[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CS5G1B64M8KEP42WR550FV":3,"skill-same:01M7CS5G1B64M8KEP42WR550FV":57},{"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":34,"commitSha":35,"skillDirectory":36,"forksCount":37,"repositoryHtmlUrl":38,"repositoryHomepageUrl":39,"ownerHtmlUrl":40,"ownerBlogUrl":6,"ownerDescription":41,"updatedAt":42,"install":43,"resources":56},"01M7CS5G1B64M8KEP42WR550FV","glmocr",null,"Trigger when: (1) User wants to extract text, tables, formulas, or structured data from images\u002FPDFs\u002Fscanned documents, (2) User mentions \"OCR\", \"文字识别\", \"文档解析\", (3) User has a document (screenshot, scanned page, invoice, paper, whiteboard photo) and needs its content in structured form, (4) User asks to parse, digitize, or extract content from a visual document.\n\nInvokes the GLM-OCR SDK (pip install glmocr) to parse documents via Zhipu's cloud API. No GPU required. Returns structured JSON (regions with labels + bounding boxes) and Markdown. Agent can operate entirely via CLI — no YAML files needed.\n\nNOT for: real-time camera feeds, audio transcription, or non-document images (photos, illustrations).","YYC-Cube\u002FYYC3-Knowledge-Base","YYC-Cube","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F235006543?v=4","01M79J2PH4SWD2X0825X6PN748","yyc3","01M79J2PHVK3F6ZBYT8K2SCZCW",1,"MIT","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-skills-v4\u002Fglmocr-sdk\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: glmocr\ndescription: |\n  Trigger when: (1) User wants to extract text, tables, formulas, or structured data from images\u002FPDFs\u002Fscanned documents, (2) User mentions \"OCR\", \"文字识别\", \"文档解析\", (3) User has a document (screenshot, scanned page, invoice, paper, whiteboard photo) and needs its content in structured form, (4) User asks to parse, digitize, or extract content from a visual document.\n\n  Invokes the GLM-OCR SDK (pip install glmocr) to parse documents via Zhipu's cloud API. No GPU required. Returns structured JSON (regions with labels + bounding boxes) and Markdown. Agent can operate entirely via CLI — no YAML files needed.\n\n  NOT for: real-time camera feeds, audio transcription, or non-document images (photos, illustrations).\nmetadata:\n  openclaw:\n    requires:\n      env:\n        - ZHIPU_API_KEY\n      bins:\n        - python\n    primaryEnv: ZHIPU_API_KEY\n    emoji: \"📄\"\n    homepage: https:\u002F\u002Fgithub.com\u002Fzai-org\u002FGLM-OCR\u002Ftree\u002Fmain\u002Fskills\u002Fsdk\n---\n\n# OpenClaw Skill: glmocr\n\nParses documents (images, PDFs, scans) via the GLM-OCR SDK.\n\n> **📌 On-demand**: This skill requires only `ZHIPU_API_KEY` in the environment. No YAML config files or GPU needed.\n\n## ⚡ Quick Start\n\n```bash\n# Install\npip install glmocr\n\n# Set API key (once)\nexport ZHIPU_API_KEY=sk-xxx\n# or add to .env file in working directory:\necho \"ZHIPU_API_KEY=sk-xxx\" >> .env\n```\n\n```python\n# One-liner\nimport glmocr\nresult = glmocr.parse(\"document.pdf\")\nprint(result.markdown_result)\nprint(result.to_dict())\n```\n\n```bash\n# CLI — pass API key directly (no env setup needed)\nglmocr parse image.png --api-key sk-xxx\n\n# Or load from a specific .env file\nglmocr parse image.png --env-file \u002Fpath\u002Fto\u002F.env\n\n# Or rely on env var \u002F auto-discovered .env (set once, then omit)\nglmocr parse image.png\nglmocr parse .\u002Fscans\u002F --output .\u002Foutput\u002F --stdout\n```\n\n---\n\n## Configuration Priority\n\n```\nConstructor kwargs  >  os.environ  >  .env file  >  config.yaml  >  built-in defaults\n```\n\nAgents override everything via constructor kwargs or env vars — no YAML editing needed.\n\n### Key Environment Variables\n\n| Variable               | Description                            | Example     |\n| ---------------------- | -------------------------------------- | ----------- |\n| `ZHIPU_API_KEY`        | API key (required for MaaS)            | `sk-abc123` |\n| `GLMOCR_MODEL`         | Model name                             | `glm-ocr`   |\n| `GLMOCR_TIMEOUT`       | Request timeout (seconds)              | `600`       |\n| `GLMOCR_ENABLE_LAYOUT` | Layout detection on\u002Foff                | `true`      |\n| `GLMOCR_LOG_LEVEL`     | `DEBUG` \u002F `INFO` \u002F `WARNING` \u002F `ERROR` | `INFO`      |\n\n---\n\n## Python API\n\n### Convenience function (single call)\n\n```python\nimport glmocr\n\n# Single file → PipelineResult\nresult = glmocr.parse(\"invoice.png\")\n\n# Multiple files → list[PipelineResult]\nresults = glmocr.parse([\"page1.png\", \"page2.png\", \"report.pdf\"])\n```\n\n### Class-based (multiple calls \u002F resource reuse)\n\n```python\nfrom glmocr import GlmOcr\n\nparser = GlmOcr(api_key=\"sk-xxx\")   # mode auto-set to \"maas\"\nparser = GlmOcr(mode=\"maas\")        # reads ZHIPU_API_KEY from env\n\n# Always use as context manager or call .close()\nwith GlmOcr(api_key=\"sk-xxx\") as parser:\n    result = parser.parse(\"document.png\")\n    print(result.markdown_result)\n\nparser.close()   # if not using `with`\n```\n\n### Constructor Parameters\n\n| Parameter       | Type   | Description                                     |\n| --------------- | ------ | ----------------------------------------------- |\n| `api_key`       | `str`  | API key. Providing this auto-enables MaaS mode. |\n| `api_url`       | `str`  | Override MaaS endpoint URL                      |\n| `model`         | `str`  | Model name override                             |\n| `timeout`       | `int`  | Request timeout in seconds (default: 600)       |\n| `enable_layout` | `bool` | Enable layout detection                         |\n| `log_level`     | `str`  | Logging level                                   |\n\n---\n\n## Working with `PipelineResult`\n\n### Fields\n\n```python\nresult.markdown_result    # str — full document as Markdown\nresult.json_result        # list[list[dict]] — structured regions per page\nresult.original_images    # list[str] — absolute paths of input images\n```\n\n### `json_result` structure\n\nList of pages → list of regions per page:\n\n```json\n[\n  [\n    {\n      \"index\": 0,\n      \"label\": \"title\",\n      \"content\": \"Annual Report 2024\",\n      \"bbox_2d\": [100, 50, 900, 120]\n    },\n    {\n      \"index\": 1,\n      \"label\": \"table\",\n      \"content\": \"| Q1 | Q2 |\\n|---|---|\\n| 120 | 145 |\",\n      \"bbox_2d\": [100, 140, 900, 400]\n    }\n  ]\n]\n```\n\n**Bounding boxes** (`bbox_2d`): `[x1, y1, x2, y2]` normalised to **0–1000** scale.\n\n**Region labels**: `title`, `text`, `table`, `figure`, `formula`, `header`, `footer`, `page_number`, `reference`, `seal`\n\n### Serialization\n\n```python\n# Dict (JSON-serializable, for passing to other tools)\nd = result.to_dict()\n# Keys: json_result, markdown_result, original_images, usage (MaaS), data_info (MaaS)\n\n# JSON string\njson_str = result.to_json()                 # pretty-printed, ensure_ascii=False\njson_str = result.to_json(indent=None)      # compact single line\n\n# Save to disk: writes \u003Cstem>\u002F\u003Cstem>.json + \u003Cstem>\u002F\u003Cstem>.md + layout_vis\u002F\nresult.save(output_dir=\".\u002Foutput\")\nresult.save(output_dir=\".\u002Foutput\", save_layout_visualization=False)\n```\n\n### Error Handling\n\nThe SDK **does not raise** on MaaS errors — check `to_dict()` for an `\"error\"` key:\n\n```python\nresult = parser.parse(\"image.png\")\nd = result.to_dict()\nif \"error\" in d:\n    # Handle failure\n    print(\"OCR failed:\", d[\"error\"])\nelse:\n    print(d[\"markdown_result\"])\n```\n\n---\n\n## CLI Reference\n\n> **Agent-preferred interface**: use the CLI for most operations. Set `ZHIPU_API_KEY` in env once, then invoke as needed.\n\n**Supported input formats**: `.jpg`, `.jpeg`, `.png`, `.bmp`, `.gif`, `.webp`, `.pdf`\n\n### Basic usage\n\n```bash\n# Parse a single file → saves to .\u002Foutput\u002F\u003Cstem>\u002F\n# MaaS mode is the default; ZHIPU_API_KEY must be set (or use --api-key)\nglmocr parse image.png\n\n# Pass API key directly without any env setup\nglmocr parse image.png --api-key sk-xxx\n\n# Parse a directory → saves each file to .\u002Foutput\u002F\u003Cstem>\u002F\nglmocr parse .\u002Fscans\u002F\n\n# Use self-hosted vLLM\u002FSGLang instead of cloud\nglmocr parse image.png --mode selfhosted\n\n# Specify output directory\nglmocr parse image.png --output .\u002Fresults\u002F\n```\n\n### Read results in the terminal (agent-friendly)\n\n```bash\n# Print Markdown + JSON to stdout (and still save to disk)\nglmocr parse image.png --stdout\n\n# Print to stdout ONLY — do not write any files\nglmocr parse image.png --stdout --no-save\n\n# JSON only (no Markdown output)\nglmocr parse image.png --stdout --json-only\n\n# Pipe JSON into jq for structured extraction\nglmocr parse image.png --stdout --json-only --no-save | jq '.[0] | map(select(.label==\"table\"))'\n```\n\n### Save control\n\n```bash\n# Skip layout visualization images (faster, smaller output)\nglmocr parse image.png --no-layout-vis\n\n# Parse and save only JSON + Markdown, skip layout vis\nglmocr parse image.png --no-layout-vis --output .\u002Fresults\u002F\n```\n\n### Batch processing\n\n```bash\n# All images in a folder\nglmocr parse .\u002Finvoice_scans\u002F --output .\u002Fparsed\u002F --no-layout-vis\n\n# With progress visible in logs\nglmocr parse .\u002Fdocs\u002F --output .\u002Fparsed\u002F --log-level INFO\n```\n\n### Debugging\n\n```bash\nglmocr parse image.png --log-level DEBUG\n```\n\n### Full flag reference\n\n| Flag              | Default    | Description                                           |\n| ----------------- | ---------- | ----------------------------------------------------- |\n| `--api-key \u002F -k`  | env var    | API key for MaaS mode (overrides `ZHIPU_API_KEY`)     |\n| `--mode`          | `maas`     | `maas` (cloud, default) or `selfhosted` (local GPU)   |\n| `--env-file`      | auto       | Path to `.env` file (default: auto-discover from cwd) |\n| `--output \u002F -o`   | `.\u002Foutput` | Output directory                                      |\n| `--stdout`        | off        | Print JSON + Markdown to stdout                       |\n| `--no-save`       | off        | Skip writing files (use with `--stdout`)              |\n| `--json-only`     | off        | stdout JSON only, no Markdown                         |\n| `--no-layout-vis` | off        | Skip layout visualization images                      |\n| `--config \u002F -c`   | none       | Path to YAML config override                          |\n| `--log-level`     | `INFO`     | `DEBUG` \u002F `INFO` \u002F `WARNING` \u002F `ERROR`                |\n\n---\n\n## Typical Agent Workflow\n\n```\nreceive document path \u002F URL\n       │\n       ▼\nglmocr.parse(path)            ← single call, handles PDF\u002Fimage\n       │\n       ▼\nresult.to_dict()              ← safe to pass as tool output\n       │\n       ├── markdown_result    → hand to LLM for reading \u002F summarization\n       └── json_result        → structured extraction (tables, formulas, regions by label)\n```\n\n### Filter by label\n\n```python\nresult = glmocr.parse(\"report.png\")\nregions = result.json_result[0]  # first page\n\ntables = [r for r in regions if r[\"label\"] == \"table\"]\nformulas = [r for r in regions if r[\"label\"] == \"formula\"]\nbody_text = [r for r in regions if r[\"label\"] == \"text\"]\n```\n\n### Multi-page PDF → iterate pages\n\n```python\nwith GlmOcr(api_key=\"sk-xxx\") as parser:\n    result = parser.parse(\"document.pdf\")   # all pages in one PipelineResult\n    for page_idx, page_regions in enumerate(result.json_result):\n        print(f\"Page {page_idx + 1}: {len(page_regions)} regions\")\n        for region in page_regions:\n            print(f\"  [{region['label']}] {region['content'][:60]}\")\n```\n\n### Programmatic config (no env vars)\n\n```python\nfrom glmocr.config import GlmOcrConfig\n\ncfg = GlmOcrConfig.from_env(\n    api_key=\"sk-xxx\",\n    mode=\"maas\",\n    timeout=600,\n    log_level=\"DEBUG\",\n)\n```\n\n---\n\n## Output Directory Layout\n\nAfter `result.save(output_dir)`:\n\n```\noutput_dir\u002F\n  \u003Cimage_stem>\u002F\n    \u003Cimage_stem>.json         ← structured regions\n    \u003Cimage_stem>.md           ← full Markdown (with cropped figure images)\n    imgs\u002F                     ← cropped figures referenced in Markdown\n    layout_vis\u002F               ← layout detection overlay images (if enabled)\n      \u003Cimage_stem>.jpg\n```\n\n---\n\n## Common Pitfalls\n\n- **`ZHIPU_API_KEY` not set**: SDK defaults to MaaS mode. Without a key, `parse()` will fail with a clear error message and quick-fix instructions. Set via `export ZHIPU_API_KEY=sk-xxx`, add to a `.env` file, or pass `--api-key sk-xxx` to the CLI.\n- **Large PDFs**: Default timeout is 600s. For very long documents increase with `timeout=1200`.\n- **`result.json_result` is a string**: Happens when the model returns malformed JSON. The SDK preserves the raw string — parse or log it manually.\n","# OpenClaw Skill: glmocr\n\nParses documents (images, PDFs, scans) via the GLM-OCR SDK.\n\n> **📌 On-demand**: This skill requires only `ZHIPU_API_KEY` in the environment. No YAML config files or GPU needed.\n\n## ⚡ Quick Start\n\n```bash\n# Install\npip install glmocr\n\n# Set API key (once)\nexport ZHIPU_API_KEY=sk-xxx\n# or add to .env file in working directory:\necho \"ZHIPU_API_KEY=sk-xxx\" >> .env\n```\n\n```python\n# One-liner\nimport glmocr\nresult = glmocr.parse(\"document.pdf\")\nprint(result.markdown_result)\nprint(result.to_dict())\n```\n\n```bash\n# CLI — pass API key directly (no env setup needed)\nglmocr parse image.png --api-key sk-xxx\n\n# Or load from a specific .env file\nglmocr parse image.png --env-file \u002Fpath\u002Fto\u002F.env\n\n# Or rely on env var \u002F auto-discovered .env (set once, then omit)\nglmocr parse image.png\nglmocr parse .\u002Fscans\u002F --output .\u002Foutput\u002F --stdout\n```\n\n---\n\n## Configuration Priority\n\n```\nConstructor kwargs  >  os.environ  >  .env file  >  config.yaml  >  built-in defaults\n```\n\nAgents override everything via constructor kwargs or env vars — no YAML editing needed.\n\n### Key Environment Variables\n\n| Variable               | Description                            | Example     |\n| ---------------------- | -------------------------------------- | ----------- |\n| `ZHIPU_API_KEY`        | API key (required for MaaS)            | `sk-abc123` |\n| `GLMOCR_MODEL`         | Model name                             | `glm-ocr`   |\n| `GLMOCR_TIMEOUT`       | Request timeout (seconds)              | `600`       |\n| `GLMOCR_ENABLE_LAYOUT` | Layout detection on\u002Foff                | `true`      |\n| `GLMOCR_LOG_LEVEL`     | `DEBUG` \u002F `INFO` \u002F `WARNING` \u002F `ERROR` | `INFO`      |\n\n---\n\n## Python API\n\n### Convenience function (single call)\n\n```python\nimport glmocr\n\n# Single file → PipelineResult\nresult = glmocr.parse(\"invoice.png\")\n\n# Multiple files → list[PipelineResult]\nresults = glmocr.parse([\"page1.png\", \"page2.png\", \"report.pdf\"])\n```\n\n### Class-based (multiple calls \u002F resource reuse)\n\n```python\nfrom glmocr import GlmOcr\n\nparser = GlmOcr(api_key=\"sk-xxx\")   # mode auto-set to \"maas\"\nparser = GlmOcr(mode=\"maas\")        # reads ZHIPU_API_KEY from env\n\n# Always use as context manager or call .close()\nwith GlmOcr(api_key=\"sk-xxx\") as parser:\n    result = parser.parse(\"document.png\")\n    print(result.markdown_result)\n\nparser.close()   # if not using `with`\n```\n\n### Constructor Parameters\n\n| Parameter       | Type   | Description                                     |\n| --------------- | ------ | ----------------------------------------------- |\n| `api_key`       | `str`  | API key. Providing this auto-enables MaaS mode. |\n| `api_url`       | `str`  | Override MaaS endpoint URL                      |\n| `model`         | `str`  | Model name override                             |\n| `timeout`       | `int`  | Request timeout in seconds (default: 600)       |\n| `enable_layout` | `bool` | Enable layout detection                         |\n| `log_level`     | `str`  | Logging level                                   |\n\n---\n\n## Working with `PipelineResult`\n\n### Fields\n\n```python\nresult.markdown_result    # str — full document as Markdown\nresult.json_result        # list[list[dict]] — structured regions per page\nresult.original_images    # list[str] — absolute paths of input images\n```\n\n### `json_result` structure\n\nList of pages → list of regions per page:\n\n```json\n[\n  [\n    {\n      \"index\": 0,\n      \"label\": \"title\",\n      \"content\": \"Annual Report 2024\",\n      \"bbox_2d\": [100, 50, 900, 120]\n    },\n    {\n      \"index\": 1,\n      \"label\": \"table\",\n      \"content\": \"| Q1 | Q2 |\\n|---|---|\\n| 120 | 145 |\",\n      \"bbox_2d\": [100, 140, 900, 400]\n    }\n  ]\n]\n```\n\n**Bounding boxes** (`bbox_2d`): `[x1, y1, x2, y2]` normalised to **0–1000** scale.\n\n**Region labels**: `title`, `text`, `table`, `figure`, `formula`, `header`, `footer`, `page_number`, `reference`, `seal`\n\n### Serialization\n\n```python\n# Dict (JSON-serializable, for passing to other tools)\nd = result.to_dict()\n# Keys: json_result, markdown_result, original_images, usage (MaaS), data_info (MaaS)\n\n# JSON string\njson_str = result.to_json()                 # pretty-printed, ensure_ascii=False\njson_str = result.to_json(indent=None)      # compact single line\n\n# Save to disk: writes \u003Cstem>\u002F\u003Cstem>.json + \u003Cstem>\u002F\u003Cstem>.md + layout_vis\u002F\nresult.save(output_dir=\".\u002Foutput\")\nresult.save(output_dir=\".\u002Foutput\", save_layout_visualization=False)\n```\n\n### Error Handling\n\nThe SDK **does not raise** on MaaS errors — check `to_dict()` for an `\"error\"` key:\n\n```python\nresult = parser.parse(\"image.png\")\nd = result.to_dict()\nif \"error\" in d:\n    # Handle failure\n    print(\"OCR failed:\", d[\"error\"])\nelse:\n    print(d[\"markdown_result\"])\n```\n\n---\n\n## CLI Reference\n\n> **Agent-preferred interface**: use the CLI for most operations. Set `ZHIPU_API_KEY` in env once, then invoke as needed.\n\n**Supported input formats**: `.jpg`, `.jpeg`, `.png`, `.bmp`, `.gif`, `.webp`, `.pdf`\n\n### Basic usage\n\n```bash\n# Parse a single file → saves to .\u002Foutput\u002F\u003Cstem>\u002F\n# MaaS mode is the default; ZHIPU_API_KEY must be set (or use --api-key)\nglmocr parse image.png\n\n# Pass API key directly without any env setup\nglmocr parse image.png --api-key sk-xxx\n\n# Parse a directory → saves each file to .\u002Foutput\u002F\u003Cstem>\u002F\nglmocr parse .\u002Fscans\u002F\n\n# Use self-hosted vLLM\u002FSGLang instead of cloud\nglmocr parse image.png --mode selfhosted\n\n# Specify output directory\nglmocr parse image.png --output .\u002Fresults\u002F\n```\n\n### Read results in the terminal (agent-friendly)\n\n```bash\n# Print Markdown + JSON to stdout (and still save to disk)\nglmocr parse image.png --stdout\n\n# Print to stdout ONLY — do not write any files\nglmocr parse image.png --stdout --no-save\n\n# JSON only (no Markdown output)\nglmocr parse image.png --stdout --json-only\n\n# Pipe JSON into jq for structured extraction\nglmocr parse image.png --stdout --json-only --no-save | jq '.[0] | map(select(.label==\"table\"))'\n```\n\n### Save control\n\n```bash\n# Skip layout visualization images (faster, smaller output)\nglmocr parse image.png --no-layout-vis\n\n# Parse and save only JSON + Markdown, skip layout vis\nglmocr parse image.png --no-layout-vis --output .\u002Fresults\u002F\n```\n\n### Batch processing\n\n```bash\n# All images in a folder\nglmocr parse .\u002Finvoice_scans\u002F --output .\u002Fparsed\u002F --no-layout-vis\n\n# With progress visible in logs\nglmocr parse .\u002Fdocs\u002F --output .\u002Fparsed\u002F --log-level INFO\n```\n\n### Debugging\n\n```bash\nglmocr parse image.png --log-level DEBUG\n```\n\n### Full flag reference\n\n| Flag              | Default    | Description                                           |\n| ----------------- | ---------- | ----------------------------------------------------- |\n| `--api-key \u002F -k`  | env var    | API key for MaaS mode (overrides `ZHIPU_API_KEY`)     |\n| `--mode`          | `maas`     | `maas` (cloud, default) or `selfhosted` (local GPU)   |\n| `--env-file`      | auto       | Path to `.env` file (default: auto-discover from cwd) |\n| `--output \u002F -o`   | `.\u002Foutput` | Output directory                                      |\n| `--stdout`        | off        | Print JSON + Markdown to stdout                       |\n| `--no-save`       | off        | Skip writing files (use with `--stdout`)              |\n| `--json-only`     | off        | stdout JSON only, no Markdown                         |\n| `--no-layout-vis` | off        | Skip layout visualization images                      |\n| `--config \u002F -c`   | none       | Path to YAML config override                          |\n| `--log-level`     | `INFO`     | `DEBUG` \u002F `INFO` \u002F `WARNING` \u002F `ERROR`                |\n\n---\n\n## Typical Agent Workflow\n\n```\nreceive document path \u002F URL\n       │\n       ▼\nglmocr.parse(path)            ← single call, handles PDF\u002Fimage\n       │\n       ▼\nresult.to_dict()              ← safe to pass as tool output\n       │\n       ├── markdown_result    → hand to LLM for reading \u002F summarization\n       └── json_result        → structured extraction (tables, formulas, regions by label)\n```\n\n### Filter by label\n\n```python\nresult = glmocr.parse(\"report.png\")\nregions = result.json_result[0]  # first page\n\ntables = [r for r in regions if r[\"label\"] == \"table\"]\nformulas = [r for r in regions if r[\"label\"] == \"formula\"]\nbody_text = [r for r in regions if r[\"label\"] == \"text\"]\n```\n\n### Multi-page PDF → iterate pages\n\n```python\nwith GlmOcr(api_key=\"sk-xxx\") as parser:\n    result = parser.parse(\"document.pdf\")   # all pages in one PipelineResult\n    for page_idx, page_regions in enumerate(result.json_result):\n        print(f\"Page {page_idx + 1}: {len(page_regions)} regions\")\n        for region in page_regions:\n            print(f\"  [{region['label']}] {region['content'][:60]}\")\n```\n\n### Programmatic config (no env vars)\n\n```python\nfrom glmocr.config import GlmOcrConfig\n\ncfg = GlmOcrConfig.from_env(\n    api_key=\"sk-xxx\",\n    mode=\"maas\",\n    timeout=600,\n    log_level=\"DEBUG\",\n)\n```\n\n---\n\n## Output Directory Layout\n\nAfter `result.save(output_dir)`:\n\n```\noutput_dir\u002F\n  \u003Cimage_stem>\u002F\n    \u003Cimage_stem>.json         ← structured regions\n    \u003Cimage_stem>.md           ← full Markdown (with cropped figure images)\n    imgs\u002F                     ← cropped figures referenced in Markdown\n    layout_vis\u002F               ← layout detection overlay images (if enabled)\n      \u003Cimage_stem>.jpg\n```\n\n---\n\n## Common Pitfalls\n\n- **`ZHIPU_API_KEY` not set**: SDK defaults to MaaS mode. Without a key, `parse()` will fail with a clear error message and quick-fix instructions. Set via `export ZHIPU_API_KEY=sk-xxx`, add to a `.env` file, or pass `--api-key sk-xxx` to the CLI.\n- **Large PDFs**: Default timeout is 600s. For very long documents increase with `timeout=1200`.\n- **`result.json_result` is a string**: Happens when the model returns malformed JSON. The SDK preserves the raw string — parse or log it manually.\n",{"name":5,"metadata":24,"description":33},{"openclaw":25},{"emoji":26,"homepage":27,"requires":28,"primaryEnv":30},"📄","https:\u002F\u002Fgithub.com\u002Fzai-org\u002FGLM-OCR\u002Ftree\u002Fmain\u002Fskills\u002Fsdk",{"env":29,"bins":31},[30],"ZHIPU_API_KEY",[32],"python","Trigger when: (1) User wants to extract text, tables, formulas, or structured data from images\u002FPDFs\u002Fscanned documents, (2) User mentions \"OCR\", \"文字识别\", \"文档解析\", (3) User has a document (screenshot, scanned page, invoice, paper, whiteboard photo) and needs its content in structured form, (4) User asks to parse, digitize, or extract content from a visual document.\n\nInvokes the GLM-OCR SDK (pip install glmocr) to parse documents via Zhipu's cloud API. No GPU required. Returns structured JSON (regions with labels + bounding boxes) and Markdown. Agent can operate entirely via CLI — no YAML files needed.\n\nNOT for: real-time camera feeds, audio transcription, or non-document images (photos, illustrations).\n","main","87afe6ce96220accf50a345dea46784c8c9f210e","YYC3-Skills-技能包\u002FYYC3-Skills-技能包-skills-v4\u002Fglmocr-sdk",0,"https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base","https:\u002F\u002Fkb.yyc3.top\u002F","https:\u002F\u002Fgithub.com\u002FYYC-Cube","万象归元于云枢 | 深栈智启新纪元","2026-11-10 10:37:32",{"source":8,"skillName":5,"commands":44,"prompt":54,"zipAvailable":55},[45,48,51],{"tool":46,"command":47},"npx","npx skills add YYC-Cube\u002FYYC3-Knowledge-Base --skill glmocr",{"tool":49,"command":50},"bunx","bunx skills add YYC-Cube\u002FYYC3-Knowledge-Base --skill glmocr",{"tool":52,"command":53},"pnpm","pnpm dlx skills add YYC-Cube\u002FYYC3-Knowledge-Base --skill glmocr","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-skills-v4\u002Fglmocr-sdk\u002FSKILL.md",true,[],[58,65,72,79,86,93],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M7CTDWHERCXPQCP0K03A86AS","vercel-sandbox","Run agent-browser + Chrome inside Vercel Sandbox microVMs for browser automation from any Vercel-deployed app. Use when the user needs browser automation in a Vercel app (Next.js, SvelteKit, Nuxt, Remix, Astro, etc.), wants to run headless Chrome without binary size limits, needs persistent browser sessions across commands, or wants ephemeral isolated browser environments. Triggers include \"Vercel Sandbox browser\", \"microVM Chrome\", \"agent-browser in sandbox\", \"browser automation on Vercel\", or any task requiring Chrome in a Vercel Sandbox.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fvercel-sandbox\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M7CTD8D2BNJY54S65M47HEJA","slack","Interact with Slack workspaces using browser automation. Use when the user needs to check unread channels, navigate Slack, send messages, extract data, find information, search conversations, or automate any Slack task. Triggers include \"check my Slack\", \"what channels have unreads\", \"send a message to\", \"search Slack for\", \"extract from Slack\", \"find who said\", or any task requiring programmatic Slack interaction.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fslack\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M7CTCSSB8ZP3Q58C3YZDZK1W","electron","Automate Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify, etc.) using agent-browser via Chrome DevTools Protocol. Use when the user needs to interact with an Electron app, automate a desktop app, connect to a running app, control a native app, or test an Electron application. Triggers include \"automate Slack app\", \"control VS Code\", \"interact with Discord app\", \"test this Electron app\", \"connect to desktop app\", or any task requiring automation of a native Electron application.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Felectron\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M7CTCBJY76688V4DCZGWFHGQ","dogfood","Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to \"dogfood\", \"QA\", \"exploratory test\", \"find issues\", \"bug hunt\", \"test this app\u002Fsite\u002Fplatform\", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fdogfood\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M7CTC4H0FFGVEAQR9PA4GA1A","agent-browser","Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to \"open a website\", \"fill out a form\", \"click a button\", \"take a screenshot\", \"scrape data from a page\", \"test this web app\", \"login to a site\", \"automate browser actions\", or any task requiring programmatic web interaction.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fagent-browser\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":19},"01M7CTBKVFYBG78FDSABSDJTTQ","shadcn","Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI, including chat interfaces. Provides project context, component docs, and usage examples. Applies when working with shadcn\u002Fui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for \"shadcn init\", \"create an app with --preset\", \"switch to --preset\", and for building a design system or applying a DESIGN.md or brand spec to shadcn\u002Fui.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-UI-界面组件\u002FYYC3-UI-界面组件-ui\u002Fskills\u002Fshadcn\u002FSKILL.md",[],[]]