling agent runs this checklist against the user's request. **If any answer is \"no \u002F unknown\", clarify with the user first.**\n\n**Required signal (must be present or inferable):**\n- [ ] **Subject**: what is the primary thing in the image? (object, person, scene)\n- [ ] **Setting \u002F backdrop**: where is it? (context, environment)\n\n**Strongly recommended (ask if absent AND not inferable from context):**\n- [ ] **Style**: photorealistic, illustration, 3D render, oil painting, concept art, flat vector, …?\n- [ ] **Mood \u002F lighting**: bright vs moody, warm vs cool, dramatic vs minimal\n- [ ] **Usage context**: hero image, icon, thumbnail, product shot, poster? (dictates aspect ratio + composition)\n- [ ] **Aspect ratio \u002F resolution**: any `WxH` where each edge is a multiple of 16 between 16 and 3840 and aspect ∈ [1:3, 3:1] (e.g. `1024x1024` square, `2048x1152` 16:9, `3840x2160` 4K UHD, `1024x1536` portrait), or `auto`.\n\n**Amplification shortcut.** For brief prompts (e.g. \"a red apple\"), do not pop clarifying questions if the request is genuinely that simple. Instead **amplify inline and show the user** the expanded version before invoking:\n\n> User: \"a red apple\"\n> Agent: \"I'll generate this as: *a single glossy red apple centered on a clean white background, soft studio lighting, photorealistic, shallow depth of field, 1024×1024*. Shall I proceed, or would you like a different style\u002Fcomposition?\"\n\nSkip both clarification and amplification when the user has clearly authored a full creative brief (≥ 2 of: subject + style + lighting + composition). Respect their prompt verbatim.\n\n**Category-specific briefs** (app mockup, poster, thumbnail, infographic, comic panel, avatar): consult `resources\u002Fprompt-tips.md` → *External Prompt Libraries*.\n\n**Output language.** Generation prompts are sent to the provider in English (image models are trained predominantly on English captions). Translate the user's request if they wrote in another language, and show them the translated version during amplification so they can correct misreadings.\n\n### Vendors\n\nThis skill follows oh-my-agent's CLI-first concept: whenever a vendor's native CLI can drive generation (and return raw bytes), the subprocess path is preferred over direct API keys. Direct API is only used as a fallback for vendors whose CLI can't yet emit raw image bytes.\n\n| Vendor | Strategy | Models | Trigger |\n|--------|----------|--------|---------|\n| `codex` | CLI-first via `codex exec` over ChatGPT OAuth (`codex login`), built-in `image_gen` | `gpt-image-2` | Logged in via Codex CLI (no API key) |\n| `pollinations` | Direct HTTP via `gen.pollinations.ai\u002Fv1\u002Fimages\u002Fgenerations` (free signup for key) | Free: `flux`, `zimage`. Credit-gated: `qwen-image`, `wan-image`, `gpt-image-2`, `klein`, `kontext`, `gptimage`, `gptimage-large` | `POLLINATIONS_API_KEY` set (free at https:\u002F\u002Fenter.pollinations.ai). No native CLI exists. |\n| `antigravity` | `agy -p --dangerously-skip-permissions --add-dir \u003CoutDir>` — Antigravity's agentic CLI runs over the user's Gemini Code Assist subscription. agy writes raw bytes to absolute target paths we embed in the prompt; the provider sniffs format via magic bytes and renames the file extension to match. Model selection is opaque — agy picks internally, we never name a model. | (opaque — chosen by agy) | `agy` CLI installed + signed in. No API key, no per-image charge. |\n\n> The direct Gemini path (`gemini -p` stream, `generativelanguage.googleapis.com` API) is deprecated. `agy` is the supported Gemini image route — it's free with Gemini Code Assist and doesn't require billing on AI Studio.\n\n### Invocation\n\n#### Standalone\n\n```\n\u002Foma-image a red apple on white background\n\u002Foma-image --vendor all --size 1536x1024 jeju coastline at sunset\n\u002Foma-image -n 3 --quality high --out .\u002Fhero \"minimalist dashboard hero illustration\"\n```\n\n#### Shell CLI\n\n```\noma image generate \"\u003Cprompt>\" [--vendor auto|codex|pollinations|antigravity|all] [-n 1..5] \\\n                             [--size WxH|auto] \\\n                             [--quality low|medium|high|auto] \\\n                             [--model \u003Cname>] \\\n                             [--output-dir \u003Cdir>] [--allow-external-output] \\\n                             [-r \u003Cpath>]... \\\n                             [--timeout 180] [-y] [--no-prompt-in-manifest] \\\n                             [--dry-run] [--output text|json]\noma image doctor\noma image vendor list\n```\n\n`--model \u003Cname>` overrides the vendor's default model for this run — e.g. `--vendor pollinations --model zimage`, or a credit-gated Pollinations model like `gpt-image-2`. It applies to every vendor in the run set, so combine it with an explicit `--vendor`; `antigravity` ignores it (agy picks its model internally).\n\n#### Reference-guided regeneration (`-r`, `--reference`)\n\nAttach up to 10 reference images (PNG\u002FJPEG\u002FGIF\u002FWebP, ≤ 5MB each) to guide a **newly generated** image's style, subject identity, lighting, or composition. Repeatable or comma-separated. The output is a new generation, not a pixel-preserving edit of the input.\n\n```\noma image generate -r ~\u002FDownloads\u002Fotter.jpeg \"same otter in dramatic lighting\"\noma image generate -r a.png -r b.png \"blend these two styles\"\n```\n\nSupported vendors:\n\n| Vendor | Support | How |\n|--------|---------|-----|\n| `codex` (gpt-image-2) | PASS | Passes `-i \u003Cpath>` to `codex exec` |\n| `antigravity` | PASS | Refs copied to a per-run temp dir, `agy --add-dir \u003Ctmpdir>` grants access, paths inlined into the prompt |\n| `pollinations` | N\u002FA | Rejected with exit code 4 when explicitly selected (requires URL hosting; see PR #2 roadmap). Under `--vendor auto`, reference-unsupported vendors are silently dropped from the run set instead. |\n\n**Paths**: absolute or relative to `$CWD`. Host CLIs usually expose attached images via:\n- **Claude Code**: `~\u002F.claude\u002Fimage-cache\u002F\u003Csession>\u002FN.png` (surfaced in system messages as `[Image: source: \u003Cpath>]`)\n- **Antigravity**: workspace upload directory (exact path shown in IDE)\n- **Codex CLI as host**: user must pass the filesystem path explicitly; in-conversation attachments are not forwarded\n\n#### Agent Behavior: Auto-forward Attached References (MANDATORY)\n\nWhen ALL of the following are true, the calling agent MUST pass the attached image via `--reference \u003Cpath>` automatically. Never describe the image in prose as a workaround.\n\n1. The user asks for a new generated variation (for example, \"same subject in dramatic lighting\", \"make a new illustration in this style\", or \"regenerate this at a different angle\").\n2. A host-surfaced attached image is visible to the agent (e.g. a Claude Code system message with `[Image: source: \u003Cpath>]`, or an Antigravity workspace upload path, or an explicit filesystem path in the user's message).\n3. The selected vendor supports references (`codex` or `antigravity`).\n\n**Required action**: invoke `oma image generate --reference \u003Cabsolute-path> --vendor \u003Ccodex|antigravity> \"\u003Cprompt>\"`. If the user didn't specify a vendor, default to `codex` (CLI-first, widest availability). Do NOT:\n\n- Fall back to prose description (\"I'll describe the otter's appearance...\").\n- Ask the user to re-type or re-attach the path.\n- Claim the CLI doesn't support references without first running `oma image generate --help` to verify.\n\n**If the local CLI is outdated** (`--reference` is missing from `--help`): tell the user to run `oma update` once, then retry. Do not silently degrade to prose.\n\n**If the reference path is from Claude Code's `image-cache`**: note to the user that the path is session-scoped and suggest copying the file to a durable location if they want to reuse it later. Still proceed with the generation.\n\n**Unsupported edit requests**: Do not route requests such as \"remove this object\", \"change only the background\", \"paint inside this mask\", \"keep every pixel except\", \"crop\", \"resize\", or \"convert this file\" through `--reference`. They need pixel\u002Fmask editing or deterministic image processing, neither of which this CLI provides. Say whether the request can instead be phrased as a reference-guided regeneration; otherwise route to an appropriate image editor or processing tool.\n\n#### Shared Infrastructure (from other skills)\n\nOther skills call `oma image generate --output json` and parse the JSON manifest from stdout.\n\n### Output Layout\n\nFilenames follow `\u003Cvendor>[-\u003Cmodel>]-\u003Cshortid>[-\u003Cn>].\u003Cext>` — the model segment is omitted for `antigravity` (opaque model), the extension reflects the sniffed format (JPEG is common), and `-\u003Cn>` appears only when `-n` > 1.\n\n```\n.agents\u002Fresults\u002Fimages\u002F\n├── 20260424-143052-ab12cd\u002F                    # single-vendor run\n│   ├── pollinations-flux-ab12cd.jpg\n│   │   (or codex-gpt-image-2-ab12cd.png, antigravity-ab12cd.jpg)\n│   └── manifest.json\n└── 20260424-143122-7z9kqw-compare\u002F            # --vendor all run\n    ├── codex-gpt-image-2-7z9kqw.png\n    ├── pollinations-flux-7z9kqw.jpg\n    ├── antigravity-7z9kqw.jpg\n    └── manifest.json\n```\n\n## References\n\n### Configuration\n\nProject-specific settings: the `image:` section of `.agents\u002Foma-config.yaml`, which `oma update` preserves. Shipped defaults live in the CLI (`DEFAULTS` in `cli\u002Fcommands\u002Fimage\u002Fconfig.ts`) — write only the keys you change. The legacy `config\u002Fimage-config.yaml` is no longer read by the CLI; migration 022 moves anything you had changed there into oma-config (and deletes the file when it was never edited).\nEnv vars: `OMA_IMAGE_DEFAULT_VENDOR`, `OMA_IMAGE_DEFAULT_OUT`, `OMA_IMAGE_YES`, `POLLINATIONS_API_KEY`.\n\n- Execution steps (follow for the selected task): `resources\u002Fexecution-protocol.md`\n- Vendor matrix: `resources\u002Fvendor-matrix.md`\n- Prompt tips: `resources\u002Fprompt-tips.md`\n- Checklist (run before handoff): `resources\u002Fchecklist.md`\n- Context loading: `..\u002F_shared\u002Fcore\u002Fcontext-loading.md`","# Image Agent - Multi-Vendor Image Router\n\n## Scheduling\n\n### Goal\nGenerate images and visual assets through authenticated multi-vendor routing while preserving prompt clarity, reference-guided regeneration, cost controls, and reproducible output manifests.\n\n### Intent signature\n- User asks to generate images, visual assets, illustrations, product photos, concept art, mockups, or AI art.\n- Another skill needs shared image-generation infrastructure.\n- User provides reference images for a new generated variation or asks for vendor comparison.\n\n### When to use\n\n- Generating images, visual assets, illustrations, product photos, concept art\n- Comparing output between multiple image models for the same prompt\n- Producing images from prompts within editor workflows (Claude Code, Codex, Gemini CLI)\n- Other skills needing image generation infrastructure (shared invocation)\n- Regenerating a new image that follows a supplied reference's subject, style, lighting, or composition\n\n### When NOT to use\n\n- Pixel-level editing, masking, inpainting, object removal, compositing, cropping, resizing, or format conversion -> out of scope; use an image editor or a tool that explicitly supports that operation.\n- Generating videos or audio -> out of scope\n- Inline vector art \u002F SVG composition from structured data -> use a templating skill\n- Simple asset resizing or format conversion -> use a dedicated image library\n\n### Expected inputs\n- Image prompt or creative brief\n- Optional vendor, size, quality, count, output directory, and reference images\n- Authentication\u002Fenvironment state for Codex, Pollinations, or Gemini\n\n### Expected outputs\n- Generated image files under `.agents\u002Fresults\u002Fimages\u002F` or requested output directory\n- `manifest.json` with prompt, vendor, model, and reproducibility metadata\n- Vendor comparison outputs when `--vendor all` is used\n\n### Dependencies\n- `oma image generate` CLI and vendor authentication\n- Codex image generation, Pollinations API, or Gemini API\u002FCLI strategy\n- `resources\u002Fvendor-matrix.md`, `resources\u002Fprompt-tips.md`, and the `image:` section of `.agents\u002Foma-config.yaml`\n\n### Control-flow features\n- Branches by prompt ambiguity, vendor auth, cost threshold, reference-guided regeneration support, path safety, and safety\u002Ftimeout exit codes\n- Calls external vendor APIs\u002FCLIs\n- Reads reference images and writes generated images plus manifests\n\n## Structural Flow\n\n### Entry\n1. Validate that the request contains enough subject, setting, style, usage, and aspect-ratio signal.\n2. Classify a supplied image as a reference for a newly generated variation, or as a request for unsupported pixel\u002Fmask editing.\n3. Check authentication, cost guardrails, output path, and count limits.\n\n### Scenes\n1. **PREPARE**: Clarify or amplify prompt and choose vendor strategy.\n2. **ACQUIRE**: Validate auth, references, output path, and provider availability.\n3. **ACT**: Invoke `oma image generate` with selected vendor(s), prompt, references, and options.\n4. **VERIFY**: Check manifest, output files, exit code, and provider result.\n5. **FINALIZE**: Return output paths and relevant warnings.\n\n### Transitions\n- If prompt lacks required signal, clarify or show amplified prompt before generation.\n- If `--vendor all` is requested, require every requested vendor to be available.\n- If the request is reference-guided regeneration and the selected vendor supports references, pass it automatically.\n- If the request requires pixel, mask, crop, resize, or format editing, stop before generation and name the unsupported operation and an appropriate editor route.\n- If estimated cost exceeds guardrail, require confirmation unless bypassed.\n\n### Failure and recovery\n- If auth is missing, report vendor-specific authentication requirement.\n- If reference support is unavailable for the selected vendor, reject with actionable guidance.\n- If local CLI is outdated, ask user to run `oma update`.\n- If generation times out or is blocked, surface exit code and provider status.\n\n### Exit\n- Success: images and manifest exist in the output directory.\n- Partial success: some vendors fail in comparison mode and failures are reported.\n- Failure: no image is produced and the route\u002Fcost\u002Fauth\u002Fsafety blocker is explicit.\n\n## Logical Operations\n\n### Actions\n| Action | SSL primitive | Evidence |\n|--------|---------------|----------|\n| Validate prompt completeness | `VALIDATE` | Clarification protocol |\n| Select vendor strategy | `SELECT` | Vendor matrix and auth state |\n| Read reference images | `READ` | `--reference` paths |\n| Call generation CLI\u002FAPI | `CALL_TOOL` | `oma image generate` |\n| Write image outputs | `WRITE` | Image files and manifest |\n| Validate result | `VALIDATE` | Exit code, manifest, files |\n| Report output | `NOTIFY` | Final path summary |\n\n### Tools and instruments\n- `oma image generate`, `oma image doctor`, `oma image vendor list`\n- Codex, Pollinations, and Gemini provider paths\n- Prompt tips, vendor matrix, and image config\n\n### Canonical command path\n```bash\noma image doctor\noma image generate \"\u003Cprompt>\" --vendor auto --size auto --quality auto --output json\n```\n\nWith reference images:\n```bash\noma image generate --reference \"\u003Cabsolute-path>\" --vendor codex \"\u003Cprompt>\"\n```\n\n### Resource scope\n| Scope | Resource target |\n|-------|-----------------|\n| `LOCAL_FS` | Reference images, generated images, manifests |\n| `PROCESS` | Provider CLIs and image router commands |\n| `NETWORK` | Pollinations\u002FGemini or provider APIs |\n| `CREDENTIALS` | Provider auth and API keys |\n\n### Preconditions\n- Prompt is sufficiently specified or user approves amplification.\n- Required vendor auth and output permissions exist.\n- Reference paths are accessible when used.\n\n### Effects and side effects\n- Creates image files and manifests.\n- May call paid or rate-limited provider APIs.\n- May read attached\u002Freference images.\n\n### Guardrails\n\n1. **Clarify before invoking**: if the user's request is ambiguous about subject, style, composition, or usage context, **ask the user first** or **amplify the prompt explicitly** (showing the user the expanded version for approval). Do NOT silently generate from a vague prompt. See `Clarification Protocol` below.\n2. **Authentication-aware dispatch**: detect which vendor CLIs are available and run only those; with `--vendor all`, every requested vendor must be available (strict). Caveat: the `antigravity` health check verifies installation only (`agy --version`) — a signed-out agy passes health and fails at generate time with agy's own error.\n3. **Cost guardrail**: confirm before executing runs whose estimated cost is ≥ `$0.20` (configurable). `--yes` \u002F `OMA_IMAGE_YES=1` bypass. Default vendors `pollinations` (flux\u002Fzimage) and `antigravity` (nano-banana via Gemini Code Assist) are free, so auto-triggering on keywords is safe. **Non-interactive contexts** (agents, CI — no TTY on stdin): the CLI cannot prompt, so a run at\u002Fover the threshold exits 1 with a message naming `--yes`. Calling agents must confirm the estimated cost with the user in-conversation (use `--dry-run` to get the estimate), then re-run with `-y`.\n4. **Path safety**: output paths outside `$PWD` require `--allow-external-out`.\n5. **Cancellable**: SIGINT\u002FSIGTERM aborts in-flight provider calls and the orchestrator.\n6. **Deterministic outputs**: every run writes `manifest.json` next to the images for reproducibility.\n7. **Max `n` = 5**: wall-time bound.\n8. **Exit codes align with `oma search fetch`** (0, 1, 2=safety, 3=not-found, 4=invalid-input, 5=auth-required, 6=timeout).\n\n### Clarification Protocol\n\nBefore invoking `oma image generate`, the calling agent runs this checklist against the user's request. **If any answer is \"no \u002F unknown\", clarify with the user first.**\n\n**Required signal (must be present or inferable):**\n- [ ] **Subject**: what is the primary thing in the image? (object, person, scene)\n- [ ] **Setting \u002F backdrop**: where is it? (context, environment)\n\n**Strongly recommended (ask if absent AND not inferable from context):**\n- [ ] **Style**: photorealistic, illustration, 3D render, oil painting, concept art, flat vector, …?\n- [ ] **Mood \u002F lighting**: bright vs moody, warm vs cool, dramatic vs minimal\n- [ ] **Usage context**: hero image, icon, thumbnail, product shot, poster? (dictates aspect ratio + composition)\n- [ ] **Aspect ratio \u002F resolution**: any `WxH` where each edge is a multiple of 16 between 16 and 3840 and aspect ∈ [1:3, 3:1] (e.g. `1024x1024` square, `2048x1152` 16:9, `3840x2160` 4K UHD, `1024x1536` portrait), or `auto`.\n\n**Amplification shortcut.** For brief prompts (e.g. \"a red apple\"), do not pop clarifying questions if the request is genuinely that simple. Instead **amplify inline and show the user** the expanded version before invoking:\n\n> User: \"a red apple\"\n> Agent: \"I'll generate this as: *a single glossy red apple centered on a clean white background, soft studio lighting, photorealistic, shallow depth of field, 1024×1024*. Shall I proceed, or would you like a different style\u002Fcomposition?\"\n\nSkip both clarification and amplification when the user has clearly authored a full creative brief (≥ 2 of: subject + style + lighting + composition). Respect their prompt verbatim.\n\n**Category-specific briefs** (app mockup, poster, thumbnail, infographic, comic panel, avatar): consult `resources\u002Fprompt-tips.md` → *External Prompt Libraries*.\n\n**Output language.** Generation prompts are sent to the provider in English (image models are trained predominantly on English captions). Translate the user's request if they wrote in another language, and show them the translated version during amplification so they can correct misreadings.\n\n### Vendors\n\nThis skill follows oh-my-agent's CLI-first concept: whenever a vendor's native CLI can drive generation (and return raw bytes), the subprocess path is preferred over direct API keys. Direct API is only used as a fallback for vendors whose CLI can't yet emit raw image bytes.\n\n| Vendor | Strategy | Models | Trigger |\n|--------|----------|--------|---------|\n| `codex` | CLI-first via `codex exec` over ChatGPT OAuth (`codex login`), built-in `image_gen` | `gpt-image-2` | Logged in via Codex CLI (no API key) |\n| `pollinations` | Direct HTTP via `gen.pollinations.ai\u002Fv1\u002Fimages\u002Fgenerations` (free signup for key) | Free: `flux`, `zimage`. Credit-gated: `qwen-image`, `wan-image`, `gpt-image-2`, `klein`, `kontext`, `gptimage`, `gptimage-large` | `POLLINATIONS_API_KEY` set (free at https:\u002F\u002Fenter.pollinations.ai). No native CLI exists. |\n| `antigravity` | `agy -p --dangerously-skip-permissions --add-dir \u003CoutDir>` — Antigravity's agentic CLI runs over the user's Gemini Code Assist subscription. agy writes raw bytes to absolute target paths we embed in the prompt; the provider sniffs format via magic bytes and renames the file extension to match. Model selection is opaque — agy picks internally, we never name a model. | (opaque — chosen by agy) | `agy` CLI installed + signed in. No API key, no per-image charge. |\n\n> The direct Gemini path (`gemini -p` stream, `generativelanguage.googleapis.com` API) is deprecated. `agy` is the supported Gemini image route — it's free with Gemini Code Assist and doesn't require billing on AI Studio.\n\n### Invocation\n\n#### Standalone\n\n```\n\u002Foma-image a red apple on white background\n\u002Foma-image --vendor all --size 1536x1024 jeju coastline at sunset\n\u002Foma-image -n 3 --quality high --out .\u002Fhero \"minimalist dashboard hero illustration\"\n```\n\n#### Shell CLI\n\n```\noma image generate \"\u003Cprompt>\" [--vendor auto|codex|pollinations|antigravity|all] [-n 1..5] \\\n                             [--size WxH|auto] \\\n                             [--quality low|medium|high|auto] \\\n                             [--model \u003Cname>] \\\n                             [--output-dir \u003Cdir>] [--allow-external-output] \\\n                             [-r \u003Cpath>]... \\\n                             [--timeout 180] [-y] [--no-prompt-in-manifest] \\\n                             [--dry-run] [--output text|json]\noma image doctor\noma image vendor list\n```\n\n`--model \u003Cname>` overrides the vendor's default model for this run — e.g. `--vendor pollinations --model zimage`, or a credit-gated Pollinations model like `gpt-image-2`. It applies to every vendor in the run set, so combine it with an explicit `--vendor`; `antigravity` ignores it (agy picks its model internally).\n\n#### Reference-guided regeneration (`-r`, `--reference`)\n\nAttach up to 10 reference images (PNG\u002FJPEG\u002FGIF\u002FWebP, ≤ 5MB each) to guide a **newly generated** image's style, subject identity, lighting, or composition. Repeatable or comma-separated. The output is a new generation, not a pixel-preserving edit of the input.\n\n```\noma image generate -r ~\u002FDownloads\u002Fotter.jpeg \"same otter in dramatic lighting\"\noma image generate -r a.png -r b.png \"blend these two styles\"\n```\n\nSupported vendors:\n\n| Vendor | Support | How |\n|--------|---------|-----|\n| `codex` (gpt-image-2) | PASS | Passes `-i \u003Cpath>` to `codex exec` |\n| `antigravity` | PASS | Refs copied to a per-run temp dir, `agy --add-dir \u003Ctmpdir>` grants access, paths inlined into the prompt |\n| `pollinations` | N\u002FA | Rejected with exit code 4 when explicitly selected (requires URL hosting; see PR #2 roadmap). Under `--vendor auto`, reference-unsupported vendors are silently dropped from the run set instead. |\n\n**Paths**: absolute or relative to `$CWD`. Host CLIs usually expose attached images via:\n- **Claude Code**: `~\u002F.claude\u002Fimage-cache\u002F\u003Csession>\u002FN.png` (surfaced in system messages as `[Image: source: \u003Cpath>]`)\n- **Antigravity**: workspace upload directory (exact path shown in IDE)\n- **Codex CLI as host**: user must pass the filesystem path explicitly; in-conversation attachments are not forwarded\n\n#### Agent Behavior: Auto-forward Attached References (MANDATORY)\n\nWhen ALL of the following are true, the calling agent MUST pass the attached image via `--reference \u003Cpath>` automatically. Never describe the image in prose as a workaround.\n\n1. The user asks for a new generated variation (for example, \"same subject in dramatic lighting\", \"make a new illustration in this style\", or \"regenerate this at a different angle\").\n2. A host-surfaced attached image is visible to the agent (e.g. a Claude Code system message with `[Image: source: \u003Cpath>]`, or an Antigravity workspace upload path, or an explicit filesystem path in the user's message).\n3. The selected vendor supports references (`codex` or `antigravity`).\n\n**Required action**: invoke `oma image generate --reference \u003Cabsolute-path> --vendor \u003Ccodex|antigravity> \"\u003Cprompt>\"`. If the user didn't specify a vendor, default to `codex` (CLI-first, widest availability). Do NOT:\n\n- Fall back to prose description (\"I'll describe the otter's appearance...\").\n- Ask the user to re-type or re-attach the path.\n- Claim the CLI doesn't support references without first running `oma image generate --help` to verify.\n\n**If the local CLI is outdated** (`--reference` is missing from `--help`): tell the user to run `oma update` once, then retry. Do not silently degrade to prose.\n\n**If the reference path is from Claude Code's `image-cache`**: note to the user that the path is session-scoped and suggest copying the file to a durable location if they want to reuse it later. Still proceed with the generation.\n\n**Unsupported edit requests**: Do not route requests such as \"remove this object\", \"change only the background\", \"paint inside this mask\", \"keep every pixel except\", \"crop\", \"resize\", or \"convert this file\" through `--reference`. They need pixel\u002Fmask editing or deterministic image processing, neither of which this CLI provides. Say whether the request can instead be phrased as a reference-guided regeneration; otherwise route to an appropriate image editor or processing tool.\n\n#### Shared Infrastructure (from other skills)\n\nOther skills call `oma image generate --output json` and parse the JSON manifest from stdout.\n\n### Output Layout\n\nFilenames follow `\u003Cvendor>[-\u003Cmodel>]-\u003Cshortid>[-\u003Cn>].\u003Cext>` — the model segment is omitted for `antigravity` (opaque model), the extension reflects the sniffed format (JPEG is common), and `-\u003Cn>` appears only when `-n` > 1.\n\n```\n.agents\u002Fresults\u002Fimages\u002F\n├── 20260424-143052-ab12cd\u002F                    # single-vendor run\n│   ├── pollinations-flux-ab12cd.jpg\n│   │   (or codex-gpt-image-2-ab12cd.png, antigravity-ab12cd.jpg)\n│   └── manifest.json\n└── 20260424-143122-7z9kqw-compare\u002F            # --vendor all run\n    ├── codex-gpt-image-2-7z9kqw.png\n    ├── pollinations-flux-7z9kqw.jpg\n    ├── antigravity-7z9kqw.jpg\n    └── manifest.json\n```\n\n## References\n\n### Configuration\n\nProject-specific settings: the `image:` section of `.agents\u002Foma-config.yaml`, which `oma update` preserves. Shipped defaults live in the CLI (`DEFAULTS` in `cli\u002Fcommands\u002Fimage\u002Fconfig.ts`) — write only the keys you change. The legacy `config\u002Fimage-config.yaml` is no longer read by the CLI; migration 022 moves anything you had changed there into oma-config (and deletes the file when it was never edited).\nEnv vars: `OMA_IMAGE_DEFAULT_VENDOR`, `OMA_IMAGE_DEFAULT_OUT`, `OMA_IMAGE_YES`, `POLLINATIONS_API_KEY`.\n\n- Execution steps (follow for the selected task): `resources\u002Fexecution-protocol.md`\n- Vendor matrix: `resources\u002Fvendor-matrix.md`\n- Prompt tips: `resources\u002Fprompt-tips.md`\n- Checklist (run before handoff): `resources\u002Fchecklist.md`\n- Context loading: `..\u002F_shared\u002Fcore\u002Fcontext-loading.md`",{"name":5,"description":7},"master","5fab8004af1390aa17b3447f2780766c8bf2482c",".agents\u002Fskills\u002Foma-image",0,"https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate","https:\u002F\u002Fgithub.com\u002Fvasceannie","thingswithstuff.io","2026-11-12 20:22:46",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add vasceannie\u002Fslopgate --skill oma-image",{"tool":63,"command":64},"bunx","bunx skills add vasceannie\u002Fslopgate --skill oma-image",{"tool":66,"command":67},"pnpm","pnpm dlx skills add vasceannie\u002Fslopgate --skill oma-image","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002FSKILL.md",true,[71,77,81,85,89],{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"config\u002Fimage-config.yaml","REFERENCE","参考文档",2308,"https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002Fconfig\u002Fimage-config.yaml",{"path":78,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":79,"sourceHtmlUrl":80},"resources\u002Fchecklist.md",1138,"https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002Fresources\u002Fchecklist.md",{"path":82,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":83,"sourceHtmlUrl":84},"resources\u002Fexecution-protocol.md",7643,"https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002Fresources\u002Fexecution-protocol.md",{"path":86,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":87,"sourceHtmlUrl":88},"resources\u002Fprompt-tips.md",4696,"https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002Fresources\u002Fprompt-tips.md",{"path":90,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":91,"sourceHtmlUrl":92},"resources\u002Fvendor-matrix.md",6134,"https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Foma-image\u002Fresources\u002Fvendor-matrix.md",[94,123,150,187,212,233],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":103,"publishedAt":44},"01M71951JDWCYMS8GDC2C7395D","type-strictness","Enforce strict type compliance by eliminating `Any` types and `# type: ignore` comments. Use when:\n(1) Editing code that contains `Any` as a type annotation\n(2) Code has `# type: ignore` or `# pyright: ignore` comments\n(3) User requests type fixes or type strictness enforcement\n(4) Importing from untyped third-party libraries\n(5) User asks to \"fix types\", \"remove Any\", or \"add proper typing\"\n(6) Pyrefly or mypy reports type errors that need resolution\n(7) Code review reveals weak typing patterns\nFollows hierarchical resolution: codebase inference → type stubs → library inspection → type guards → casting (last resort).","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002Fbundle\u002Fshared\u002Fskills\u002Ftype-strictness\u002FSKILL.md",[100],{"categoryNo":101,"categoryName":102,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000007","代码审查",[104,105,106,107,108,109,110,111,112,113,38,114,115,116,117,118,119,120,121,122],"type","type strictness","typing","types","remove","code review","fix","errors","inspection","when","review","reports","libraries","casting","pyrefly","mypy","pyright","library","patterns",{"skillNo":124,"name":125,"nameZh":6,"description":126,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":133,"publishedAt":44},"01M7194XGBQXSE7Y5BZ4K5CSTK","slopgate-test-extender","Extend Python test suites with parametrization, fixture reuse, and quality enforcement. Use when asked to add tests, extend existing tests, improve test coverage, consolidate duplicate tests, fix test quality issues, or when Slopgate reports untested-production-code, missing test references, or coverage gaps. Triggers on requests like \"add tests for X\", \"extend these tests\", \"parametrize tests\", \"consolidate tests\", \"fix test quality\", \"improve test coverage\", \"write tests for this function\", \"test this module\", \"behavior tests\", \"integration tests\", \"missing tests\", \"coverage holes\", \"untested code\", or when a hook denies with test-related findings.","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002Fbundle\u002Fshared\u002Fskills\u002Fslopgate-test-extender\u002FSKILL.md",[129,130],{"categoryNo":101,"categoryName":102,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":131,"categoryName":132,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量",[134,135,136,137,138,110,139,140,141,142,143,144,113,38,145,146,115,147,148,149],"slopgate test extender","test","coverage","fixture","testing","issues","write tests","improve","only","cleanup","loop","candidate","consolidate","references","opencode","orchestrator",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":171,"publishedAt":44},"01M7194RMVQS5YJJ4ZEFXX20VS","slopgate-intelligent-coding-patterns","Pattern-selection playbook for agents editing code with complex branching, stateful parsers\u002Fworkflows, duplicated skeletons, policy checks, or ordered transforms. Load before refactoring if\u002Felif dispatch chains, boolean-flag state machines, repeated validation predicates, parallel provider\u002Flanguage implementations, or regex\u002Fcontent transformation pipelines. Also load when asked to \"apply a design pattern\", \"which pattern should I use here\", \"refactor this switch statement\", \"replace if-elif chain\", \"state machine pattern\", \"strategy pattern\", \"template method\", \"pipeline pattern\", \"specification pattern\", \"dispatch table\", \"guard clause refactor\", \"reduce cyclomatic complexity\", or when Slopgate fires PY-CODE-014, PY-CODE-017, PY-CODE-018, QUALITY-LINT-001, or PY-QUALITY-010. Use with behavior-locking tests and avoid pattern theater for simple one-off code.","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002Fbundle\u002Fshared\u002Fskills\u002Fslopgate-intelligent-coding-patterns\u002FSKILL.md",[156,157,160,163,168],{"categoryNo":101,"categoryName":102,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":158,"categoryName":159,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":161,"categoryName":162,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":164,"categoryName":165,"taxonomyType":166,"taxonomyTypeDesc":167},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",{"categoryNo":169,"categoryName":170,"taxonomyType":166,"taxonomyTypeDesc":167},"01JBSEEDCAT000000000000032","设计师",[172,173,122,174,175,176,177,178,179,180,181,182,183,142,143,184,185,113,186,38],"coding patterns","slopgate intelligent coding patterns","coding","stateful","complexity","branch","which","chain","parsers","pipeline","selection","design","utility","workflows","reduce",{"skillNo":188,"name":189,"nameZh":6,"description":190,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":191,"categories":192,"keywords":195,"publishedAt":44},"01M7194KTXBM85GB0W7JKFQD55","slopgate-hygiene-orchestrator","Orchestrate systematic elimination of linter warnings and errors across Python, TypeScript, and Rust codebases. Use when asked to \"clean up lints\", \"fix all warnings\", \"run hygiene\", \"eliminate type errors\", \"orchestrate code quality fixes\", \"fix lint errors\", \"resolve quality gate failures\", \"pass CI checks\", \"reduce tech debt\", \"refactor to pass hooks\", or when a Slopgate hook denies with QUALITY-LINT-001, PY-CODE-017, PY-CODE-018, or any collector finding. This skill coordinates multiple code-hygiene-enforcer agents working in parallel on non-conflicting file groupings, tracks progress persistently via JSON exports, and loops until all issues are resolved. CRITICAL: Never run linters directly—always use `slopgate lint check` as the primary gate, discover project-specific tools from config files, and export findings to JSON for tracking.","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002Fbundle\u002Fshared\u002Fskills\u002Fslopgate-hygiene-orchestrator\u002FSKILL.md",[193,194],{"categoryNo":101,"categoryName":102,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":161,"categoryName":162,"taxonomyType":21,"taxonomyTypeDesc":22},[196,197,149,198,110,199,200,201,202,203,143,186,204,205,206,207,208,209,210,211],"slopgate hygiene orchestrator","hygiene","project","ci","local","clean","never run","files","run","lint check","systematic","config","typescript","debt","quality-gates","hook",{"skillNo":213,"name":214,"nameZh":6,"description":215,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":216,"categories":217,"keywords":220,"publishedAt":44},"01M7194G50FZYXXT8VNK360RPA","slopgate-code-smell-utility-locator","Autoactivate when code-smell hooks or reviews mention feature envy, thin wrappers, duplicate helpers, duplicate code,\nscattered helpers, builders\u002Ffactories\u002Fconfig\u002Fconstants\u002Fdataclass reuse, PY-CODE-012, PY-CODE-013, QUALITY-LINT-001,\nor repeated slopgate code-quality denials. Also load when asked to \"find existing helpers\", \"consolidate utilities\",\n\"search for duplicate code\", \"where is this already implemented\", \"reuse existing logic\", \"don't duplicate this\",\n\"find similar functions\", \"utility inventory\", \"code smell radar\", or before adding new helpers\u002Fwrappers so agents can\nfind existing shared utilities and consolidate instead of duplicating.","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002Fbundle\u002Fshared\u002Fskills\u002Fslopgate-code-smell-utility-locator\u002FSKILL.md",[218,219],{"categoryNo":101,"categoryName":102,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":164,"categoryName":165,"taxonomyType":166,"taxonomyTypeDesc":167},[184,221,38,222,136,223,224,225,226,142,227,113,228,146,148,229,230,207,231,232],"slopgate code smell utility locator","shared","reviews","inventory","search","hooks","new","coordination","radar","code-quality","Utilities","dataclass",{"skillNo":234,"name":235,"nameZh":6,"description":236,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":237,"categories":238,"keywords":241,"publishedAt":44},"01M7194AWG1Q37F9756H4SFDH9","slopgate-code-hygiene-refactor","Recovery playbook for Slopgate and project quality failures. Load after hook\u002Flint denials such as PY-CODE-017 flat sibling modules, PY-CODE-018 oversized modules, QUALITY-LINT-001 post-edit quality backstops, PY-QUALITY-010 magic numbers \u002F repeated literals, PY-CODE-012\u002FPY-CODE-013 smell findings, type-suppression bans, duplicate helpers, or repeated hygiene retry loops. Also load when asked to \"refactor this module\", \"split this file\", \"extract a class\", \"reduce complexity\", \"fix god class\", \"flatten module hierarchy\", \"move code to a package\", \"extract helpers\", \"consolidate duplicates\", \"replace magic numbers with constants\", or when a hook blocks an edit with structural complaints. Focuses on preserving public APIs while refactoring toward packages, cohesive modules, typed seams, and focused tests.","https:\u002F\u002Fgithub.com\u002Fvasceannie\u002Fslopgate\u002Fblob\u002Fmaster\u002Fbundle\u002Fshared\u002Fskills\u002Fslopgate-code-hygiene-refactor\u002FSKILL.md",[239,240],{"categoryNo":101,"categoryName":102,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":164,"categoryName":165,"taxonomyType":166,"taxonomyTypeDesc":167},[242,243,38,244,197,245,176,246,247,198,248,224,249,110,142,184,113,186,228,146],"code hygiene","slopgate code hygiene refactor","refactor","modules","duplicates","retry","move","structural"]