[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79RSCXJ4YA0M57PV4F876R5":3,"skill-same:01M79RSCXJ4YA0M57PV4F876R5":49},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":30,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M79RSCXJ4YA0M57PV4F876R5","api-server-mcp",null,"REST API server and MCP protocol integration","jamon8888\u002Fhacienda-engine","jamon8888","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F145190?v=4","01M79RNFE7K7JZ1AZG8TACAV98","Jamin Rubio","01M79RNFEV6ZC1SZ4K04SHDY8Z",2,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Fapi-server-mcp\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\ndescription: \"REST API server and MCP protocol integration\"\nname: api-server-mcp\npriority: critical\n---\n\n# API Server & MCP Protocol\n\n**Axum server design for document extraction endpoints, middleware, async processing, and Model Context Protocol integration for AI agents**\n\n> **Copied from upstream `xberg` and now known-stale in one place** (same class of gap\n> the Unreleased changelog already flagged for `crate-structure`\u002F`alef-generated-bindings`):\n> the \"MCP Tools\" section below lists three tools (`extract`, `extract_batch`,\n> `get_capabilities`). Reading the pinned xberg commit's actual source\n> (`crates\u002Fxberg\u002Fsrc\u002Fmcp\u002Fserver.rs`) shows eight: `extract`, `extract_batch`,\n> `detect_mime_type`, `cache_stats`, `cache_clear`, `cache_manifest`, `cache_warm`,\n> `get_version`. This also does not describe hacienda's own MCP server —\n> `hacienda-mcp` (`hacienda mcp serve`), which wraps `HaciendaFacade`, not xberg's API —\n> see that crate's module doc and\n> `superpowers\u002Fspecs\u002F2026-08-13-M1-mcp-server-and-cli-sdk-parity-design.md`.\n\n## Xberg API Architecture\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002F`, `crates\u002Fxberg-cli\u002F`\n\nXberg provides a dual REST API + MCP server built with Axum + Tokio.\n\n```text\nRequest Flow:\nHTTP Client \u002F AI Agent (Claude)\n    |\n[Transport Layer]\n├── REST API (Axum HTTP)\n└── MCP Protocol (HTTP or Stdio)\n    |\n[Middleware Layer]\n├── CORS, Request Logging (TraceLayer)\n├── Request\u002FResponse size limits\n└── Rate limiting (optional)\n    |\n[Router]\n├── REST Endpoints\n│   ├── POST \u002Fextract - File upload extraction\n│   ├── POST \u002Fextract-url - URL-based extraction\n│   ├── GET \u002Fformats - List supported formats\n│   ├── GET \u002Fhealth - Server health check\n│   ├── POST \u002Fbatch - Batch document processing\n│   ├── GET \u002Fcache\u002Fstats - Cache statistics\n│   └── DELETE \u002Fcache - Clear extraction cache\n├── MCP Endpoints\n│   ├── POST \u002Fmcp\u002Ftools - List available tools\n│   ├── POST \u002Fmcp\u002Ftools\u002Fcall - Call a tool\n│   ├── GET \u002Fmcp\u002Fresources - List resources\n│   ├── GET \u002Fmcp\u002Fresources\u002F:uri - Read resource\n│   ├── GET \u002Fmcp\u002Fprompts - List prompts\n│   └── GET \u002Fmcp\u002Fprompts\u002F:name - Get prompt\n    |\n[Handler \u002F Tool Layer]\n├── extract_handler \u002F extract tool\n├── extract_async_handler \u002F extract_batch tool\n├── health_handler \u002F get_capabilities tool\n└── format_handler\n    |\n[Extraction Core]\n├── Format detection\n├── Extraction pipeline\n├── Post-processing (chunking, embeddings)\n└── Result formatting\n    |\nJSON Response \u002F MCP ToolResult\n```\n\n## Server Setup & Configuration\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002Fserver.rs`\n\nServer initialization pattern: Create `ApiState` (holds `ExtractionConfig` + `ExtractionCache`), build Axum `Router` with all REST + MCP routes, apply middleware layers (body limits, CORS, tracing), serve via `tokio::net::TcpListener`.\n\nKey middleware layers applied in order:\n\n- `DefaultBodyLimit::max(100MB)` + `RequestBodyLimitLayer` -- configurable via env vars\n- `CorsLayer::permissive()` -- restrict in production via `CORS_ALLOWED_ORIGINS`\n- `TraceLayer::new_for_http()` -- request\u002Fresponse logging\n\n## Core REST Handlers\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002Fhandlers.rs`\n\n| Handler               | Method            | Description                                                                                            |\n| --------------------- | ----------------- | ------------------------------------------------------------------------------------------------------ |\n| `extract_handler`     | POST \u002Fextract       | Multipart files, URL fields, or JSON inputs; build `ExtractInput` and call `extract()` \u002F `extract_batch()` |\n| `extract_async_handler` | POST \u002Fextract-async | Queue the same unified extraction input shape for async processing                                      |\n| `health_handler`      | GET \u002Fhealth       | Report status, version, uptime, feature availability (OCR, embeddings), cache stats                    |\n| `formats_handler`     | GET \u002Fformats      | Return supported format categories (office, pdf, images, web, email, archives, academic)               |\n| `cache_stats_handler` | GET \u002Fcache\u002Fstats  | Hit\u002Fmiss counts and hit rate                                                                           |\n| `cache_clear_handler` | DELETE \u002Fcache     | Clear LRU cache                                                                                        |\n\n## Caching Strategy\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fcache\u002Fmod.rs`\n\nLRU cache keyed by `SHA256(file_content)`, stores `Arc\u003CExtractionResult>`. Default 1000 entries. Thread-safe via `RwLock`. Tracks hit\u002Fmiss counters with `AtomicU64` for stats endpoint.\n\n## Error Handling\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002Ferror.rs`\n\n`ApiError` enum maps to HTTP status codes:\n\n- `MissingFile` -> 400, `FileNotFound` -> 404\n- `OnnxRuntimeMissing` \u002F `TesseractMissing` -> 503 (with remediation message)\n- `PayloadTooLarge` -> 413\n- `ExtractionFailed` \u002F `InvalidConfig` \u002F `UnsupportedFormat` -> 500\n\n## MCP Server Implementation\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fmcp\u002Fserver.rs`\n\nThe MCP server allows Claude and other AI agents to call Xberg extraction functions through the Model Context Protocol.\n\n### MCP Tools (Callable Functions)\n\nThree tools are registered:\n\n| Tool               | Purpose                                                   | Required Params |\n| ------------------ | --------------------------------------------------------- | --------------- |\n| `extract`          | Extract text\u002Ftables\u002Fmetadata from bytes, paths, file URIs, or URLs | `input`        |\n| `extract_batch`    | Extract from multiple unified inputs in parallel                 | `inputs[]`     |\n| `get_capabilities` | List supported formats, features, backends                | (none)          |\n\n**Tool registration pattern** (example: `extract`):\n\n```rust\n\u002F\u002F Define Tool with name, description, JSON Schema inputSchema\n\u002F\u002F Register with server.register_tool(tool, handler_fn)\n\u002F\u002F Handler: parse params -> build ExtractInput + ExtractionConfig -> call extract() -> return ToolResult as JSON\n```\n\n`extract` optional params: `mime_type`, `filename`, `extract_tables`, `extract_images`, `ocr_enabled`, `extract_metadata`, `chunking_preset`, `generate_embeddings`, and URL ingestion options.\n\n### MCP Resources (Static Knowledge)\n\nThree resources provide static information to agents:\n\n- `xberg:\u002F\u002Fformats` -- Supported format list as JSON\n- `xberg:\u002F\u002Ffeatures` -- Cross-binding feature matrix (from `FEATURE_MATRIX.md`)\n- `xberg:\u002F\u002Fapi-reference` -- Generated API documentation\n\n### MCP Prompts (Agent Templates)\n\nTwo prompts guide agent extraction workflows:\n\n- `extract_for_rag` -- Document type-specific RAG extraction guidance (research paper, contract, report). Recommends chunking preset and embedding config.\n- `batch_document_processing` -- Optimal concurrency, grouping, and error handling for batch workflows.\n\n### MCP Transport Protocols\n\n- **HTTP\u002FREST**: MCP routes mounted alongside REST API on separate `\u002Fmcp\u002F` prefix\n- **Stdio**: JSON-RPC 2.0 over stdin\u002Fstdout for local CLI integration (e.g., Claude Desktop)\n\n### Integration with Claude Desktop\n\n```json\n{\n  \"mcpServers\": {\n    \"xberg\": {\n      \"command\": \"xberg-mcp\",\n      \"env\": {\n        \"XBERG_API_BASE\": \"http:\u002F\u002Flocalhost:8000\",\n        \"XBERG_MCP_TRANSPORT\": \"stdio\"\n      }\n    }\n  }\n}\n```\n\n### MCP Error Handling\n\n`ToolError` variants: `FileNotFound`, `UnsupportedFormat`, `ExtractionFailed`, `OnnxRuntimeMissing`, `TesseractMissing`, `Timeout`. Each maps to an MCP `ToolResultError` with descriptive code and message.\n\n## Environment Configuration\n\nSee `.env.example` for all configurable variables. Key categories:\n\n- **Server**: `XBERG_HOST`, `XBERG_PORT`\n- **Size limits**: `XBERG_MAX_REQUEST_BODY_BYTES` (default 100MB), `XBERG_MAX_MULTIPART_FIELD_BYTES`\n- **Features**: `XBERG_ENABLE_OCR`, `XBERG_ENABLE_EMBEDDINGS`, `XBERG_ENABLE_KEYWORDS`\n- **Cache**: `XBERG_CACHE_ENABLED`, `XBERG_CACHE_SIZE`\n- **CORS**: `CORS_ALLOWED_ORIGINS` (comma-separated)\n- **MCP**: `XBERG_MCP_HOST`, `XBERG_MCP_PORT`, `XBERG_MCP_TRANSPORT` (stdio\u002Fhttp)\n- **Logging**: `RUST_LOG=xberg=info,tower_http=debug`\n\n## Critical Rules\n\n### REST API Rules\n\n1. **Always validate multipart file uploads** - Check MIME type, size, magic bytes\n2. **Timeout long-running extractions** - Set per-handler timeout (5 min default)\n3. **Stream large files** - Never buffer entire multi-GB file in memory\n4. **Cache aggressively** - Identical files should return from cache in \u003C1ms\n5. **Parallel extraction is CPU-bound** - Limit workers to CPU count + 1\n6. **Error responses must be actionable** - Include error code and remediation suggestion\n7. **Health checks must verify features** - Report missing dependencies (ONNX, Tesseract)\n8. **Size limits are configurable** - Allow override via env var for large deployments\n9. **CORS is permissive by default** - Restrict in production via env var\n10. **Logging all requests** - Track extraction metrics for observability\n\n### MCP Rules\n\n1. **All tools must have timeout** - Prevent hanging on large files (default 5 min)\n2. **Error responses must be detailed** - Include suggestions for missing dependencies\n3. **Feature gates must be checked** - Return helpful message if feature unavailable (embeddings, OCR)\n4. **Resources should be static** - Don't query external services in resource handlers\n5. **Prompts guide agents** - Provide clear examples and best practices\n6. **Batch tools must support cancellation** - Allow agent to stop long-running batch operations\n7. **Logging all tool calls** - Track usage for analytics and debugging\n\n## Related Skills\n\n- **extraction-pipeline-patterns** - Core extraction called by handlers and MCP tools\n- **chunking-embeddings** - Optional chunking\u002Fembedding parameters in extraction\n- **ocr-backend-management** - OCR engine selection and image preprocessing\n","# API Server & MCP Protocol\n\n**Axum server design for document extraction endpoints, middleware, async processing, and Model Context Protocol integration for AI agents**\n\n> **Copied from upstream `xberg` and now known-stale in one place** (same class of gap\n> the Unreleased changelog already flagged for `crate-structure`\u002F`alef-generated-bindings`):\n> the \"MCP Tools\" section below lists three tools (`extract`, `extract_batch`,\n> `get_capabilities`). Reading the pinned xberg commit's actual source\n> (`crates\u002Fxberg\u002Fsrc\u002Fmcp\u002Fserver.rs`) shows eight: `extract`, `extract_batch`,\n> `detect_mime_type`, `cache_stats`, `cache_clear`, `cache_manifest`, `cache_warm`,\n> `get_version`. This also does not describe hacienda's own MCP server —\n> `hacienda-mcp` (`hacienda mcp serve`), which wraps `HaciendaFacade`, not xberg's API —\n> see that crate's module doc and\n> `superpowers\u002Fspecs\u002F2026-08-13-M1-mcp-server-and-cli-sdk-parity-design.md`.\n\n## Xberg API Architecture\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002F`, `crates\u002Fxberg-cli\u002F`\n\nXberg provides a dual REST API + MCP server built with Axum + Tokio.\n\n```text\nRequest Flow:\nHTTP Client \u002F AI Agent (Claude)\n    |\n[Transport Layer]\n├── REST API (Axum HTTP)\n└── MCP Protocol (HTTP or Stdio)\n    |\n[Middleware Layer]\n├── CORS, Request Logging (TraceLayer)\n├── Request\u002FResponse size limits\n└── Rate limiting (optional)\n    |\n[Router]\n├── REST Endpoints\n│   ├── POST \u002Fextract - File upload extraction\n│   ├── POST \u002Fextract-url - URL-based extraction\n│   ├── GET \u002Fformats - List supported formats\n│   ├── GET \u002Fhealth - Server health check\n│   ├── POST \u002Fbatch - Batch document processing\n│   ├── GET \u002Fcache\u002Fstats - Cache statistics\n│   └── DELETE \u002Fcache - Clear extraction cache\n├── MCP Endpoints\n│   ├── POST \u002Fmcp\u002Ftools - List available tools\n│   ├── POST \u002Fmcp\u002Ftools\u002Fcall - Call a tool\n│   ├── GET \u002Fmcp\u002Fresources - List resources\n│   ├── GET \u002Fmcp\u002Fresources\u002F:uri - Read resource\n│   ├── GET \u002Fmcp\u002Fprompts - List prompts\n│   └── GET \u002Fmcp\u002Fprompts\u002F:name - Get prompt\n    |\n[Handler \u002F Tool Layer]\n├── extract_handler \u002F extract tool\n├── extract_async_handler \u002F extract_batch tool\n├── health_handler \u002F get_capabilities tool\n└── format_handler\n    |\n[Extraction Core]\n├── Format detection\n├── Extraction pipeline\n├── Post-processing (chunking, embeddings)\n└── Result formatting\n    |\nJSON Response \u002F MCP ToolResult\n```\n\n## Server Setup & Configuration\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002Fserver.rs`\n\nServer initialization pattern: Create `ApiState` (holds `ExtractionConfig` + `ExtractionCache`), build Axum `Router` with all REST + MCP routes, apply middleware layers (body limits, CORS, tracing), serve via `tokio::net::TcpListener`.\n\nKey middleware layers applied in order:\n\n- `DefaultBodyLimit::max(100MB)` + `RequestBodyLimitLayer` -- configurable via env vars\n- `CorsLayer::permissive()` -- restrict in production via `CORS_ALLOWED_ORIGINS`\n- `TraceLayer::new_for_http()` -- request\u002Fresponse logging\n\n## Core REST Handlers\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002Fhandlers.rs`\n\n| Handler               | Method            | Description                                                                                            |\n| --------------------- | ----------------- | ------------------------------------------------------------------------------------------------------ |\n| `extract_handler`     | POST \u002Fextract       | Multipart files, URL fields, or JSON inputs; build `ExtractInput` and call `extract()` \u002F `extract_batch()` |\n| `extract_async_handler` | POST \u002Fextract-async | Queue the same unified extraction input shape for async processing                                      |\n| `health_handler`      | GET \u002Fhealth       | Report status, version, uptime, feature availability (OCR, embeddings), cache stats                    |\n| `formats_handler`     | GET \u002Fformats      | Return supported format categories (office, pdf, images, web, email, archives, academic)               |\n| `cache_stats_handler` | GET \u002Fcache\u002Fstats  | Hit\u002Fmiss counts and hit rate                                                                           |\n| `cache_clear_handler` | DELETE \u002Fcache     | Clear LRU cache                                                                                        |\n\n## Caching Strategy\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fcache\u002Fmod.rs`\n\nLRU cache keyed by `SHA256(file_content)`, stores `Arc\u003CExtractionResult>`. Default 1000 entries. Thread-safe via `RwLock`. Tracks hit\u002Fmiss counters with `AtomicU64` for stats endpoint.\n\n## Error Handling\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fapi\u002Ferror.rs`\n\n`ApiError` enum maps to HTTP status codes:\n\n- `MissingFile` -> 400, `FileNotFound` -> 404\n- `OnnxRuntimeMissing` \u002F `TesseractMissing` -> 503 (with remediation message)\n- `PayloadTooLarge` -> 413\n- `ExtractionFailed` \u002F `InvalidConfig` \u002F `UnsupportedFormat` -> 500\n\n## MCP Server Implementation\n\n**Location**: `crates\u002Fxberg\u002Fsrc\u002Fmcp\u002Fserver.rs`\n\nThe MCP server allows Claude and other AI agents to call Xberg extraction functions through the Model Context Protocol.\n\n### MCP Tools (Callable Functions)\n\nThree tools are registered:\n\n| Tool               | Purpose                                                   | Required Params |\n| ------------------ | --------------------------------------------------------- | --------------- |\n| `extract`          | Extract text\u002Ftables\u002Fmetadata from bytes, paths, file URIs, or URLs | `input`        |\n| `extract_batch`    | Extract from multiple unified inputs in parallel                 | `inputs[]`     |\n| `get_capabilities` | List supported formats, features, backends                | (none)          |\n\n**Tool registration pattern** (example: `extract`):\n\n```rust\n\u002F\u002F Define Tool with name, description, JSON Schema inputSchema\n\u002F\u002F Register with server.register_tool(tool, handler_fn)\n\u002F\u002F Handler: parse params -> build ExtractInput + ExtractionConfig -> call extract() -> return ToolResult as JSON\n```\n\n`extract` optional params: `mime_type`, `filename`, `extract_tables`, `extract_images`, `ocr_enabled`, `extract_metadata`, `chunking_preset`, `generate_embeddings`, and URL ingestion options.\n\n### MCP Resources (Static Knowledge)\n\nThree resources provide static information to agents:\n\n- `xberg:\u002F\u002Fformats` -- Supported format list as JSON\n- `xberg:\u002F\u002Ffeatures` -- Cross-binding feature matrix (from `FEATURE_MATRIX.md`)\n- `xberg:\u002F\u002Fapi-reference` -- Generated API documentation\n\n### MCP Prompts (Agent Templates)\n\nTwo prompts guide agent extraction workflows:\n\n- `extract_for_rag` -- Document type-specific RAG extraction guidance (research paper, contract, report). Recommends chunking preset and embedding config.\n- `batch_document_processing` -- Optimal concurrency, grouping, and error handling for batch workflows.\n\n### MCP Transport Protocols\n\n- **HTTP\u002FREST**: MCP routes mounted alongside REST API on separate `\u002Fmcp\u002F` prefix\n- **Stdio**: JSON-RPC 2.0 over stdin\u002Fstdout for local CLI integration (e.g., Claude Desktop)\n\n### Integration with Claude Desktop\n\n```json\n{\n  \"mcpServers\": {\n    \"xberg\": {\n      \"command\": \"xberg-mcp\",\n      \"env\": {\n        \"XBERG_API_BASE\": \"http:\u002F\u002Flocalhost:8000\",\n        \"XBERG_MCP_TRANSPORT\": \"stdio\"\n      }\n    }\n  }\n}\n```\n\n### MCP Error Handling\n\n`ToolError` variants: `FileNotFound`, `UnsupportedFormat`, `ExtractionFailed`, `OnnxRuntimeMissing`, `TesseractMissing`, `Timeout`. Each maps to an MCP `ToolResultError` with descriptive code and message.\n\n## Environment Configuration\n\nSee `.env.example` for all configurable variables. Key categories:\n\n- **Server**: `XBERG_HOST`, `XBERG_PORT`\n- **Size limits**: `XBERG_MAX_REQUEST_BODY_BYTES` (default 100MB), `XBERG_MAX_MULTIPART_FIELD_BYTES`\n- **Features**: `XBERG_ENABLE_OCR`, `XBERG_ENABLE_EMBEDDINGS`, `XBERG_ENABLE_KEYWORDS`\n- **Cache**: `XBERG_CACHE_ENABLED`, `XBERG_CACHE_SIZE`\n- **CORS**: `CORS_ALLOWED_ORIGINS` (comma-separated)\n- **MCP**: `XBERG_MCP_HOST`, `XBERG_MCP_PORT`, `XBERG_MCP_TRANSPORT` (stdio\u002Fhttp)\n- **Logging**: `RUST_LOG=xberg=info,tower_http=debug`\n\n## Critical Rules\n\n### REST API Rules\n\n1. **Always validate multipart file uploads** - Check MIME type, size, magic bytes\n2. **Timeout long-running extractions** - Set per-handler timeout (5 min default)\n3. **Stream large files** - Never buffer entire multi-GB file in memory\n4. **Cache aggressively** - Identical files should return from cache in \u003C1ms\n5. **Parallel extraction is CPU-bound** - Limit workers to CPU count + 1\n6. **Error responses must be actionable** - Include error code and remediation suggestion\n7. **Health checks must verify features** - Report missing dependencies (ONNX, Tesseract)\n8. **Size limits are configurable** - Allow override via env var for large deployments\n9. **CORS is permissive by default** - Restrict in production via env var\n10. **Logging all requests** - Track extraction metrics for observability\n\n### MCP Rules\n\n1. **All tools must have timeout** - Prevent hanging on large files (default 5 min)\n2. **Error responses must be detailed** - Include suggestions for missing dependencies\n3. **Feature gates must be checked** - Return helpful message if feature unavailable (embeddings, OCR)\n4. **Resources should be static** - Don't query external services in resource handlers\n5. **Prompts guide agents** - Provide clear examples and best practices\n6. **Batch tools must support cancellation** - Allow agent to stop long-running batch operations\n7. **Logging all tool calls** - Track usage for analytics and debugging\n\n## Related Skills\n\n- **extraction-pipeline-patterns** - Core extraction called by handlers and MCP tools\n- **chunking-embeddings** - Optional chunking\u002Fembedding parameters in extraction\n- **ocr-backend-management** - OCR engine selection and image preprocessing\n",{"name":5,"priority":24,"description":7},"critical","main","613e6deee7a2f31f95bac8beec31dd5171193c8d",".ai-rulez\u002Fskills\u002Fapi-server-mcp",0,"https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine","https:\u002F\u002Fhacienda-engine-hacienda-studio.vercel.app","https:\u002F\u002Fgithub.com\u002Fjamon8888","hacienda.diy","Maker","2026-11-15 05:01:55",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add jamon8888\u002Fhacienda-engine --skill api-server-mcp",{"tool":41,"command":42},"bunx","bunx skills add jamon8888\u002Fhacienda-engine --skill api-server-mcp",{"tool":44,"command":45},"pnpm","pnpm dlx skills add jamon8888\u002Fhacienda-engine --skill api-server-mcp","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Fapi-server-mcp\u002FSKILL.md",true,[],[50,57,64,71,78,85],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M79RTKG5HB5PQ251W7J9C4S6","impeccable","Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fimpeccable\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M79RTFDA4MAD9PPF3N15B68Y","wasm-constraints","WASM build constraints for the crates\u002Fxberg-wasm crate — the wasm-target feature set, no-tokio sync-only internal APIs, the mandatory SyncExtractor trait for WASM-compatible extractors, the 2 MB HTML size limit, size-optimized build config (opt-level=\"z\"), and the async-wrapper\u002Fsync-internal API pattern. Load when building for wasm32, adding or modifying a WASM-compatible extractor, or debugging WASM build\u002Fruntime failures.","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Fwasm-constraints\u002FSKILL.md",[],[],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M79RTC4SV7NENKJXK2PSHK7H","release-versioning","How xberg versions are synced and released — Cargo.toml is the single source of truth, `task version:sync` propagates it to alef-managed binding manifests AND the integrations under integrations\u002F, which are versioned and published in lockstep with core (including -rc.N). Load before bumping a version, editing the version-sync task, or touching an integration's version\u002Fxberg dependency.","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Frelease-versioning\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M79RT94KWTS5R5T61S71H1AD","plugin-architecture-patterns","Plugin architecture, registration, and trait patterns","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Fplugin-architecture-patterns\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M79RT3X076QPZYK10QF6Q22X","mime-detection-routing","MIME type detection and extractor routing in core\u002Fmime.rs and core\u002Fextractor\u002Fbytes.rs — the extension→EXT_TO_MIME→validate→registry→extractor flow, key detection functions, the 118+ case-insensitive extension map, priority-based registry selection, wildcard MIME families, and how to add a new MIME type. Load when adding a format, wiring an extractor to a MIME type, or debugging why a file routes to the wrong (or no) extractor.","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Fmime-detection-routing\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M79RSZYWZ6RSAM31N2Y0TXMN","format-specific-extraction","Format-specific document extraction workflows","https:\u002F\u002Fgithub.com\u002Fjamon8888\u002Fhacienda-engine\u002Fblob\u002Fmain\u002F.ai-rulez\u002Fskills\u002Fformat-specific-extraction\u002FSKILL.md",[],[]]