[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MJ168NHETSR3H0VF3N56DF":3,"skill-same:01M6MJ168NHETSR3H0VF3N56DF":69},{"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":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":48,"repositoryHtmlUrl":49,"repositoryHomepageUrl":50,"ownerHtmlUrl":51,"ownerBlogUrl":52,"ownerDescription":53,"updatedAt":54,"install":55,"resources":68},"01M6MJ168NHETSR3H0VF3N56DF","extracting-keywords",null,"Use when extracting keywords (YAKE\u002FRAKE) from documents — and, secondarily, when detecting document language or generating embeddings for RAG and search. Covers the keyword config (and its feature gating), `--detect-language`, and the standalone `embed` command with real flags.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fkreuzberg-dev\u002Fplugins\u002Fplugins\u002Fkreuzberg\u002Fskills\u002Fextracting-keywords\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"extracting keywords","extracting","detect","when","gating","for","standalone","yake","detecting","with","embed","config","feature","flags","document","use","command","embeddings","search","rag","documents","2026-11-10 10:12:30",false,"---\nname: extracting-keywords\ndescription: Use when extracting keywords (YAKE\u002FRAKE) from documents — and, secondarily, when detecting document language or generating embeddings for RAG and search. Covers the keyword config (and its feature gating), `--detect-language`, and the standalone `embed` command with real flags.\n---\n\n# Extracting keywords, language, and embeddings\n\nUse this for the enrichment surface around extraction: statistical keyword\nextraction, language detection, and vector embeddings. Keywords and\nlanguage detection ride along with extraction and land on the result;\nembeddings are produced by a dedicated `embed` command.\n\n## Keywords (YAKE \u002F RAKE)\n\nKeyword extraction is configured via the `[keywords]` config block (or\ninline JSON) — there is no single `--keywords` CLI flag. When enabled,\nextracted keywords appear on `result.keywords`. Two algorithms are\navailable:\n\n- **YAKE** (`\"yake\"`) — statistical, unsupervised single-document\n  extraction. Good general default.\n- **RAKE** (`\"rake\"`) — co-occurrence \u002F phrase-based. Favors multi-word\n  key phrases.\n\n> Feature-gated: keyword extraction requires the CLI to be built with the\n> `keywords-yake` and\u002For `keywords-rake` Cargo features (both are in the\n> default\u002F`full` build). If the CLI was built without them, the `[keywords]`\n> config block is silently ignored — `result.keywords` simply stays empty\n> rather than erroring. The `\"yake\"` algorithm needs `keywords-yake`; `\"rake\"`\n> needs `keywords-rake`.\n\nEnable via inline JSON on the CLI:\n\n```bash\nkreuzberg extract paper.pdf --format json \\\n  --config-json '{\"keywords\":{\"algorithm\":\"yake\",\"max_keywords\":15,\"language\":\"en\"}}' \\\n  | jq '.keywords'\n```\n\nOr in a config file:\n\n```toml\n[keywords]\nalgorithm = \"rake\"       # \"yake\" or \"rake\"\nmax_keywords = 10        # default 10\nmin_score = 0.0          # filter below this score (ranges differ per algorithm)\nngram_range = [1, 3]     # unigrams..trigrams (default)\nlanguage = \"en\"          # stopword language; omit to skip stopword filtering\n```\n\n```bash\nkreuzberg extract report.pdf --config kreuzberg.toml --format json | jq '.keywords'\n```\n\nField notes:\n\n- `max_keywords` caps how many keywords are returned (default 10).\n- `min_score` filters low-scoring keywords; note YAKE scores are\n  *lower-is-better* while RAKE scores are *higher-is-better*, so a single\n  threshold behaves differently per algorithm.\n- `ngram_range` is `[min, max]`: `[1,1]` unigrams only, `[1,2]` adds\n  bigrams, `[1,3]` (default) adds trigrams.\n- `language` enables stopword filtering for that language; omit it to\n  disable stopword filtering entirely.\n\n## Language detection\n\nLanguage detection is a real CLI flag: `--detect-language`. Detected\nlanguages appear on `result.detected_languages`:\n\n```bash\nkreuzberg extract multilingual.pdf --detect-language true --format json \\\n  | jq '.detected_languages'\n```\n\nIn a config file it lives under `[language_detection]`:\n\n```toml\n[language_detection]\nenabled = true\nmin_confidence = 0.8\ndetect_multiple = false\n```\n\nThe CLI flag enables detection with `min_confidence = 0.8` and\nsingle-language mode; use the config block to detect multiple languages or\ntune confidence.\n\n## Embeddings (`embed` command)\n\nThe standalone `embed` command produces vector embeddings for text from\n`--text` (repeatable) or stdin. It does not run extraction — pipe\nextracted content in if you want document embeddings.\n\n```bash\n# Local ONNX preset model (default provider)\nkreuzberg embed --text \"first passage\" --text \"second passage\" --preset balanced\n\n# Embed extracted document text\nkreuzberg extract report.pdf | kreuzberg embed --preset quality\n```\n\nPresets for the local provider: `fast`, `balanced` (default), `quality`,\n`multilingual`. Output defaults to JSON (`--format json`).\n\n`--provider` selects the embedding source:\n\n| Provider | Flag                                  | Notes                                         |\n| -------- | ------------------------------------- | --------------------------------------------- |\n| `local`  | `--preset \u003Cfast\\|balanced\\|quality\\|multilingual>` | **Default.** ONNX model, no API key. |\n| `llm`    | `--model \u003Cid>` `--api-key \u003Ckey>`      | liter-llm routing, e.g. `openai\u002Ftext-embedding-3-small`. |\n| `plugin` | `--plugin \u003Cname>`                     | A backend pre-registered in-process via the plugin API. |\n\n```bash\n# Provider-hosted embeddings via an LLM\nkreuzberg embed --text \"query text\" \\\n  --provider llm --model openai\u002Ftext-embedding-3-small --api-key \"$OPENAI_API_KEY\"\n```\n\nLocal embedding presets must be downloaded first if not cached. Pre-warm\nthem with the cache command:\n\n```bash\nkreuzberg cache warm --embedding-model balanced   # one preset\nkreuzberg cache warm --all-embeddings             # all four presets\n```\n\n## Programmatic access\n\nKeywords and detected languages live on the extraction result:\n\n```python\nfrom kreuzberg import extract_file_sync, ExtractionConfig\n\nresult = extract_file_sync(\n    \"paper.pdf\",\n    config=ExtractionConfig(),  # configure keywords\u002Flanguage_detection on the config\n)\nprint(result.keywords)             # extracted keywords (when enabled)\nprint(result.detected_languages)   # detected languages (when enabled)\n```\n\nSee `references\u002Fpython-api.md` and `references\u002Fconfiguration.md` in the\nsibling `kreuzberg` skill for the keyword \u002F language-detection config\nclasses and the embedding presets.\n\n## Common pitfalls\n\n- **No `--keywords` flag** — keyword extraction is config-only. Use\n  `--config-json '{\"keywords\":{...}}'` or a `[keywords]` config block.\n- **`min_score` direction** — lower is better for YAKE, higher is better\n  for RAKE; pick the threshold to match the algorithm.\n- **Embeddings ≠ extraction** — `embed` only takes raw text. Pipe\n  `kreuzberg extract` output into it for document vectors.\n- **Cold embedding models** — first local run downloads the preset; run\n  `kreuzberg cache warm --all-embeddings` to pre-populate.\n\nSee `references\u002Fadvanced-features.md` for the embeddings pipeline and\n`references\u002Fcli-reference.md` for the `embed` and `cache warm` flag sets.\n","# Extracting keywords, language, and embeddings\n\nUse this for the enrichment surface around extraction: statistical keyword\nextraction, language detection, and vector embeddings. Keywords and\nlanguage detection ride along with extraction and land on the result;\nembeddings are produced by a dedicated `embed` command.\n\n## Keywords (YAKE \u002F RAKE)\n\nKeyword extraction is configured via the `[keywords]` config block (or\ninline JSON) — there is no single `--keywords` CLI flag. When enabled,\nextracted keywords appear on `result.keywords`. Two algorithms are\navailable:\n\n- **YAKE** (`\"yake\"`) — statistical, unsupervised single-document\n  extraction. Good general default.\n- **RAKE** (`\"rake\"`) — co-occurrence \u002F phrase-based. Favors multi-word\n  key phrases.\n\n> Feature-gated: keyword extraction requires the CLI to be built with the\n> `keywords-yake` and\u002For `keywords-rake` Cargo features (both are in the\n> default\u002F`full` build). If the CLI was built without them, the `[keywords]`\n> config block is silently ignored — `result.keywords` simply stays empty\n> rather than erroring. The `\"yake\"` algorithm needs `keywords-yake`; `\"rake\"`\n> needs `keywords-rake`.\n\nEnable via inline JSON on the CLI:\n\n```bash\nkreuzberg extract paper.pdf --format json \\\n  --config-json '{\"keywords\":{\"algorithm\":\"yake\",\"max_keywords\":15,\"language\":\"en\"}}' \\\n  | jq '.keywords'\n```\n\nOr in a config file:\n\n```toml\n[keywords]\nalgorithm = \"rake\"       # \"yake\" or \"rake\"\nmax_keywords = 10        # default 10\nmin_score = 0.0          # filter below this score (ranges differ per algorithm)\nngram_range = [1, 3]     # unigrams..trigrams (default)\nlanguage = \"en\"          # stopword language; omit to skip stopword filtering\n```\n\n```bash\nkreuzberg extract report.pdf --config kreuzberg.toml --format json | jq '.keywords'\n```\n\nField notes:\n\n- `max_keywords` caps how many keywords are returned (default 10).\n- `min_score` filters low-scoring keywords; note YAKE scores are\n  *lower-is-better* while RAKE scores are *higher-is-better*, so a single\n  threshold behaves differently per algorithm.\n- `ngram_range` is `[min, max]`: `[1,1]` unigrams only, `[1,2]` adds\n  bigrams, `[1,3]` (default) adds trigrams.\n- `language` enables stopword filtering for that language; omit it to\n  disable stopword filtering entirely.\n\n## Language detection\n\nLanguage detection is a real CLI flag: `--detect-language`. Detected\nlanguages appear on `result.detected_languages`:\n\n```bash\nkreuzberg extract multilingual.pdf --detect-language true --format json \\\n  | jq '.detected_languages'\n```\n\nIn a config file it lives under `[language_detection]`:\n\n```toml\n[language_detection]\nenabled = true\nmin_confidence = 0.8\ndetect_multiple = false\n```\n\nThe CLI flag enables detection with `min_confidence = 0.8` and\nsingle-language mode; use the config block to detect multiple languages or\ntune confidence.\n\n## Embeddings (`embed` command)\n\nThe standalone `embed` command produces vector embeddings for text from\n`--text` (repeatable) or stdin. It does not run extraction — pipe\nextracted content in if you want document embeddings.\n\n```bash\n# Local ONNX preset model (default provider)\nkreuzberg embed --text \"first passage\" --text \"second passage\" --preset balanced\n\n# Embed extracted document text\nkreuzberg extract report.pdf | kreuzberg embed --preset quality\n```\n\nPresets for the local provider: `fast`, `balanced` (default), `quality`,\n`multilingual`. Output defaults to JSON (`--format json`).\n\n`--provider` selects the embedding source:\n\n| Provider | Flag                                  | Notes                                         |\n| -------- | ------------------------------------- | --------------------------------------------- |\n| `local`  | `--preset \u003Cfast\\|balanced\\|quality\\|multilingual>` | **Default.** ONNX model, no API key. |\n| `llm`    | `--model \u003Cid>` `--api-key \u003Ckey>`      | liter-llm routing, e.g. `openai\u002Ftext-embedding-3-small`. |\n| `plugin` | `--plugin \u003Cname>`                     | A backend pre-registered in-process via the plugin API. |\n\n```bash\n# Provider-hosted embeddings via an LLM\nkreuzberg embed --text \"query text\" \\\n  --provider llm --model openai\u002Ftext-embedding-3-small --api-key \"$OPENAI_API_KEY\"\n```\n\nLocal embedding presets must be downloaded first if not cached. Pre-warm\nthem with the cache command:\n\n```bash\nkreuzberg cache warm --embedding-model balanced   # one preset\nkreuzberg cache warm --all-embeddings             # all four presets\n```\n\n## Programmatic access\n\nKeywords and detected languages live on the extraction result:\n\n```python\nfrom kreuzberg import extract_file_sync, ExtractionConfig\n\nresult = extract_file_sync(\n    \"paper.pdf\",\n    config=ExtractionConfig(),  # configure keywords\u002Flanguage_detection on the config\n)\nprint(result.keywords)             # extracted keywords (when enabled)\nprint(result.detected_languages)   # detected languages (when enabled)\n```\n\nSee `references\u002Fpython-api.md` and `references\u002Fconfiguration.md` in the\nsibling `kreuzberg` skill for the keyword \u002F language-detection config\nclasses and the embedding presets.\n\n## Common pitfalls\n\n- **No `--keywords` flag** — keyword extraction is config-only. Use\n  `--config-json '{\"keywords\":{...}}'` or a `[keywords]` config block.\n- **`min_score` direction** — lower is better for YAKE, higher is better\n  for RAKE; pick the threshold to match the algorithm.\n- **Embeddings ≠ extraction** — `embed` only takes raw text. Pipe\n  `kreuzberg extract` output into it for document vectors.\n- **Cold embedding models** — first local run downloads the preset; run\n  `kreuzberg cache warm --all-embeddings` to pre-populate.\n\nSee `references\u002Fadvanced-features.md` for the embeddings pipeline and\n`references\u002Fcli-reference.md` for the `embed` and `cache warm` flag sets.\n",{"name":5,"description":7},"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fkreuzberg-dev\u002Fplugins\u002Fplugins\u002Fkreuzberg\u002Fskills\u002Fextracting-keywords",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":56,"prompt":66,"zipAvailable":67},[57,60,63],{"tool":58,"command":59},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill extracting-keywords",{"tool":61,"command":62},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill extracting-keywords",{"tool":64,"command":65},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill extracting-keywords","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fkreuzberg-dev\u002Fplugins\u002Fplugins\u002Fkreuzberg\u002Fskills\u002Fextracting-keywords\u002FSKILL.md",true,[],[70,80,88,96,104,112],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":75,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":79},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":84,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":79},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":92,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":79},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":100,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":79},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":108,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":79},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":116,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":79},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]