[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6ACFHJR3612ZE6RNX1P5WG9":3,"skill-same:01M6ACFHJR3612ZE6RNX1P5WG9":79},{"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":26,"publishedAt":51,"linkOnly":52,"markdownContent":53,"markdownBody":54,"frontMatter":55,"refName":56,"commitSha":57,"skillDirectory":58,"forksCount":59,"repositoryHtmlUrl":60,"repositoryHomepageUrl":61,"ownerHtmlUrl":62,"ownerBlogUrl":63,"ownerDescription":6,"updatedAt":64,"install":65,"resources":78},"01M6ACFHJR3612ZE6RNX1P5WG9","file-to-markdown-docker",null,"Run Apache Tika as a Docker container when you need guaranteed OCR (scanned PDFs, images) or geospatial raster support with zero local install — `apache\u002Ftika:\u003Cversion>-full` bundles Tesseract, GDAL, ImageMagick, and fonts. Also covers the minimal image, port\u002Fvolume\u002Fmemory conventions, the path-identity mount gotcha, and how to confirm OCR actually ran rather than silently returning no text. Powered by Apache Tika. Use when a local `tika-app`\u002F`tika-server` doesn't have Tesseract installed, or you want a disposable, self-contained parsing environment. Companion to the `file-to-markdown` skill, which covers the parsing calls themselves once a server is up.","apache\u002Ftika","apache","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F47359?v=4","01M6AC83E6K4FDKK7HZ8WNMQKE","The Apache Software Foundation","01M6ACEMPJD6J74AGZBFFEGEYV",4145,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fusers\u002Ffile-to-markdown-docker\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000002","文档处理","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,9],"docker","markdown","file","file to markdown","file to markdown docker","images","when","which","fonts","you","how to","app","local","volume","self","run","up","text","image","ocr","memory","support","identity","tika","2026-11-10 10:12:30",false,"---\nname: file-to-markdown-docker\ndescription: >\n  Run Apache Tika as a Docker container when you need guaranteed OCR (scanned\n  PDFs, images) or geospatial raster support with zero local install —\n  `apache\u002Ftika:\u003Cversion>-full` bundles Tesseract, GDAL, ImageMagick, and\n  fonts. Also covers the minimal image, port\u002Fvolume\u002Fmemory conventions, the\n  path-identity mount gotcha, and how to confirm OCR actually ran rather than\n  silently returning no text. Powered by Apache Tika. Use when a local `tika-app`\u002F`tika-server`\n  doesn't have Tesseract installed, or you want a disposable, self-contained\n  parsing environment. Companion to the `file-to-markdown` skill, which covers the\n  parsing calls themselves once a server is up.\n---\n\n\u003C!--\nLicensed to the Apache Software Foundation (ASF) under one or more\ncontributor license agreements.  See the NOTICE file distributed with\nthis work for additional information regarding copyright ownership.\nThe ASF licenses this file to You under the Apache License, Version 2.0\n(the \"License\"); you may not use this file except in compliance with\nthe License.  You may obtain a copy of the License at\n\n    http:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0\n\nUnless required by applicable law or agreed to in writing, software\ndistributed under the License is distributed on an \"AS IS\" BASIS,\nWITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\nSee the License for the specific language governing permissions and\nlimitations under the License.\n-->\n\nLocal override: `$TIKA_SKILLS_LOCAL\u002Ffile-to-markdown-docker\u002FLOCAL.md` (default `~\u002F.tika-skills`),\nread after this file, wins on conflict.\n\n# Running Apache Tika via Docker\n\nTwo images on Docker Hub: `apache\u002Ftika` (REST server, port 9998) and\n`apache\u002Ftika-grpc` (gRPC, port 9090). This skill covers `apache\u002Ftika`; the\n`file-to-markdown` companion skill covers the parsing calls in more depth once the\ncontainer is up (same HTTP API either way), but the examples below are\nenough to parse on their own.\n\nThe Docker route needs **no local Java at all** — the container brings its\nown. If the host lacks Java 17+ (Tika 4.x's requirement), this is the\neasiest path, not just the OCR path. This skill describes Tika 4.x images;\non a 3.x image the default output is XHTML, not Markdown.\n\n## Minimal vs `-full` — the choice that matters\n\n```\napache\u002Ftika:\u003Cversion>        # JRE + tika-server-standard, pure-Java parsers only\napache\u002Ftika:\u003Cversion>-full   # + Tesseract OCR, GDAL, ImageMagick, font sets\n```\n\n**Take `-full` if the task involves OCR (scanned PDFs, photos of documents,\nimage-only PDFs) or geospatial rasters — otherwise take minimal.** A scanned\nPDF parsed against the minimal image doesn't error; it just silently returns\nlittle or no text, because there's no Tesseract to run. If output looks\nsuspiciously short for a document that's clearly a scan, that's the signal\nyou're on the wrong image, not that the file has no text — see **Confirming\nOCR actually ran**, below.\n\nTag forms: `\u003Cversion>` rolls forward on rebuild of the same release,\n`\u003Cversion>-\u003CN>` is immutable (pin this in anything long-lived), `latest`\ntracks newest stable. Published for `linux\u002Famd64`, `linux\u002Farm64`,\n`linux\u002Fs390x` — `docker pull`\u002F`docker run` picks the right one automatically.\n\n## Starting it\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 apache\u002Ftika:latest-full\n# or pin a real version: apache\u002Ftika:4.0.0-full\ncurl -T document.pdf http:\u002F\u002Flocalhost:9998\u002Ftika\n```\n\n**Bind to `127.0.0.1`, not `0.0.0.0` or a bare port mapping, unless you\nspecifically mean to expose it.** Docker writes its own iptables rules, so\n`-p 9998:9998` (no host part) can publish the server past your host\nfirewall onto the network — a real, easy-to-hit surprise, not a theoretical\none. Tika parses untrusted input by design; the server itself does no\nauthentication, so treat network exposure as a deliberate decision, not a\ndefault.\n\nGive it a few seconds to start before the first request — `curl -sf\nhttp:\u002F\u002Flocalhost:9998\u002Fversion` is a simple readiness check to poll in a\nscript.\n\n## Mounting files: get the path identity right\n\nPlain `curl -T` uploads need **no mount at all** — the document travels in\nthe HTTP body. Mounts matter when the container must read paths itself: a\n`-c` config file, pipes fetchers reading from a directory, extra jars. The\npath-identity rule below applies to **directories of documents that callers\nreference by path**; a single config or jar mounted at a fixed container\npath (`-v ...\u002Fmy.json:\u002Fmy.json:ro`) is fine and normal — nothing translates\nthose paths back and forth. When a caller (an agent, a script) passes\nfilesystem paths, mount the directory at the **same absolute path inside\nthe container** that the caller uses outside it:\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 \\\n  -v \"\u002Fhome\u002Fme\u002Fproject:\u002Fhome\u002Fme\u002Fproject:ro\" \\\n  apache\u002Ftika:\u003Cversion>-full\n```\n\nDon't remap to something like `\u002Fdata` — if you do, every path the caller\npasses needs translating before Tika can see it, and every response\n(embedded-file paths from `\u002Funpack`, resource names) needs translating back.\nMatching the host path exactly makes the container a transparent stand-in\nfor a locally-installed Tika: paths just work in both directions. `:ro` is\nworth defaulting to — Tika only needs to read the input, and a read-only\nmount is a real containment boundary if a parse goes wrong, on top of (not\ninstead of) the process isolation Tika's own forked workers already give\nyou.\n\nThe container runs as a **non-root user, UID\u002FGID `35002:35002`**. Mounted\ninput files must be readable, and any directory Tika writes to must be\nwritable, by that UID — a mount that's `0600` owned by your host user will\nfail inside the container even though it works fine outside it.\n\n## Confirming OCR actually ran\n\nDon't assume the `-full` image means OCR fired on a given file — confirm it,\nespecially the first time you stand one up.\n\nThe signal is `tk:parsed-by-full-set`: it lists\n`org.apache.tika.parser.ocr.TesseractOCRParser` when Tesseract was run on the\nfile (a PDF page or an image), and leaves it out when it wasn't:\n\n```bash\ncurl -T scanned.pdf http:\u002F\u002Flocalhost:9998\u002Frmeta | jq '.[0].\"tk:parsed-by-full-set\"'\n```\n\nDo **not** use `pdf:ocr-page-count` for this. It counts the pages that looked\nlike they needed OCR, whether or not an engine ran, so it is non-zero with OCR\nturned off too. `tk:ocr-warnings` names input the engine was not given:\n`skipped:\u003Creason>` from an image check, `skipped:max-pages` once a\n`pages.ocr.maxPages` cap is reached (4.3.0+).\n\nNo engine listed means OCR didn't run — check you're on `-full` (not\nminimal), that the file actually needs it (a PDF with a real text layer\ncorrectly skips OCR — that's not a bug), and that no mounted config turns it\noff (see below). If you see camelCase keys (`pdf:ocrPageCount`) or `X-TIKA:*`\nkeys, your image predates the 4.0.0 metadata-key renames; upgrade the image\nrather than adapting to the old spellings.\n\n## Configuration (turning OCR off, or anything else)\n\nMount a `tika-config.json` and point `-c` at it — anything after the image\nname is appended to the entry point, which already sets `-h 0.0.0.0` (don't\npass `-h` again):\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 \\\n  -v \"$(pwd)\u002Ftika-config.json:\u002Ftika-config.json\" \\\n  apache\u002Ftika:\u003Cversion>-full -c \u002Ftika-config.json\n```\n\nThrough 4.2.x the server refuses a config file without a `server` block\n(`Couldn't find 'server' element`), so include one, even if empty; it is\nharmless on later versions.\n\nTo keep `-full`'s other parsers but disable OCR specifically (e.g. you only\nwanted GDAL), empty the `text-recognizers` list (4.1.0+):\n\n```json\n{\n  \"server\": {},\n  \"text-recognizers\": []\n}\n```\n\nSince 4.2.0 an OCR engine under `parsers` fails the config load; engines go\nunder `engines` and are named in `text-recognizers`.\n\n## Memory\n\nSize the container's `--memory` limit; don't pass `-Xmx` — the JVM sizes its\nheap from the container's own limit. Tika Pipes forks additional JVMs\n*inside* the same container for isolation, and each fork's heap comes out of\nthat same limit, so size for the forks, not just the parent:\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 --memory 4g apache\u002Ftika:\u003Cversion>-full\n```\n\n## Disposable \u002F one-shot use\n\nFor a short-lived session, name the container so you can stop it\ndeterministically — backgrounding `docker run` does NOT stop the container\nwhen your script exits, and a leaked container keeps port 9998 occupied for\nyour next attempt:\n\n```bash\ndocker run -d --rm --name tika-tmp -p 127.0.0.1:9998:9998 apache\u002Ftika:latest-full\nuntil curl -sf http:\u002F\u002Flocalhost:9998\u002Fversion >\u002Fdev\u002Fnull; do sleep 1; done\ncurl -T document.pdf http:\u002F\u002Flocalhost:9998\u002Ftika\ndocker stop tika-tmp   # --rm removes it on stop\ndocker ps --filter name=tika-tmp   # verify: should list nothing\n```\n\nFor many parses in a session, start it once and reuse it — the JVM warm-up\ncost is worth paying only once, not per file.\n\n## Troubleshooting\n\n- **`500` with an empty body** from `\u002Ftika` or `\u002Frmeta`: the response tells\n  you nothing — go straight to `docker logs \u003Ccontainer>`. A log line like\n  `Unknown fetcher type: 'file-system-fetcher' ... Available types: []` plus\n  a pf4j `No 'plugins' root` warning means the server can't find its plugins\n  directory — a symptom of an outdated or hand-built image whose working\n  directory isn't the distribution root. Official 4.0.0+ images set the\n  working directory correctly; upgrade the image (workaround for a broken\n  one: add `--workdir \u002Fopt\u002Ftika-server`).\n- **Old metadata key spellings** (`pdf:ocrPageCount`, `X-TIKA:*`): image\n  predates 4.0.0 — upgrade (see the OCR section).\n- **Connection refused right after start**: the JVM is still booting; poll\n  `curl -sf http:\u002F\u002Flocalhost:9998\u002Fversion` rather than sleeping a fixed time.\n- **Permission denied reading a mounted file**: the container runs as UID\n  35002 — see the mount section.\n\n## What's not built yet\n\nThere is no `apache\u002Ftika-app` image and no MCP mode for the Tika Docker\nimages as of this writing — the container always runs `tika-server`. If a\ntask specifically wants a stdio-spoken tool (MCP) rather than an HTTP server,\nthat isn't available through Docker today; use the `file-to-markdown` skill's\ntika-app path directly on the host instead.\n","\u003C!--\nLicensed to the Apache Software Foundation (ASF) under one or more\ncontributor license agreements.  See the NOTICE file distributed with\nthis work for additional information regarding copyright ownership.\nThe ASF licenses this file to You under the Apache License, Version 2.0\n(the \"License\"); you may not use this file except in compliance with\nthe License.  You may obtain a copy of the License at\n\n    http:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0\n\nUnless required by applicable law or agreed to in writing, software\ndistributed under the License is distributed on an \"AS IS\" BASIS,\nWITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\nSee the License for the specific language governing permissions and\nlimitations under the License.\n-->\n\nLocal override: `$TIKA_SKILLS_LOCAL\u002Ffile-to-markdown-docker\u002FLOCAL.md` (default `~\u002F.tika-skills`),\nread after this file, wins on conflict.\n\n# Running Apache Tika via Docker\n\nTwo images on Docker Hub: `apache\u002Ftika` (REST server, port 9998) and\n`apache\u002Ftika-grpc` (gRPC, port 9090). This skill covers `apache\u002Ftika`; the\n`file-to-markdown` companion skill covers the parsing calls in more depth once the\ncontainer is up (same HTTP API either way), but the examples below are\nenough to parse on their own.\n\nThe Docker route needs **no local Java at all** — the container brings its\nown. If the host lacks Java 17+ (Tika 4.x's requirement), this is the\neasiest path, not just the OCR path. This skill describes Tika 4.x images;\non a 3.x image the default output is XHTML, not Markdown.\n\n## Minimal vs `-full` — the choice that matters\n\n```\napache\u002Ftika:\u003Cversion>        # JRE + tika-server-standard, pure-Java parsers only\napache\u002Ftika:\u003Cversion>-full   # + Tesseract OCR, GDAL, ImageMagick, font sets\n```\n\n**Take `-full` if the task involves OCR (scanned PDFs, photos of documents,\nimage-only PDFs) or geospatial rasters — otherwise take minimal.** A scanned\nPDF parsed against the minimal image doesn't error; it just silently returns\nlittle or no text, because there's no Tesseract to run. If output looks\nsuspiciously short for a document that's clearly a scan, that's the signal\nyou're on the wrong image, not that the file has no text — see **Confirming\nOCR actually ran**, below.\n\nTag forms: `\u003Cversion>` rolls forward on rebuild of the same release,\n`\u003Cversion>-\u003CN>` is immutable (pin this in anything long-lived), `latest`\ntracks newest stable. Published for `linux\u002Famd64`, `linux\u002Farm64`,\n`linux\u002Fs390x` — `docker pull`\u002F`docker run` picks the right one automatically.\n\n## Starting it\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 apache\u002Ftika:latest-full\n# or pin a real version: apache\u002Ftika:4.0.0-full\ncurl -T document.pdf http:\u002F\u002Flocalhost:9998\u002Ftika\n```\n\n**Bind to `127.0.0.1`, not `0.0.0.0` or a bare port mapping, unless you\nspecifically mean to expose it.** Docker writes its own iptables rules, so\n`-p 9998:9998` (no host part) can publish the server past your host\nfirewall onto the network — a real, easy-to-hit surprise, not a theoretical\none. Tika parses untrusted input by design; the server itself does no\nauthentication, so treat network exposure as a deliberate decision, not a\ndefault.\n\nGive it a few seconds to start before the first request — `curl -sf\nhttp:\u002F\u002Flocalhost:9998\u002Fversion` is a simple readiness check to poll in a\nscript.\n\n## Mounting files: get the path identity right\n\nPlain `curl -T` uploads need **no mount at all** — the document travels in\nthe HTTP body. Mounts matter when the container must read paths itself: a\n`-c` config file, pipes fetchers reading from a directory, extra jars. The\npath-identity rule below applies to **directories of documents that callers\nreference by path**; a single config or jar mounted at a fixed container\npath (`-v ...\u002Fmy.json:\u002Fmy.json:ro`) is fine and normal — nothing translates\nthose paths back and forth. When a caller (an agent, a script) passes\nfilesystem paths, mount the directory at the **same absolute path inside\nthe container** that the caller uses outside it:\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 \\\n  -v \"\u002Fhome\u002Fme\u002Fproject:\u002Fhome\u002Fme\u002Fproject:ro\" \\\n  apache\u002Ftika:\u003Cversion>-full\n```\n\nDon't remap to something like `\u002Fdata` — if you do, every path the caller\npasses needs translating before Tika can see it, and every response\n(embedded-file paths from `\u002Funpack`, resource names) needs translating back.\nMatching the host path exactly makes the container a transparent stand-in\nfor a locally-installed Tika: paths just work in both directions. `:ro` is\nworth defaulting to — Tika only needs to read the input, and a read-only\nmount is a real containment boundary if a parse goes wrong, on top of (not\ninstead of) the process isolation Tika's own forked workers already give\nyou.\n\nThe container runs as a **non-root user, UID\u002FGID `35002:35002`**. Mounted\ninput files must be readable, and any directory Tika writes to must be\nwritable, by that UID — a mount that's `0600` owned by your host user will\nfail inside the container even though it works fine outside it.\n\n## Confirming OCR actually ran\n\nDon't assume the `-full` image means OCR fired on a given file — confirm it,\nespecially the first time you stand one up.\n\nThe signal is `tk:parsed-by-full-set`: it lists\n`org.apache.tika.parser.ocr.TesseractOCRParser` when Tesseract was run on the\nfile (a PDF page or an image), and leaves it out when it wasn't:\n\n```bash\ncurl -T scanned.pdf http:\u002F\u002Flocalhost:9998\u002Frmeta | jq '.[0].\"tk:parsed-by-full-set\"'\n```\n\nDo **not** use `pdf:ocr-page-count` for this. It counts the pages that looked\nlike they needed OCR, whether or not an engine ran, so it is non-zero with OCR\nturned off too. `tk:ocr-warnings` names input the engine was not given:\n`skipped:\u003Creason>` from an image check, `skipped:max-pages` once a\n`pages.ocr.maxPages` cap is reached (4.3.0+).\n\nNo engine listed means OCR didn't run — check you're on `-full` (not\nminimal), that the file actually needs it (a PDF with a real text layer\ncorrectly skips OCR — that's not a bug), and that no mounted config turns it\noff (see below). If you see camelCase keys (`pdf:ocrPageCount`) or `X-TIKA:*`\nkeys, your image predates the 4.0.0 metadata-key renames; upgrade the image\nrather than adapting to the old spellings.\n\n## Configuration (turning OCR off, or anything else)\n\nMount a `tika-config.json` and point `-c` at it — anything after the image\nname is appended to the entry point, which already sets `-h 0.0.0.0` (don't\npass `-h` again):\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 \\\n  -v \"$(pwd)\u002Ftika-config.json:\u002Ftika-config.json\" \\\n  apache\u002Ftika:\u003Cversion>-full -c \u002Ftika-config.json\n```\n\nThrough 4.2.x the server refuses a config file without a `server` block\n(`Couldn't find 'server' element`), so include one, even if empty; it is\nharmless on later versions.\n\nTo keep `-full`'s other parsers but disable OCR specifically (e.g. you only\nwanted GDAL), empty the `text-recognizers` list (4.1.0+):\n\n```json\n{\n  \"server\": {},\n  \"text-recognizers\": []\n}\n```\n\nSince 4.2.0 an OCR engine under `parsers` fails the config load; engines go\nunder `engines` and are named in `text-recognizers`.\n\n## Memory\n\nSize the container's `--memory` limit; don't pass `-Xmx` — the JVM sizes its\nheap from the container's own limit. Tika Pipes forks additional JVMs\n*inside* the same container for isolation, and each fork's heap comes out of\nthat same limit, so size for the forks, not just the parent:\n\n```bash\ndocker run -d -p 127.0.0.1:9998:9998 --memory 4g apache\u002Ftika:\u003Cversion>-full\n```\n\n## Disposable \u002F one-shot use\n\nFor a short-lived session, name the container so you can stop it\ndeterministically — backgrounding `docker run` does NOT stop the container\nwhen your script exits, and a leaked container keeps port 9998 occupied for\nyour next attempt:\n\n```bash\ndocker run -d --rm --name tika-tmp -p 127.0.0.1:9998:9998 apache\u002Ftika:latest-full\nuntil curl -sf http:\u002F\u002Flocalhost:9998\u002Fversion >\u002Fdev\u002Fnull; do sleep 1; done\ncurl -T document.pdf http:\u002F\u002Flocalhost:9998\u002Ftika\ndocker stop tika-tmp   # --rm removes it on stop\ndocker ps --filter name=tika-tmp   # verify: should list nothing\n```\n\nFor many parses in a session, start it once and reuse it — the JVM warm-up\ncost is worth paying only once, not per file.\n\n## Troubleshooting\n\n- **`500` with an empty body** from `\u002Ftika` or `\u002Frmeta`: the response tells\n  you nothing — go straight to `docker logs \u003Ccontainer>`. A log line like\n  `Unknown fetcher type: 'file-system-fetcher' ... Available types: []` plus\n  a pf4j `No 'plugins' root` warning means the server can't find its plugins\n  directory — a symptom of an outdated or hand-built image whose working\n  directory isn't the distribution root. Official 4.0.0+ images set the\n  working directory correctly; upgrade the image (workaround for a broken\n  one: add `--workdir \u002Fopt\u002Ftika-server`).\n- **Old metadata key spellings** (`pdf:ocrPageCount`, `X-TIKA:*`): image\n  predates 4.0.0 — upgrade (see the OCR section).\n- **Connection refused right after start**: the JVM is still booting; poll\n  `curl -sf http:\u002F\u002Flocalhost:9998\u002Fversion` rather than sleeping a fixed time.\n- **Permission denied reading a mounted file**: the container runs as UID\n  35002 — see the mount section.\n\n## What's not built yet\n\nThere is no `apache\u002Ftika-app` image and no MCP mode for the Tika Docker\nimages as of this writing — the container always runs `tika-server`. If a\ntask specifically wants a stdio-spoken tool (MCP) rather than an HTTP server,\nthat isn't available through Docker today; use the `file-to-markdown` skill's\ntika-app path directly on the host instead.\n",{"name":5,"description":7},"main","9f8b48425fea3630020924f6ae96e98eabb9ce12",".skills\u002Fusers\u002Ffile-to-markdown-docker",1039,"https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika","https:\u002F\u002Ftika.apache.org\u002F","https:\u002F\u002Fgithub.com\u002Fapache","https:\u002F\u002Fwww.apache.org\u002F","2026-11-13 08:35:47",{"source":8,"skillName":5,"commands":66,"prompt":76,"zipAvailable":77},[67,70,73],{"tool":68,"command":69},"npx","npx skills add apache\u002Ftika --skill file-to-markdown-docker",{"tool":71,"command":72},"bunx","bunx skills add apache\u002Ftika --skill file-to-markdown-docker",{"tool":74,"command":75},"pnpm","pnpm dlx skills add apache\u002Ftika --skill file-to-markdown-docker","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fusers\u002Ffile-to-markdown-docker\u002FSKILL.md",true,[],[80,109,136,163,188,203],{"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":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":93,"publishedAt":51},"01M6ACFKZR081Z7QX5XB1ZA2R2","file-to-markdown","Turn almost any file into Markdown plus metadata — PDF, Office, HTML, email, archives, images, audio\u002Fvideo, 1000+ formats — powered by Apache Tika, either via the tika-app CLI (zero setup, one file) or a running tika-server (curl, warm process, many calls). Leads with rmeta (structured, embedded-item-aware output) as the default operation rather than flat concatenated text, since you can't tell whether a file has embedded content from its extension. Covers metadata-only triage, type and language detection, OCR, and output-size discipline for agent context. Use whenever a task involves reading the content of a file whose format you don't want to hand-parse.","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fusers\u002Ffile-to-markdown\u002FSKILL.md",[86,87,90],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":88,"categoryName":89,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":91,"categoryName":92,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[30,28,29,94,95,9,96,97,98,99,50,32,100,38,46,101,102,103,104,105,36,106,44,107,108],"PDF","cli","triage","setup","content","metadata","Video","Audio","parse","default","only","either","hand","server","discipline",{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":116,"publishedAt":51},"01M6ACFE1684DE184VE0D7PQVY","file-forensics","Examine what a file claims about itself and what it actually contains — powered by Apache Tika. True content-based type detection (extensions lie), provenance claims (authors, dates, creating application), revision and tamper signals (PDF incremental updates, tracked changes, hidden slides, zip integrity), hidden and embedded content (attachments, macros), risk indicators (PDF JavaScript actions, encryption), and content digests. Evidence gathering, not verdicts: Tika reports what the file asserts and what parsing observed; it does not attribute authorship or validate signatures. Use for triaging suspicious files, provenance questions, e-discovery-style review, or \"is this file what it claims to be.\"","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fusers\u002Ffile-forensics\u002FSKILL.md",[115],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[117,29,118,119,120,94,9,121,98,122,50,123,124,125,126,127,128,129,130,131,132,133,134,135],"file forensics","forensics","javascript","suspicious","validate","review","actions","signatures","files","updates","encryption","reports","claims","signals","revision","integrity","zip","evidence","risk",{"skillNo":137,"name":138,"nameZh":6,"description":139,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":140,"categories":141,"keywords":142,"publishedAt":51},"01M6ACFB92SPADF0FY4490B847","update-site-for-release","Update\u002Fpublish the Apache Tika website (tika-site SVN repo) for a release — step 17 of the Release Process. Handles the 4.x track (Changes page + aggregate javadoc + Antora docs branch) vs the 3.x maintenance track (full per-version apt docs + javadoc). Use for \"update the site\", \"publish the site for X.Y.Z\", \"the website part of the release\".","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fdevs\u002Fupdate-site-for-release\u002FSKILL.md",[],[143,144,145,146,147,9,148,149,50,150,151,152,153,154,155,156,157,158,159,160,161,162],"update","release","update site for release","for","site","release process","docs","branch","publish","track","website","svn","full","apt","page","repo","use","version","changes","maintenance",{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":175,"publishedAt":51},"01M6ACF9G52NYHXQ0BVSJ3VH6E","tika-eval-h2-query","Query the tika-eval H2 database directly for counts and joins the canned reports do not compute — connection gotchas, key tables, example queries. Use when the xlsx\u002Fsummary.md reports are not enough.","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fdevs\u002Ftika-eval-h2-query\u002FSKILL.md",[169,172],{"categoryNo":170,"categoryName":171,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000004","电子表格",{"categoryNo":173,"categoryName":174,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[176,50,177,178,179,180,181,146,182,159,33,183,128,184,185,186,187],"tika eval h2 query","eval","query","database","xlsx","do","example","compute","queries","summary","connection","tables",{"skillNo":189,"name":190,"nameZh":6,"description":191,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":192,"categories":193,"keywords":194,"publishedAt":51},"01M6ACF7V56CCWHNVR614VBNX5","tika-eval-encoding-regression","Condensed tika-eval pattern for charset-detector regression hunts (\"A picks encoding X, B picks Y\") using one build and two configs — encoding-pair flip queries, OOV\u002Flanguageness\u002FFFFD signals, per-file detector attribution.","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fdevs\u002Ftika-eval-encoding-regression\u002FSKILL.md",[],[195,50,196,177,197,146,198,29,184,130,199,200,201,202],"tika eval encoding regression","regression","encoding","attribution","build","detector","pattern","pair",{"skillNo":204,"name":205,"nameZh":6,"description":206,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":207,"categories":208,"keywords":209,"publishedAt":51},"01M6ACF53J5MJCDNAEMCBV0RPP","tika-eval-compare","Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for \"compare before\u002Fafter extracts\", \"eval this change against the corpus\".","https:\u002F\u002Fgithub.com\u002Fapache\u002Ftika\u002Fblob\u002Fmain\u002F.skills\u002Fdevs\u002Ftika-eval-compare\u002FSKILL.md",[],[210,50,177,211,212,98,146,159,213,214,197,215],"tika eval compare","compare","change","document","detect","embedded"]