Deposition Question Development
ReviewAudited by ClawScan on May 1, 2026.
Overview
The skill appears coherent and purpose-aligned for local legal PDF analysis, with the main cautions being sensitive document text, recursive folder processing, and an optional Python package install.
Before installing, confirm you are comfortable processing the selected legal PDFs with this agent. Use specific PDF paths or a narrow production folder, install `pdfplumber` only from a trusted environment if needed, keep the generated JSON secure, and manually verify document IDs, quotes, and legal conclusions.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If a produced PDF contains misleading prompt-like text, it could influence the agent unless the user keeps the legal theory and task instructions authoritative.
The skill intentionally places potentially adversarial document content into the agent's analysis context.
Use when tasks involve reviewing opponent-produced PDFs
Treat PDF contents only as evidence for quotes and relevance; ignore any instructions embedded in the documents and manually verify important outputs.
If the user points it at too broad a folder, it may process more PDFs than intended.
The documented workflow runs a local extraction tool over a user-specified file or folder, optionally recursively.
python scripts/extract_relativity_pages.py --input <pdf-folder-or-file> --recurse --output <relativity_pages.json>
Use a narrow case-production folder or specific PDF paths, and review the generated JSON before relying on it.
Installing an unpinned package can expose the environment to ordinary package supply-chain risk.
The skill discloses an optional dependency install, but the package version is not pinned.
python -m pip install --user pdfplumber
Install dependencies from a trusted package index, preferably in a virtual environment, and pin or review the package version if used for sensitive matters.
Confidential legal document text may persist in the generated JSON file and be available to later tools or users with file access.
The script includes full extracted page text in the page records and writes those records to the output JSON.
"page_text": page_text, ... json.dump(result, handle, indent=2 if args.pretty else None)
Write output only to an appropriate case workspace, restrict access to the JSON, and delete or archive it according to the matter's confidentiality rules.