ClawHub

pdf-extract-skill

v0.0.10

OpenClaw PDF extraction skill using OpenDataLoader. Use when the user wants to extract and process PDF content for RAG, embeddings, or coordinate-based citat...

⭐ 0· 92·0 current·0 all-time
byLucas Moyano@secondport
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
✓
Purpose & Capability
Name/description, required binaries (java, python3, opendataloader-pdf), and the SKILL.md all describe running the OpenDataLoader CLI and hybrid backend — this is coherent and expected for a PDF extraction skill. The skill does not request unrelated services or credentials.
✓
Instruction Scope
SKILL.md only instructs running local CLI commands, starting a local hybrid backend, and using flags for OCR, pages, formats, etc. It does not ask to read unrelated files, access unrelated environment variables, or send data to external endpoints. It explicitly recommends binding the hybrid backend to localhost and contains a security checklist.
ℹ
Install Mechanism
There is no install spec (instruction-only), which limits the surface written to disk — low intrinsic risk. However, because opendataloader-pdf must be installed by the user, the lack of an author-provided pinned install command or source/homepage in the registry is notable; the docs do advise verifying PyPI/GitHub metadata and using pinned installs in an isolated environment.
✓
Credentials
The skill requests no environment variables, no credentials, and no config paths. This is proportionate to its stated CLI-only purpose.
✓
Persistence & Privilege
The skill is not flagged always:true, is user-invocable, and contains no instructions to persistently modify agent/system configuration or other skills. Autonomous invocation is allowed by default but is not combined with other concerning flags.
Assessment
This skill appears to be a coherent CLI guide for the opendataloader-pdf tool and its hybrid backend, and it contains reasonable security advice (use venv/container, bind hybrid backend to localhost, verify package provenance). Before installing or running anything: (1) verify the opendataloader-pdf project on PyPI/GitHub matches an official source (the registry entry for this skill has no homepage/source URL), (2) install only into an isolated environment (venv/container/VM) and use a pinned version, (3) inspect the package/repository and its dependencies for unexpected network behavior (especially if you enable image-description or hybrid flags), (4) run the hybrid backend with --host 127.0.0.1 and verify listeners before processing sensitive PDFs. If you need higher assurance, request the skill author to provide an explicit install spec and source URL.

Like a lobster shell, security has layers — review code before you run it.

latestvk97dmm6w37angspyp5cr134v9n83m00d

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

📄 Clawdis
Binsjava, python3, opendataloader-pdf

Comments