daily-scan
v1.0.4Scan photographed documents into searchable PDFs with OCR and stable file naming. Use when the user sends document photos and asks to scan, save, archive, OC...
⭐ 0· 30·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description describe turning photos into searchable PDFs. Bundled scripts perform image cleanup, OCR (Tesseract or PaddleOCR), PDF assembly, and local search; the requested local binaries and libraries (tesseract, ocrmypdf or reportlab, OpenCV, Pillow) are consistent with that purpose.
Instruction Scope
SKILL.md limits actions to local processing, saving, and searching. The scripts follow that: they read local image files, write PDFs into daily-scan-storage, and run local OCR. Notes: build_searchable_pdf.py calls external CLIs (tesseract, ocrmypdf) via subprocess (arguments passed as lists, avoiding shell interpolation). The PaddleOCR path may implicitly download model weights at runtime (not explicitly declared in SKILL.md) — this could cause outbound network activity on first use.
Install Mechanism
No install spec (instruction-only) — lower risk. The skill bundles Python scripts but does not itself download or install remote code. Host must ensure required binaries/packages are present; installing those may pull external packages (pip, system packages), but that is outside the skill bundle.
Credentials
The skill requests no environment variables, credentials, or config paths. All file I/O is local and limited to the configured storage path. No unrelated secrets or services are requested.
Persistence & Privilege
always:false and no code modifies other skills or global agent settings. The skill only writes its own output under daily-scan-storage and does not request persistent elevated privileges.
Assessment
This skill appears to do what it claims: local image preprocessing, OCR, PDF creation, and local search. Before installing/using: 1) Ensure tesseract and either ocrmypdf or reportlab + dependencies are installed on the host (ocrmypdf may require ghostscript); 2) Be aware the experimental PaddleOCR path can automatically download model weights (outbound network activity) — disable or sandbox if you require strict offline operation; 3) Validate storage location and file permissions (daily-scan-storage) to avoid accidentally exposing sensitive images; 4) Test in a safe environment first to confirm dependencies and behavior; 5) If you need strict no-network guarantees, avoid enabling the PaddleOCR path or pre-install its models offline.Like a lobster shell, security has layers — review code before you run it.
latestvk97fh5r06efcrvfzbyvmdn05r984c3ar
License
MIT-0
Free to use, modify, and redistribute. No attribution required.