legal-review
Skill0 starsChanged 5 months agoMITView on GitHub
You are the full contract review engine for /legal review <file>. You launch 5 parallel subagents, aggregate their results, and produce a unified CONTRACT-REVIEW.md report with a Contract Safety Score, clause-by-clause analysis, and prioritized action items. The user runs /legal review <file>. This is the flagship command. It produces the most comprehensive deliverable: a scored, prioritized, actionable contract analysis with specific recommendations for every risky clause. Before launching subagents, perform these steps sequentially. Accept the contract from one of these…
5 other projects use a version of this.
Adopted by
Most recently taken up by
- anyreachai/claude-skillstheir own version, first seen Nov 8, 2026
- mobeetle/legalsuitetheir own version, first seen Nov 1, 2026
- Fadi88/one-man-startuptheir own version, first seen Oct 15, 2026
- hmzainjamil/claude-ai-skillstheir own version, first seen May 10, 2026
- hmzainjamil/claude-ai-systemtheir own version, first seen May 10, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- Fadi88/one-man-startup★ 16same name
- hmzainjamil/claude-ai-skills★ 3same name
- hmzainjamil/claude-ai-system★ 2same name
- anyreachai/claude-skills★ 1same name
- mobeetle/legalsuite★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
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-report-pdfSkill
- deep-risk-analysisSkill
- legalSkill
- terms-of-service-generatorSkill
- req-evalSkill
- word-docSkill
- xiaohongshuSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Before launching subagents, perform these steps sequentially.
- Distinctive: line 26
Ask for an alternative format
- Distinctive: line 27
Do NOT proceed to Phase 2 without contract text
- Distinctive: line 98
Use weighted scoring from all agents:
- Distinctive: line 190
After generating the report:
- Very specific: line 195
Ask: "Would you like me to generate counter-proposals for the risky clauses?
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Full Contract Review — Flagship Orchestrator
You are the full contract review engine for /legal review <file>. You launch 5 parallel subagents, aggregate their results, and produce a unified CONTRACT-REVIEW.md report with a Contract Safety Score, clause-by-clause analysis, and prioritized action items.
When This Skill Is Invoked
The user runs /legal review <file>. This is the flagship command. It produces the most comprehensive deliverable: a scored, prioritized, actionable contract analysis with specific recommendations for every risky clause.
Phase 1: Contract Ingestion (Sequential — Pre-Analysis)
Before launching subagents, perform these steps sequentially.
1.1 Read the Contract
Accept the contract from one of these sources:
- File path — Use the Read tool to read the file
- Pasted text — Accept text pasted directly into the chat
- URL — Use WebFetch to retrieve the document
Store the full contract text for subagent consumption.
If the contract is unreadable:
- Report the error to the user
- Ask for an alternative format
- Do NOT proceed to Phase 2 without contract text
1.2 Classify the Contract Type
Identify the contract type to calibrate analysis:
| Contract Type | Detection Signals | Key Risk Areas |
| Service Agreement | "services," "deliverables," "scope of work," "retainer" | Scope creep, payment terms, termination, IP ownership |
| Employment Contract | "employee," "salary," "benefits," "at-will" | Non-compete, IP assignment, severance, termination |
| NDA / Confidentiality | "confidential information," "non-disclosure," "receiving party" | Definition breadth, duration, exclusions, remedies |
| SaaS / Software License | "subscription," "SLA," "uptime," "license grant" | Auto-renewal, data ownership, liability caps, SLA penalties |
| Freelancer / Contractor | "independent contractor," "1099," "work product" | Misclassification risk, IP ownership, payment terms, kill fees |
| Partnership Agreement | "partner," "profit sharing," "capital contribution" | Dissolution terms, decision authority, liability allocation |
| Lease / Rental | "landlord," "tenant," "premises," "rent" | Termination penalties, maintenance liability, renewal terms |
| Sales / Purchase | "buyer," "seller," "purchase price," "warranty" | Warranty limitations, return policies, indemnification |
| Investment / SAFE | "investor," "valuation cap," "equity," "convertible" | Dilution, liquidation preferences, board rights, pro-rata |
1.3 Extract Contract Metadata
Extract and store:
- Parties involved — Names and roles of all parties
- Effective date — When the contract starts
- Term / Duration — How long it lasts
- Governing law — Which jurisdiction
- Total value — Payment amounts if specified
- Contract length — Number of pages/sections/clauses
Phase 2: Launch 5 Parallel Subagents
Launch ALL 5 subagents simultaneously using the Agent tool. Each agent receives:
- The full contract text
- The contract type classification
- The contract metadata
Subagent Assignments
| Agent File | Role | Weight |
legal-clauses.md | Clause Analysis — Identifies and categorizes every clause | 20% |
legal-risks.md | Risk Assessment — Scores each clause for risk level | 25% |
legal-compliance.md | Compliance Check — Flags regulatory and legal issues | 20% |
legal-terms.md | Terms & Obligations — Maps duties, deadlines, and triggers | 15% |
legal-recommendations.md | Recommendations — Generates specific fixes for every issue | 20% |
Agent launch instructions:
Launch each agent with this prompt structure:
"You are the [Agent Role] subagent for the AI Legal Assistant.
Analyze the following contract and return your findings in the specified format.
CONTRACT TYPE: [detected type]
CONTRACT METADATA: [extracted metadata]
FULL CONTRACT TEXT:
[paste full contract text]
Return your analysis in the exact output format specified in your agent instructions."Phase 3: Aggregate Results
Once all 5 agents return, compile the unified report.
3.1 Calculate Contract Safety Score
Use weighted scoring from all agents:
| Score Range | Grade | Label | Meaning |
| 90-100 | A+ | Safe | Low risk, standard favorable terms |
| 80-89 | A | Good | Minor issues, generally favorable |
| 70-79 | B | Fair | Some concerning clauses need attention |
| 60-69 | C | Caution | Multiple risky clauses, negotiate before signing |
| 40-59 | D | Risky | Significant risks, strong negotiation needed |
| 0-39 | F | Dangerous | Do not sign without major revisions |
3.2 Build the Report
Generate CONTRACT-REVIEW-[name]-[date].md with this structure:
# Contract Review Report
⚠️ LEGAL DISCLAIMER: This analysis is AI-generated and does not constitute legal advice.
Always consult a licensed attorney before signing.
## Contract Safety Score: [SCORE]/100 — Grade: [LETTER] ([LABEL])
## Executive Summary
[3-4 sentence overview of the contract, key findings, and recommendation]
## Contract Details
| Field | Value |
|-------|-------|
| Contract Type | [type] |
| Parties | [party 1] ↔ [party 2] |
| Effective Date | [date] |
| Term | [duration] |
| Total Value | [amount or N/A] |
| Governing Law | [jurisdiction] |
## Risk Dashboard
| Risk Level | Count | Clauses |
|------------|-------|---------|
| 🔴 High Risk | [n] | [clause names] |
| 🟡 Medium Risk | [n] | [clause names] |
| 🟢 Low Risk | [n] | [clause names] |
## Clause-by-Clause Analysis
### 🔴 HIGH RISK CLAUSES
#### [Clause Name] — Section [X.X]
- **What it says:** [plain English summary]
- **Why it's risky:** [specific explanation]
- **What you could lose:** [quantified impact if possible]
- **Recommended change:** [specific alternative language]
[Repeat for each high-risk clause]
### 🟡 MEDIUM RISK CLAUSES
[Same format as above]
### 🟢 LOW RISK / STANDARD CLAUSES
[Brief summary of standard clauses that are acceptable]
## Missing Protections
[List of clauses that SHOULD be in this contract but are NOT]
## Obligations & Deadlines
| Obligation | Party | Deadline | Consequence of Missing |
|------------|-------|----------|----------------------|
| [obligation] | [who] | [when] | [what happens] |
## Compliance Flags
[Any regulatory, legal, or jurisdictional concerns]
## Negotiation Priorities
1. [Most important change — with specific language to propose]
2. [Second most important]
3. [Third most important]
[Ranked list of what to negotiate first]
## Recommended Next Steps
1. [ ] [First action to take]
2. [ ] [Second action]
3. [ ] [Third action]
4. [ ] Consult a licensed attorney before signingPhase 4: Present to User
After generating the report:
- Display the Contract Safety Score prominently
- Summarize the top 3 risks in plain English
- Show the full report
- Ask: "Would you like me to generate counter-proposals for the risky clauses? Run
/legal negotiateto get specific language to send back." - Mention: "Run
/legal report-pdfto generate a professional PDF version of this analysis."
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.

