FAQ
1. Too much garbage data?
Method 1: Adjust rules in the workspace config.toml
Method 2: Ignore resource files (-n)2. Error: This application has shell, the retrieval results may not be accurate, Please remove the shell and try again!
The app is packed/shelled. Unpack/dump it first:
Android:
Xposed module: dexdump
frida module: FRIDA-DEXDump
No-root unpacking: blackdex
iOS:
frida module:
Windows: frida-ipa-dump
macOS: frida-ios-dumpSee Shell Detection & Unpacking for the built-in --unpack flow and fix_magic Repair for repairing dumped files.
3. Error: File download failed! Please download the file manually and try again.
File download failed.
1) Check if the URL is correct
2) Check network issues, or configure headers, data, and method in
the workspace config.toml and re-run4. Error: Decompilation failed, please submit error information at https://github.com/kelvinBen/AppInfoScanner/issues
File decompilation failed. This is the most frequent cluster in the issue tracker (#14 / #22 / #37 / #39 / #50, …). Troubleshoot in this order:
1) Confirm Java 11+ (required by apktool 3.0.3; most legacy issues were Java 8 or below)
2) Since V1.0.10 a repair-retry is built in: on failure it verifies the EOCD first,
repairs corrupted zip/dex/manifest magic (with a .bak backup) and retries once —
most magic-corruption cases now need no manual work
3) Still failing? Repair the sample standalone first:
python3 libs/core/fix_magic.py detect/fix <file> (see the fix_magic guide),
then re-scan the repaired artifact
4) Otherwise submit the error screenshot and the APK at
https://github.com/kelvinBen/AppInfoScanner/issues5. What are the runtime requirements?
Python : 3.11+ (3.14 verified), dependencies in requirements.txt
Java : apktool 3.0.3 requires Java 11+ (the old "Java <= 1.8" note applied to legacy jars)
Tools : Windows uses the bundled tools/; macOS/Linux auto-installs missing Java/adb/frida6. Where are the workspace and scan results?
The default output root is ~/Documents/AppInfoScanner/ (falls back to ~/AppInfoScanner/ on minimal Linux without ~/Documents):
config.toml user config (generated on first run)
out/ scan results (result/<sample>_<timestamp>/ with all three report formats)
download/ auto-downloaded APK/IPA/H5
history/ history (always follows the default root; -o does not change it)
logs/ run logs (only the latest 20 kept)The -o option only changes the scan-result directory.
7. How is the legacy config.py migrated?
On first run the legacy workspace config.py is migrated to config.toml automatically (the original is kept as config.py.bak). Since V1.0.11 the config uses a 3-tier deep merge — built-in defaults → workspace config.toml → CLI options — with missing keys falling back to defaults and versioned auto-migration. See the config reference.
8. Frida version mismatch during unpacking?
The python-side frida core and the device-side frida-server must be the same version (currently pinned to 17.18.0). --unpack pushes a version-matched frida-server automatically; if you deployed one manually, make sure the versions match. See Packer Detection & Unpacking.
9. CentOS pip install fails with "No module named '_ctypes'" after building Python?
The system is missing the libffi development package: run yum install libffi-devel -y, then rebuild with make && make install in the Python source directory. Full steps in the migrated article.
10. Scan stuck at Searching for strings that match the rules?
See #27. Usually an oversized file dragging down the full-string scan. Since V1.0.10 there are guard limits: files over 10 MB are skipped with [-] Skip large file, and extracted strings are truncated at 512 characters. If it still feels slow, raise threads with -t or narrow rules with -r.
11. Dumped dex is empty / extraction returns nothing?
See #42 and #44. Dumped artifacts often have damaged headers (magic / file size / checksum wiped), so decompiling or scanning yields nothing. Repair with the built-in module first:
python3 libs/core/fix_magic.py detect <dumped-artifact> # detect damage
python3 libs/core/fix_magic.py fix dex <file> # repair dex (zip/axml likewise)Then scan the existing dump directly with --prefer-dump — no re-unpacking needed. See the fix_magic guide.
12. Errors installing requirements.txt or frida?
1) Use a CN PyPI mirror: pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple
2) The frida trio is version-pinned (frida==17.18.0 / frida-tools==14.10.4 /
frida-dexdump==2.0.1); do not mix versions, and always upgrade all three together
3) Python 3.11+ is required13. Does it work on macOS / Linux?
Yes (#1, #24). All three platforms are supported: Windows uses the bundled tools/ toolchain; macOS / Linux auto-installs missing Java, adb and frida (brew / apt, …); the strings binary needed for iOS parsing resolves from PATH on macOS / Linux (with macOS section-aware handling) and from the bundled strings.exe on Windows. Only the Windows-only binaries under tools/unpacker/ (adb.exe, aapt.exe, …) are Windows-specific — their capabilities have equivalents on macOS / Linux.
14. How to batch-scan multiple APPs or directories?
See #16 and #10. Pass a directory to -i for directory-level batch scanning on Android / iOS / Web alike; each sample gets its own timestamped result directory:
python app.py android -i /Work/APKs/
python app.py web -i /Work/sources/15. adb: no devices/emulators found or Windows missing AdbWinApi.dll?
1) Normal scanning (URL/IP/component/credential extraction) needs no device —
this error only affects --unpack
2) Before unpacking: enable USB debugging, authorize the computer, use a data-capable
cable, and confirm the device shows up in adb devices
3) On Windows use the bundled tools/unpacker/adb.exe (its DLL dependencies sit in the
same directory) — do not copy adb.exe elsewhere to run it16. Error: cannot find the Bootstrapper module?
See #13. Usually an incomplete clone or files mixed between versions. Re-clone the repo completely (or fetch a full source archive from the download center) and run again.