Convert Kindle (AZW3) to MediaWiki
Turn Kindle (AZW3), Amazon's KF8 Kindle format used by modern Kindle devices, into MediaWiki, the wiki markup that powers Wikipedia and MediaWiki sites. FileChanger runs the document conversion in one call, no local tooling to install.
In the browser
Upload a file and download the result, no pipeline required. Open the converter.
One API call
POST the file with from=azw3 and to=mediawiki, get the converted bytes back. Jump to the code.
From an agent
FileChanger ships an MCP server, so agents can run this conversion themselves. See the docs for setup.
Convert Kindle (AZW3) to MediaWiki in code
The same call from any language: POST the file to /api/convert, the response body is the converted file.
curl -X POST https://filechanger.io/api/convert \
-H "X-Api-Key: fc_your_api_key_here" \
-F "file=@input.azw3" \
-F "from=azw3" \
-F "to=mediawiki" \
-o output.mediawikiimport requests
with open("input.azw3", "rb") as f:
response = requests.post(
"https://filechanger.io/api/convert",
headers={"X-Api-Key": "fc_your_api_key_here"},
files={"file": f},
data={"from": "azw3", "to": "mediawiki"},
)
response.raise_for_status()
with open("output.mediawiki", "wb") as out:
out.write(response.content)const formData = new FormData();
formData.append("file", fileInput.files[0]);
formData.append("from", "azw3");
formData.append("to", "mediawiki");
const response = await fetch("https://filechanger.io/api/convert", {
method: "POST",
headers: { "X-Api-Key": "fc_your_api_key_here" },
body: formData,
});
const output = await response.blob();import java.net.http.HttpRequest.BodyPublishers;
import java.net.http.HttpResponse.BodyHandlers;
import java.nio.file.Path;
String boundary = "----FormBoundary" + System.currentTimeMillis();
byte[] fileBytes = java.nio.file.Files.readAllBytes(Path.of("input.azw3"));
String body = "--" + boundary + "\r\n"
+ "Content-Disposition: form-data; name=\"file\"; filename=\"input.azw3\"\r\n"
+ "Content-Type: application/octet-stream\r\n\r\n"
+ new String(fileBytes) + "\r\n"
+ "--" + boundary + "\r\nContent-Disposition: form-data; name=\"from\"\r\n\r\nazw3\r\n"
+ "--" + boundary + "\r\nContent-Disposition: form-data; name=\"to\"\r\n\r\nmediawiki\r\n"
+ "--" + boundary + "--";
HttpClient client = HttpClient.newHttpClient();
HttpRequest request = HttpRequest.newBuilder()
.uri(URI.create("https://filechanger.io/api/convert"))
.header("X-Api-Key", "fc_your_api_key_here")
.header("Content-Type", "multipart/form-data; boundary=" + boundary)
.POST(BodyPublishers.ofString(body))
.build();
HttpResponse<byte[]> response = client.send(
request, BodyHandlers.ofByteArray());
java.nio.file.Files.write(Path.of("output.mediawiki"), response.body());import Network.HTTP.Client
import Network.HTTP.Client.MultipartFormData
import qualified Data.ByteString.Lazy as BL
main :: IO ()
main = do
manager <- newManager defaultManagerSettings
initialReq <- parseRequest "POST https://filechanger.io/api/convert"
let req = initialReq
{ requestHeaders = [("X-Api-Key", "fc_your_api_key_here")] }
request <- formDataBody
[ partFileSource "file" "input.azw3"
, partBS "from" "azw3"
, partBS "to" "mediawiki"
] req
response <- httpLbs request manager
BL.writeFile "output.mediawiki" (responseBody response)About MediaWiki
MediaWiki markup is the wikitext syntax behind Wikipedia and every other MediaWiki-powered site. Editors, archivists, and companies running internal MediaWiki instances use it for collaboratively edited articles with templates, tables, and citations.
Frequently asked questions
How do I convert Kindle (AZW3) to MediaWiki?
Upload your file in the browser converter and pick MediaWiki as the target, or POST it to /api/convert with from=azw3 and to=mediawiki. The response body is the converted file.
Is converting Kindle (AZW3) to MediaWiki free?
Yes. The free tier includes 2 GB of conversions per month, metered on input size. See pricing for the paid tiers.
Is there an API for Kindle (AZW3) to MediaWiki conversion?
Yes, this whole conversion is one POST to /api/convert (the code examples above show it in five languages), plus an MCP server so AI agents can run it directly.