What does this skill do, and when should you use it?
This skill is the reporting module of the AI Legal Assistant collection (one of 14 skills), invoked via /legal report-pdf. It first locates the most recent contract-review output file in the working directory (e.g. CONTRACT-REVIEW-*.md), parses the safety score, risk distribution, clause analysis, missing protections, and negotiation priorities, then uses Python + ReportLab to produce a multi-page PDF with cover page, score gauge, risk dashboard, and legal disclaimer. It performs no legal analysis itself — a review command must run first to generate the data. The whole collection installs with a single curl command.
1) Uses Glob to search the working directory for review output files in priority order: CONTRACT-REVIEW-*.md, FREELANCER-REVIEW-*.md, COMPLIANCE-AUDIT-*.md, NDA-REVIEW-*.md; 2) parses score, grade, contract type, parties, risk dashboard, clause analysis, obligations timeline, compliance flags, and negotiation priorities via regex/string matching; 3) prefers running the repo's scripts/generate_pdf_report.py, falling back to an inline Python script that calls ReportLab; 4) outputs a Letter-size PDF with cover (semi-circular score gauge), executive summary, risk dashboard, clause-by-clause entries, missing-protections checklist, obligations table, compliance flags, negotiation priorities, next-steps checklist, and a final legal disclaimer, with CONFIDENTIAL page footers; 5) verifies the file exists and is a reasonable size (over 10KB).
- A freelancer who has run /legal review and needs a formal PDF to send to a client or attorney
- An AI automation agency selling contract review as a paid service and needing branded client deliverables
- A small-business owner who wants a structured contract-risk record including deadlines and consequences
- A compliance person who wants /legal compliance audit results fixed into a written report with flags
- A negotiator who wants the ranked negotiation priorities rendered as a printable brief before a call
How do you install this skill?
- When the script is missing, the skill auto-installs reportlab and executes inline-generated Python; verify dependency provenance and inspect generated code before use.
- README claims like '60-second contract review' lack execution evidence; this is a static review — the PDF pipeline was not executed.
- Repository license is unknown; the script filename in the skill differs from README — verify the actual script location after install.
- Core function depends on Claude Code and PyPI, which may be unreachable from mainland-China networks; no Chinese-language support is documented.
- The PDF contains legal analysis content; always have it reviewed by a licensed attorney before relying on it.
- Shell / CLI
- Local filesystem
Python 3.8+reportlab
Install the whole collection (including this skill and the PDF script): curl -fsSL https://raw.githubusercontent.com/zubair-trabzada/ai-legal-claude/main/install.sh | bash. The skill lives at skills/legal-report-pdf/SKILL.md in the repo. An uninstall.sh is also provided. Requirements: Claude Code, Python 3.8+, and reportlab (pip3 install reportlab). Manual standalone installation of just this skill is not documented.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/zubair-trabzada/ai-legal-claude.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/legal-report-pdf" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Run any review command first (e.g. /legal review <file>), then run /legal report-pdf in the same working directory. The skill auto-locates the most recent review file and outputs CONTRACT-REVIEW-REPORT.pdf. If no review data is found it prompts you to run /legal review first. In environments without slash commands, follow the SKILL.md workflow manually: run generate_pdf_report.py with --input pointing at the analysis markdown and --output for the PDF path.
What are this skill's strengths and limitations?
- Complete report structure: score gauge, risk dashboard, clause-by-clause analysis, obligations timeline, negotiation priorities
- Built-in fallbacks: generates Python inline if the script is missing, and degrades the gauge to a color-coded score box if drawing fails
- Handles missing data sections gracefully and reports which sections were skipped
- Includes legal disclaimer pages, suitable for client delivery
- Not standalone — depends on /legal review, /legal freelancer, or similar commands producing data first
- Requires Python 3.8+ and ReportLab; cannot run in pure web or Python-less environments
- License unknown, so commercial-use terms are unclear
- No test suite or sample PDF in the docs to evidence output quality
- Parsing depends on the specific markdown markers of upstream review output; format changes upstream would break it
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Legal Analysis PDF Report Generator this page | 47 · Use with care | ★ 1.8k | 7mo ago | — |
| Open Cowork PDF Skill | 69 · Recommended | ★ 2.2k | 12d ago | MIT |
| AI Legal Assistant (legal/SKILL.md Orchestrator) | 43 · Not recommended | ★ 1.8k | 7mo ago | — |
| Missing Protections Finder (AI Legal Assistant) | 45 · Not recommended | ★ 1.8k | 7mo ago | — |
| AI Legal Assistant: Full Contract Review | 41 · Not recommended | ★ 1.8k | 7mo ago | — |
The source names no other PDF report tools; within the repo it serves as the shared report output for the 13 analysis/generation skills.
How did FollowSkills review this skill?
The skill itself is fairly restrained: it only reads existing analysis files in the working directory and produces a PDF, and includes explicit legal disclaimers. Deductions: Phase 2.3 auto-installs reportlab via pip without user confirmation, and instructs generating and executing inline Python when the script is missing — an unconfirmed code-execution surface; script search uses relative globs with mediocre data-flow transparency; no rollback mechanism documented.
Instructions are self-consistent and well structured, with an error-handling table and graceful fallbacks (no data/no script paths). But the key path depends entirely on a generate_pdf_report.py whose contents are not provided here, plus inline generation quality; parsing depends on exact upstream markdown formats (e.g. 'Contract Safety Score: X/100') with no handling for format drift; no tests or failure-feedback detail — static ceiling of 10 applies.
Trigger conditions are clear (/legal report-pdf after a review), prerequisites and input formats are specified, and it explicitly refuses to run without data. Deductions: compatibility boundaries with externally produced analysis files are undeclared; English-only with no Chinese support note; core function depends on Claude Code and PyPI, with mainland-China reachability risk undisclosed.
Documentation is well layered (phased, tabular), and the error-handling table is a strength. Deductions: license metadata unknown; no version or changelog; the script path inside the skill (scripts/generate_pdf_report.py) is inconsistent with README's generate_legal_pdf.py — a hidden assumption; maintenance ownership and update path unclear, and README skews toward marketing (Skool community funnel).
The Phase 3 output spec is detailed and concrete enough to yield a directly usable branded PDF, with a fallback if the gauge fails. Deductions: actual quality depends on LLM-written inline ReportLab code, unverifiable statically; '>10KB' is weak verification; usability of output without attorney review is limited; comparative-benefit evidence missing; static ceiling of 7 applies.
Only source-file specifications exist — no tests, CI evidence, real execution samples, or third-party corroboration. README statistics (82%, $10,000+ etc.) are unsourced marketing claims. Evidence types are thin under static review; 3 points.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →