Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generates customized business agreements for 10 common relationship types with plain English annotations. Use when formalizing a business relationship, creating a partnership agreement, or drafting a service contract from scratch. Trigger with "/agreement-generator" or "create a freelancer agreement".
.claude/skills/jeremylongshore-agreement-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 120% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 102% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 118% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 163% | 0% |
Generates professional business agreements for 10 common relationship types, each with type-specific clause sections, plain English annotations, and proper legal structure. Templates are benchmarked against CommonPaper standards (CC BY 4.0), Bonterms (CC BY 4.0), and Open-Agreements (MIT) to ensure market-standard language.
The skill uses an information-gathering wizard approach — collecting essential details before generating, rather than producing generic boilerplate that requires heavy editing.
> Legal Disclaimer: This skill generates template documents for informational and > educational purposes only. Generated agreements are not a substitute for legal advice. > Contract requirements vary by jurisdiction, industry, and transaction specifics. > All documents should be reviewed by a licensed attorney before execution. No > attorney-client relationship is created by using this tool.
| Type | When to Use | Key Focus | |------|-------------|-----------| | Freelancer | Hiring independent contractors | Deliverables, IP, independent contractor status | | Partnership | Forming a business partnership | Profit sharing, decision authority, exit | | NDA | Protecting confidential info | Scope, duration, remedies (use nda-generator for full NDA) | | Licensing | Granting IP usage rights | Grant scope, royalties, exclusivity | | Consulting | Engaging expert advisors | Scope of work, deliverables, hourly/project fees | | Statement of Work (SOW) | Defining project specifics | Milestones, acceptance criteria, change orders | | Master Service Agreement (MSA) | Ongoing service relationship | Framework terms, SOW attachment structure | | Joint Venture | Temporary business collaboration | Contributions, profit split, governance, dissolution | | Distribution | Product distribution rights | Territory, exclusivity, minimum orders, marketing | | Referral | Formalizing referral partnerships | Commission structure, tracking, payment triggers |
Universal fields (all types):
Type-specific fields:
required sections. Common sections across all types:
| Section | Included In | |---------|-------------| | Recitals & Definitions | All types | | Scope of Work / Services | Freelancer, Consulting, SOW, MSA | | Compensation & Payment | All types | | Intellectual Property | Freelancer, Consulting, Licensing, JV | | Confidentiality | All types | | Representations & Warranties | All types | | Indemnification | All types | | Limitation of Liability | All types | | Term & Termination | All types | | Non-Compete / Non-Solicit | Freelancer, Partnership, Consulting, JV | | Dispute Resolution | All types | | General Provisions | All types | | Signature Block | All types |
Type-specific sections:
> **Plain English:** {simple explanation of what this means for both parties}
or relationship specifics not explicitly provided by the user.
Generate a single Markdown file named {TYPE}-AGREEMENT-{PartyA}-{PartyB}-{YYYY-MM-DD}.md:
# {Type} Agreement
**Between:** {Party A} ("{Role A}")
**And:** {Party B} ("{Role B}")
**Effective Date:** {date}
---
## Table of Contents
{numbered section list}
---
## 1. Recitals and Definitions
{formal legal text}
> **Plain English:** {simple explanation}
## 2. {Type-Specific Section}
{formal legal text}
> **Plain English:** {simple explanation}
{... remaining sections ...}
---
## Signature Block
| | {Party A} | {Party B} |
|---|-----------|-----------|
| Signature | _________________ | _________________ |
| Name | {name} | {name} |
| Title | {title} | {title} |
| Date | _________________ | _________________ |
---
**[VERIFY] Tags Summary:**
{numbered list of assumptions}
**Agreement Type:** {type}
**Clause Count:** {count} sections
**Generated by:** Legal Assistant Plugin — Not a substitute for legal counsel.| Error | Cause | Solution | |-------|-------|----------| | Unclear agreement type | User describes a hybrid relationship | Recommend the closest type, note deviations | | Missing payment details | User has not set compensation terms | Provide market-rate ranges for context, add VERIFY] | | Multi-party agreement | More than 2 parties | Adapt signature block and obligations for all parties | | International parties | Cross-border relationship | Add international arbitration clause, address currency and tax | | Regulated industry | Healthcare, finance, government contracting | Flag additional compliance requirements (HIPAA, SOX, FAR) | | California freelancer | AB 5 independent contractor risks | Include ABC test analysis, recommend legal review |
Example 1: Freelancer Agreement
Request: "Create a freelancer agreement for a web developer building our new marketing site"
Result: FREELANCER-AGREEMENT-AcmeCorp-JaneDev-2026-04-02.md with:
Example 2: Master Service Agreement
Request: "Generate an MSA between our consulting firm and a new enterprise client"
Result: MSA-AGREEMENT-ConsultingCo-EnterpriseCorp-2026-04-02.md with:
Example 3: Referral Agreement
Request: "Create a referral agreement — we'll pay 10% commission for qualified leads"
Result: REFERRAL-AGREEMENT-CompanyA-PartnerB-2026-04-02.md with:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→pass | 21,565 | 32,461 | +51% | 1 | 1 | 0% | 3,790 | 8,354 | +120% | 0 | 0 | — |
case-06 | fail→pass | 19,749 | 27,180 | +38% | 1 | 1 | 0% | 3,690 | 7,456 | +102% | 0 | 0 | — |
case-07 | pass→pass | 20,246 | 26,811 | +32% | 1 | 1 | 0% | 3,956 | 7,510 | +90% | 0 | 0 | — |
case-01 | fail→pass | 23,450 | 27,470 | +17% | 1 | 1 | 0% | 4,714 | 7,557 | +60% | 0 | 0 | — |
case-02 | fail→fail | 21,742 | 31,467 | +45% | 1 | 1 | 0% | 4,458 | 8,398 | +88% | 0 | 0 | — |
case-03 | pass→pass | 13,758 | 8,406 | -39% | 1 | 1 | 0% | 2,511 | 3,701 | +47% | 0 | 0 | — |
case-04 | fail→fail | 18,042 | 28,322 | +57% | 1 | 1 | 0% | 3,662 | 7,836 | +114% | 0 | 0 | — |
case-08 | pass→pass | 13,639 | 28,165 | +107% | 1 | 1 | 0% | 2,722 | 7,546 | +177% | 0 | 0 | — |
case-09 | fail→fail | 14,095 | 13,317 | -6% | 1 | 1 | 0% | 2,519 | 4,485 | +78% | 0 | 0 | — |
case-10 | pass→pass | 15,304 | 14,329 | -6% | 1 | 1 | 0% | 2,750 | 4,658 | +69% | 0 | 0 | — |
case-11 | pass→pass | 19,826 | 30,629 | +54% | 1 | 1 | 0% | 3,875 | 8,280 | +114% | 0 | 0 | — |
case-17 | fail→pass | 19,101 | 32,831 | +72% | 1 | 1 | 0% | 3,803 | 8,274 | +118% | 0 | 0 | — |
case-12 | pass→pass | 15,230 | 12,683 | -17% | 1 | 1 | 0% | 2,529 | 4,389 | +74% | 0 | 0 | — |
case-13 | fail→pass | 15,206 | 23,480 | +54% | 1 | 1 | 0% | 2,512 | 6,602 | +163% | 0 | 0 | — |
case-14 | fail→fail | 11,672 | 10,076 | -14% | 1 | 1 | 0% | 1,921 | 3,825 | +99% | 0 | 0 | — |
case-15 | fail→pass | 12,599 | 10,095 | -20% | 1 | 1 | 0% | 2,081 | 3,973 | +91% | 0 | 0 | — |
case-16 | pass→pass | 12,968 | 22,827 | +76% | 1 | 1 | 0% | 2,507 | 6,543 | +161% | 0 | 0 | — |
case-18 | fail→fail | 16,493 | 28,446 | +72% | 1 | 1 | 0% | 3,001 | 7,471 | +149% | 0 | 0 | — |
case-19 | fail→fail | 13,950 | 20,254 | +45% | 1 | 1 | 0% | 2,857 | 5,815 | +104% | 0 | 0 | — |
case-20 | fail→fail | 16,290 | 21,719 | +33% | 1 | 1 | 0% | 2,818 | 6,221 | +121% | 0 | 0 | — |
case-21 | fail→fail | 12,681 | 20,447 | +61% | 1 | 1 | 0% | 2,295 | 5,878 | +156% | 0 | 0 | — |
case-22 | fail→fail | 19,730 | 28,591 | +45% | 1 | 1 | 0% | 3,741 | 7,740 | +107% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +27 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.