Strings Extractor
extract_stringsRead-onlyIdempotent
List the readable text inside any file, given as base64, at most 2,000,000 bytes: runs of ASCII and UTF-16LE characters (UTF-8 on request) at least min_length long, with byte offsets. Labels each string by shape as a URL, email, IP address, file path, key or token, or other, and counts every kind across the whole file; the list can be narrowed by kind or text and holds up to 2,000 strings in file order. Pure standard library, no network or model.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"file": "<file>"
},
"name": "extract_strings"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| file | Yes | Required. Any file, as base64 or a base64 data URL; at most 2,000,000 bytes once decoded. | |
| kinds | No | Only list strings of these kinds (url, email, ip, path, key, other). Counts always cover all. | |
| limit | No | Most strings to list, in file order. Default and maximum 2,000. | |
| contains | No | Only list strings containing this text (any case). | |
| filename | No | Optional file name, used only in the result text. | |
| encodings | No | Which text to look for: ascii, utf-16le and utf-8. Default ascii and utf-16le; utf-8 also reads accented and other non-English letters, and finds more false text in random bytes. | |
| min_length | No | Shortest run of characters to keep, 3 to 32. Default 4. |