Supported formats

ZahirScan detects format from content and path, then runs the appropriate metadata and (where applicable) template pipeline.

Logs & text

• Plain text logs, JSON-formatted logs, structured log files
• TXT, Markdown (.md)

Documents

• DOCX, XLSX, PPTX, PDF, EPUB

Settings & structured text

• INI (.ini, .cfg), TOML (.toml, .lock), YAML (.yaml, .yml), XML (.xml)
• CSV / TSV / tab / psv, JSON, HTML (.html, .htm)

Tabular & columnar

• Parquet; Arrow IPC / Feather (.arrow, .feather, .ipc); Avro; ORC
• NumPy (.npy, .npz); HDF5 (.h5, .hdf5); NetCDF (.nc, .cdf)
• Matrix Market (.mtx); MATLAB (.mat); Zarr (.zarr)
• Tetration (.tet) — see tetration-docs

Models

• ONNX (.onnx), GGUF (.gguf), TensorFlow Lite (.tflite), Safetensors (.safetensors)

Archives

• ZIP (.zip)
• TAR family (.tar, .tar.gz, .tgz, .tar.bz2, .tar.xz)

Code & scripts

Common source and config extensions (e.g. .py, .rs, .js, .ts, .sh, Makefile, Dockerfile, …) via linguist-style detection plus optional shebang.

Media

Type	Examples
Images	JPEG, PNG, GIF, WebP, BMP, TIFF
Video	MP4, MKV, AVI, MOV, WMV, FLV, WebM, M4V, 3GP, OGV
Audio	MP3, FLAC, WAV, M4A, AAC, OGG, Opus, WMA, APE, DSD, DSF

Databases

• SQLite (.db, .sqlite, .sqlite3)

What each category extracts is summarized on Metadata extraction.