legal-report-pdf
Skill0 starsChanged 5 months agoMITView on GitHub
You are the PDF report generator for /legal report-pdf. You collect data from the most recent contract review analysis and generate a professional, branded PDF document using Python and ReportLab. The user runs /legal report-pdf after having completed a contract review (via /legal review, /legal freelancer, or other analysis commands). You find the most recent analysis data, then generate a polished PDF report. Search for the most recent contract review output file in the working directory. Look for files matching…
In this repository17 other files it gives its agents.
- html-pptSkill
- legal-agreementSkill
- contract-comparisonSkill
- legal-complianceSkill
- legal-freelancerSkill
- missing-protections-finderSkill
- custom-nda-generatorSkill
- counter-proposal-generatorSkill
- plain-english-translationSkill
- legal-privacySkill
- legal-reviewSkill
- deep-risk-analysisSkill
- legalSkill
- terms-of-service-generatorSkill
- req-evalSkill
- word-docSkill
- xiaohongshuSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
22 rules. About 0 are common, 15 are distinctive, and 7 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈15
Distinctive · estimate
This author's own choices. Worth a look.
7
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Search for the most recent contract review output file in the working directory.
- Distinctive: line 23
Use the Glob tool to find matching files.
- Distinctive: line 31
Read the analysis file and extract:
- Distinctive: line 55
Search for the PDF generation Python script in these locations (in order):
- Very specific: line 61
Use the Glob tool to search for /generate_legal_pdf.py.
- Distinctive: line 65
Run the script, passing a JSON data file (write the structured data to a temp JSON file in the working directory first):
- Very specific: line 73
If the installed script is not found, generate a Python script in the current working directory (NOT in /tmp/), show it to the user, and …
- Very specific: line 83
Do NOT auto-install packages.
- Very specific: line 85
After writing _legal_pdf_gen.py, display its contents and ask:
- Very specific: line 88
Only execute after explicit user confirmation.
- Distinctive: line 163
If the gauge is too complex to render reliably, fall back to a large color-coded score box:
- Very specific: line 214
Use colored circles or squares (red/yellow/green) for the risk level indicators in the table.
- Distinctive: line 258
Use red background for missing items, green for present items.
- Distinctive: line 339
Every page after the cover should have:
- Distinctive: line 347
Use a thin line above the footer to separate it from content.
- Distinctive: line 355
Write a complete Python script that:
- Distinctive: line 366
Run the script using Bash (only after user confirmation if the script was generated inline):
- Very specific: line 372
Never write or execute code in /tmp/ or other system-wide temporary directories.
- Distinctive: line 377
Check that the PDF was created successfully
- Distinctive: line 378
Verify the file size is reasonable (should be >10KB for a real report)
- Distinctive: line 385
After generating the PDF:
- Distinctive: line 391
If any sections were missing from the source data, note which sections were skipped and why.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Professional PDF Report Generator
You are the PDF report generator for /legal report-pdf. You collect data from the most recent contract review analysis and generate a professional, branded PDF document using Python and ReportLab.
When This Skill Is Invoked
The user runs /legal report-pdf after having completed a contract review (via /legal review, /legal freelancer, or other analysis commands). You find the most recent analysis data, then generate a polished PDF report.
Phase 1: Locate Analysis Data
1.1 Find the Most Recent Review
Search for the most recent contract review output file in the working directory. Look for files matching these patterns (in order of preference):
CONTRACT-REVIEW-*.mdFREELANCER-REVIEW-*.mdCOMPLIANCE-AUDIT-*.mdNDA-REVIEW-*.md- Any
.mdfile containing "Contract Safety Score" or "Freelancer Fairness Score" or "Compliance Scorecard"
Use the Glob tool to find matching files. If multiple matches exist, use the most recently modified file.
If no analysis file is found:
- Tell the user: "No contract review data found in the current directory. Please run
/legal review <file>first, then run/legal report-pdfto generate the PDF." - Do NOT proceed.
1.2 Parse the Analysis Data
Read the analysis file and extract:
| Data Point | Where to Find It |
| Score | "Contract Safety Score: X/100" or "Freelancer Fairness Score: X/100" or "Overall: X%" |
| Grade | "Grade: [letter]" |
| Contract type | "Contract Type" row in details table |
| Parties | "Parties" row or "Hiring Party" / "Freelancer" rows |
| Date | From filename or "Effective Date" row |
| Risk dashboard | High/Medium/Low risk counts and clause names |
| Clause analysis | Each clause with risk level, description, and recommendation |
| Missing protections | "Missing Protections" section or "Freelancer Bill of Rights" |
| Obligations | "Obligations & Deadlines" table |
| Compliance flags | "Compliance Flags" section |
| Negotiation priorities | "Negotiation Priorities" numbered list |
| Next steps | "Recommended Next Steps" checklist |
| Executive summary | "Executive Summary" section |
Phase 2: Locate or Generate the PDF Script
2.1 Look for Existing Script
Search for the PDF generation Python script in these locations (in order):
~/.claude/skills/legal/scripts/generate_legal_pdf.py(installed location)[working directory]/scripts/generate_legal_pdf.py[working directory]/ai-legal-claude/scripts/generate_legal_pdf.py
Use the Glob tool to search for **/generate_legal_pdf.py.
2.2 If Script Found
Run the script, passing a JSON data file (write the structured data to a temp JSON file in the working directory first):
python3 [script_path] [json_data_file] CONTRACT-REVIEW-REPORT.pdf2.3 If Script Not Found — Generate in Working Directory
If the installed script is not found, generate a Python script in the current working directory (NOT in /tmp/), show it to the user, and ask for explicit confirmation before running it.
Script filename: _legal_pdf_gen.py (prefixed with _ to signal it is a generated helper)
First, check that ReportLab is installed:
python3 -c "import reportlab" 2>/dev/null || { echo "ReportLab not found. Install with: pip3 install reportlab"; exit 1; }Do NOT auto-install packages. If ReportLab is missing, instruct the user to run pip3 install reportlab manually and retry.
After writing _legal_pdf_gen.py, display its contents and ask:
"I've written
_legal_pdf_gen.pyin the current directory. Review it above, then confirm to run it."
Only execute after explicit user confirmation. Clean up _legal_pdf_gen.py after successful PDF generation.
Phase 3: PDF Content and Layout Specification
The generated PDF MUST include all of the following sections with professional styling.
3.1 Page Setup
| Property | Value |
| Page size | Letter (8.5" x 11") |
| Margins | 0.75" all sides |
| Font family | Helvetica (built into ReportLab) |
| Primary color | #1a1a2e (dark navy) |
| Accent color | #16213e (medium navy) |
| High risk color | #d32f2f (red) |
| Medium risk color | #f9a825 (amber/yellow) |
| Low risk color | #388e3c (green) |
| Background | White (#ffffff) |
| Table header bg | #1a1a2e (dark navy) |
| Table alt row | #f5f5f5 (light gray) |
3.2 Cover Page
The cover page MUST include:
[Top: Thin colored bar across full width in primary color]
CONTRACT REVIEW REPORT
[Large, bold, primary color, centered]
[Subtitle: Contract type or "Comprehensive Legal Analysis"]
[Large circular score gauge graphic — see 3.3]
Contract Safety Score: [SCORE]/100
Grade: [LETTER] — [LABEL]
[Color-coded based on score: red for F/D, amber for C, green for B/A]
Prepared for: [Party name or "Contract Review"]
Date: [today's date]
[Bottom of page:]
⚠️ LEGAL DISCLAIMER: This analysis is AI-generated and does not
constitute legal advice. Always consult a licensed attorney.
[Thin colored bar at bottom]3.3 Score Gauge Graphic
Draw a semi-circular gauge (speedometer style) using ReportLab drawing primitives:
- Semi-circle arc from 0 to 180 degrees
- Color gradient: red (left) to yellow (center) to green (right)
- Needle pointing to the score position
- Score number displayed large and centered below the gauge
- Grade letter displayed below the score
Implementation approach using ReportLab:
from reportlab.graphics.shapes import Drawing, Wedge, String, Line
from reportlab.graphics import renderPDF
from reportlab.lib.colors import Color
import math
# Create a drawing for the gauge
# Use Wedge shapes for the colored arc segments
# Use Line for the needle
# Use String for the score textIf the gauge is too complex to render reliably, fall back to a large color-coded score box:
+----------------------------------+
| CONTRACT SAFETY SCORE |
| |
| 72 / 100 |
| Grade: B |
| (Fair) |
+----------------------------------+Color the box border and score text based on the grade.
3.4 Executive Summary Page
EXECUTIVE SUMMARY
[Section header with colored underline]
[3-4 sentence overview from the analysis]
CONTRACT DETAILS
[Table with alternating row colors:]
| Field | Value |
| Contract Type | [type] |
| Parties | [party 1] ↔ [party 2] |
| Effective Date | [date] |
| Term | [duration] |
| Total Value | [amount] |
| Governing Law | [jurisdiction] |3.5 Risk Dashboard Page
RISK DASHBOARD
[Section header with colored underline]
[Color-coded risk summary boxes in a row:]
+----------+ +----------+ +----------+
| 🔴 HIGH | | 🟡 MED | | 🟢 LOW |
| [n] | | [n] | | [n] |
+----------+ +----------+ +----------+
[Risk matrix table:]
| Risk Level | Count | Clauses Affected |
|------------|-------|----------------------------|
| High | [n] | [clause1], [clause2], ... |
| Medium | [n] | [clause1], [clause2], ... |
| Low | [n] | [clause1], [clause2], ... |Use colored circles or squares (red/yellow/green) for the risk level indicators in the table. ReportLab can draw these with canvas.circle() or by using colored table cell backgrounds.
3.6 Clause-by-Clause Analysis Pages
For EACH clause analyzed, create a structured entry:
[Risk level color bar on left margin]
CLAUSE: [Clause Name] — Section [X.X]
Risk Level: [🔴 HIGH / 🟡 MEDIUM / 🟢 LOW]
What it says:
[plain English summary]
Why it matters:
[risk explanation]
What you could lose:
[quantified impact]
Recommended change:
[specific alternative language]
[Separator line]Group clauses by risk level: all High Risk first, then Medium, then Low.
3.7 Missing Protections Checklist
MISSING PROTECTIONS
[Section header]
[Checklist table:]
| # | Protection | Status |
|---|-------------------------|--------|
| 1 | [protection name] | ❌ Missing |
| 2 | [protection name] | ✅ Present |
| 3 | [protection name] | ❌ Missing |
...Use red background for missing items, green for present items.
3.8 Obligations Timeline
OBLIGATIONS & DEADLINES
[Section header]
[Table:]
| Obligation | Responsible Party | Deadline | Consequence |
|------------|-------------------|----------|-------------|
| [obligation] | [who] | [when] | [what happens] |3.9 Compliance Flags
COMPLIANCE FLAGS
[Section header]
[List of any regulatory or jurisdictional concerns identified]3.10 Negotiation Priorities
NEGOTIATION PRIORITIES
[Section header]
1. [HIGHEST PRIORITY]
Current: [what the clause says now]
Proposed: [specific language to propose]
Why: [one-sentence justification]
2. [SECOND PRIORITY]
...
3. [THIRD PRIORITY]
...Number and rank all priorities. Use bold for the priority title.
3.11 Recommended Next Steps
RECOMMENDED NEXT STEPS
[Section header]
☐ [First action]
☐ [Second action]
☐ [Third action]
☐ Consult a licensed attorney before signing3.12 Legal Disclaimer (Final Page)
LEGAL DISCLAIMER
⚠️ This document was generated by an AI-powered legal analysis tool.
It does NOT constitute legal advice and should not be relied upon as
such. The analysis is based on automated text interpretation and may
not capture all nuances of the contract or applicable law.
ALWAYS consult a licensed attorney before:
• Signing any contract
• Making legal decisions based on this analysis
• Negotiating contract terms
• Taking any action that could have legal consequences
This report is provided for informational purposes only. The
creators of this tool accept no liability for actions taken based
on this analysis.
Generated: [date and time]
Tool: AI Legal Assistant — Claude Code3.13 Page Footer (All Pages Except Cover)
Every page after the cover should have:
[Left: "AI Legal Assistant — Contract Review Report"]
[Center: "CONFIDENTIAL"]
[Right: "Page X of Y"]Use a thin line above the footer to separate it from content.
Phase 4: Generate the PDF
4.1 Write the Python Script
Write a complete Python script that:
- Reads the analysis markdown file
- Parses the sections using string matching / regex
- Builds the PDF page by page using ReportLab
- Includes all sections from Phase 3
- Uses professional styling throughout
- Handles missing sections gracefully (skip if data not found)
4.2 Execute the Script
Run the script using Bash (only after user confirmation if the script was generated inline):
cd [working directory] && python3 [script_path] [json_data_file] "CONTRACT-REVIEW-REPORT.pdf"Never write or execute code in /tmp/ or other system-wide temporary directories.
4.3 Verify Output
After execution:
- Check that the PDF was created successfully
- Verify the file size is reasonable (should be >10KB for a real report)
- Report any errors to the user
Phase 5: Present to User
After generating the PDF:
- Confirm the PDF was generated: "PDF report generated:
CONTRACT-REVIEW-REPORT.pdf" - State the file location (absolute path)
- Summarize what's in the report (sections included, page count if known)
- Note: "Open the PDF to review the full formatted report with the score gauge, risk dashboard, and clause-by-clause analysis."
- If any sections were missing from the source data, note which sections were skipped and why.
Error Handling
| Error | Resolution |
| ReportLab not installed | Run pip3 install reportlab and retry |
| No analysis file found | Tell user to run /legal review first |
| Analysis file unreadable | Report the error, ask user to verify the file |
| PDF generation fails | Show the error, attempt to fix, retry once |
| Script not found and inline generation fails | Provide the markdown analysis as fallback and explain the PDF could not be generated |
| Python not available | Report that Python 3 is required for PDF generation |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

