メニュー
Deep clause-by-clause NDA review from Recipient or Discloser perspective. Produces issue log with redlines, fallbacks, rationales, owners, deadlines. Use when reviewing NDAs for negotiation or approval.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/ThomasMoreAI/legal-skills-open --skill nda-review-borgheiコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Use when retrieving arbitration institutional rules (ICC, LCIA, SCC, SIAC, HKIAC, VIAC, МКАС/МАК при ТПП України, UNCITRAL) — fetching current version, verifying redaction applicable to the date of arbitration agreement, constructing URLs for official rule texts
Use when choosing the Polish legal regime for letters, requests, applications, complaints, petitions, public-information requests, KPA filings, PPSA complaints, RODO access requests, registry extracts, court-file access, tax/ZUS/cudzoziemcy/USC procedures, or professional lawyer letters. Prevents mixing UDIP, KPA, PPSA, RODO, registry, special-procedure, and advocate/radca letter regimes.
Use when preparing applications for recognition and enforcement of foreign arbitral awards in Poland, applications for setting aside arbitral awards under KPC art. 1205–1211, or opposing such applications — mapping Article V of the 1958 New York Convention to art. 1214–1215 of the Polish KPC, identifying grounds for refusal, structuring public policy arguments
SKILL.md を表示中
SOC 職業分類に基づく
| name | nda-review-borghei |
| title | NDA Review |
| description | Deep clause-by-clause NDA review from Recipient or Discloser perspective. Produces issue log with redlines, fallbacks, rationales, owners, deadlines. Use when reviewing NDAs for negotiation or approval. |
| author | borghei |
| author_url | https://github.com/borghei/Claude-Skills/tree/main/legal/nda-review |
| license | MIT |
| version | 0.1.0 |
| execution_mode | open |
| jurisdiction | general |
| practice | contracts |
| language | en |
⚠️ EXPERIMENTAL — This skill is provided for educational and informational purposes only. It does NOT constitute legal advice. All responsibility for usage rests with the user. Consult qualified legal professionals before acting on any output.
Deep clause-by-clause NDA review tool that analyzes agreements from Recipient or Discloser perspective. Produces structured issue logs with preferred redlines, fallback positions, rationale, owners, and deadlines.
Performs deep analysis of NDA text, extracting and classifying each clause against best practices. Detects overbroad definitions, missing carveouts, problematic residuals, IP grants, indemnification, and audit rights.
# Review from recipient perspective (default)
python scripts/nda_clause_reviewer.py nda_draft.txt
# Review from discloser perspective
python scripts/nda_clause_reviewer.py nda_draft.txt --perspective discloser
# JSON output for integration
python scripts/nda_clause_reviewer.py nda_draft.txt --json
# Save issue log
python scripts/nda_clause_reviewer.py nda_draft.txt --output issues.json --json
What it produces:
references/nda_clause_reference.md
Five deep reference modules:
references/nda_review_templates.md
Output templates and worked examples:
nda-triage skill for quick GREEN/YELLOW/RED classificationnda_clause_reviewer.py with appropriate --perspective| Perspective | Focus Areas | Key Concerns |
|---|---|---|
| Recipient | Scope of obligations, carveouts, residuals, return/destruction | Protecting freedom to operate; avoiding contamination claims |
| Discloser | Definition breadth, remedies, duration, permitted disclosures | Maximizing protection; ensuring adequate enforcement |
Stop review and escalate if any of these 7 red flags are present.
| # | Red Flag | Why It Matters | Escalation |
|---|---|---|---|
| 1 | Non-compete clause | Restricts business operations; requires separate consideration and analysis | Senior counsel immediately |
| 2 | IP assignment or license grant | Transfers rights beyond confidentiality scope | Senior counsel immediately |
| 3 | Non-solicitation of employees or customers | Employment law implications; may be unenforceable | Senior counsel within 24 hours |
| 4 | Missing 3+ standard carveouts | Fundamentally deficient NDA | Counsel review before any response |
| 5 | Liquidated damages or penalty clause | Transforms NDA into penalty contract | Senior counsel within 24 hours |
| 6 | Perpetual obligations with no termination | Indefinite legal burden with no exit | Counsel review within 48 hours |
| 7 | Exclusivity provision | Limits engagement with other parties | Business leadership + counsel |
| # | Topic | Key Questions | Risk if Missing |
|---|---|---|---|
| 1 | Definition Scope | Is confidential info bounded? Is there a marking requirement? | Overbroad definition traps all shared information |
| 2 | Standard Carveouts | Are all 5 carveouts present and properly drafted? | Missing carveouts restrict legitimate business activities |
| 3 | Permitted Use | Is use restricted to stated purpose? Can we share with advisors? | Overly restrictive use limits may impede evaluation |
| 4 | Residuals | Is there a residuals clause? Is it narrow or broad? | Broad residuals clause benefits; narrow or absent protects discloser |
| 5 | Return/Destruction | Return or destroy option? Retention exception for backups? | No retention exception is impractical for electronic data |
| 6 | Term & Survival | Reasonable term? Reasonable survival period? Termination right? | Perpetual obligations are burdensome |
| 7 | Remedies | Injunctive relief only? Or liquidated damages/indemnification? | Excessive remedies shift risk disproportionately |
| 8 | Problematic Provisions | Non-compete? Non-solicitation? IP assignment? Audit rights? | These provisions have no place in a standard NDA |
| # | Topic | Key Questions | Risk if Missing |
|---|---|---|---|
| 1 | Definition Breadth | Does definition cover all information we will share? All forms? | Gaps in definition leave information unprotected |
| 2 | Obligation Strength | Standard of care adequate? Written agreements from recipients? | Weak obligations increase risk of unauthorized disclosure |
| 3 | Remedies | Injunctive relief available? Is it meaningful in this jurisdiction? | Without adequate remedies, NDA is unenforceable in practice |
| 4 | Duration | Is the term long enough? Does survival cover our exposure window? | Short terms may expire before information loses value |
| 5 | Recipient Limits | Who can receive? Is need-to-know enforced? Downstream binding? | Unrestricted sharing exposes information to unauthorized parties |
Different NDA contexts require different review emphasis.
| Additional Concern | Reason | Recommended Position |
|---|---|---|
| Standstill provision | Prevents hostile acquisition moves during due diligence | Accept if mutual and time-limited (12-18 months) |
| Non-solicitation of employees | Standard in M&A NDAs | Accept if limited to key employees for 12 months |
| Broader definition | M&A requires extensive information sharing | Accept broader definition with strong carveouts |
| Longer survival | Sensitive strategic information shared | 3-5 year survival is appropriate |
| Residuals clause sensitivity | Competitive intelligence at stake | Resist residuals clause or narrow significantly |
| Additional Concern | Reason | Recommended Position |
|---|---|---|
| Invention assignment | Employer IP ownership | Separate from NDA; use invention assignment agreement |
| Post-employment obligations | Obligations after employment ends | Limit survival to 2 years; ensure enforceability |
| Scope of work product | What the employee creates | Define in employment agreement, not NDA |
| Non-compete enforceability | Varies by jurisdiction | Review local law before including; may be void |
| Additional Concern | Reason | Recommended Position |
|---|---|---|
| Investor portfolio conflicts | VC may have portfolio companies in same space | Include portfolio company exclusion or conflict provision |
| Residuals clause | VCs see many similar pitches | Resist; protect trade secrets and specific data |
| Term limitations | VCs want short obligations | 2-3 year term acceptable; ensure adequate survival |
| Definition scope | Founders want maximum protection | Balance with investor need for portfolio flexibility |
| Rating | Criteria | Action | Timeline |
|---|---|---|---|
| HIGH (H) | Could result in material legal or financial exposure; deal-breaker potential | Must resolve before signing | Pre-signing |
| MEDIUM (M) | Creates meaningful risk but manageable; strong preference to resolve | Should resolve; accept with documented risk if necessary | Within 30 days |
| LOW (L) | Minor preference; improves agreement but not material | Nice to resolve; concede if needed for higher-priority wins | Within 90 days |
| Issue Type | Typical Rating | Escalation |
|---|---|---|
| Missing carveout (any) | M-H | Counsel |
| Overbroad definition | M | Counsel |
| Non-compete/non-solicitation | H | Senior counsel |
| IP assignment | H | Senior counsel |
| Residuals clause (broad) | M | Counsel |
| Perpetual obligations | M-H | Counsel |
| No return/destruction | M | Counsel |
| Liquidated damages | H | Senior counsel |
| Missing governing law | L-M | Counsel |
| One-sided obligations | M | Counsel |
| Pitfall | Impact | Fix |
|---|---|---|
| Reviewing without knowing your perspective | Recipient and discloser have opposing interests on many clauses | Always set --perspective flag; review with clear role in mind |
| Treating the NDA as "just a formality" | Missing problematic provisions that create real obligations | Run full clause review on every NDA, regardless of perceived importance |
| Negotiating clause-by-clause in document order | Wastes time on early low-priority clauses; may not reach critical issues | Prioritize by risk rating; address H items first |
| Accepting "standard" NDAs without review | Every organization's "standard" is different; one party's standard favors that party | No NDA is truly standard; always review |
| Ignoring context (M&A, employment, VC) | Standard NDA review misses context-specific risks | Use variation callouts for specialized contexts |
| Not preparing fallback positions | Stuck when counterparty rejects preferred redline | Prepare preferred + fallback for every H and M item |
| Signing before resolving H-rated issues | Creates material legal exposure | Require all H items resolved or executive sign-off |
| Problem | Cause | Solution |
|---|---|---|
| All issues rated LOW | NDA is genuinely well-drafted, or text extraction lost key sections | Manually verify critical sections (definition, carveouts, remedies) are in the input file |
| Perspective flag has no effect | Tool adjusts weighting, not detection; same issues found either way | Perspective changes risk ratings and recommendations, not issue detection |
| Too many issues generated | NDA is non-standard or poorly drafted | Focus on H-rated issues first; use the issue log as a negotiation roadmap |
| Script misses embedded provisions | Non-compete or IP clause hidden in definitions or general provisions | Search full document for "compete", "assign", "license", "solicit" manually |
| Output format does not match template | Tool outputs structured data, not final deliverable | Use references/nda_review_templates.md to format the output for stakeholders |
Covers:
Does NOT cover:
nda-triage for quick GREEN/YELLOW/RED screeningcontract-review for general commercial agreements| Anti-Pattern | Why It Fails | Better Approach |
|---|---|---|
| Running deep review without triage first | Wastes time on detailed analysis of NDAs that should be rejected outright (RED triage) | Always run nda-triage first; only proceed to deep review for YELLOW or GREEN-with-complexity |
| Using Recipient perspective for both sides | Recipient perspective minimizes obligations and maximizes carveouts, which is wrong if you are the discloser | Always set the correct --perspective flag based on your role |
| Accepting all LOW-rated issues without review | Some LOW issues are low-risk individually but create cumulative exposure when combined | Review the full issue log for interaction effects; multiple LOW issues in the same area may compound to MEDIUM |
| Skipping the variation callouts for specialized contexts | Standard NDA review misses M&A standstill provisions, employment invention assignment, VC portfolio conflicts | Check the variation callouts section for your specific deal context |
Purpose: Performs deep clause-by-clause NDA analysis. Detects overbroad definitions, missing carveouts, problematic provisions, and generates an issue log with redlines, fallbacks, rationale, owners, and deadlines.
Usage:
python scripts/nda_clause_reviewer.py <nda_file> [--perspective PERSPECTIVE] [--json] [--output FILE]
Flags:
| Flag | Short | Default | Description |
|---|---|---|---|
nda_file | (positional) | Path to NDA text file (.txt or .md) | |
--perspective | -p | recipient | Review perspective: recipient or discloser |
--json | off | Output in JSON format | |
--output | -o | (stdout) | Write output to file |
Example Output (JSON):
{
"file": "vendor_nda.txt",
"perspective": "recipient",
"issues": [
{
"id": 1,
"clause": "Definition of Confidential Information",
"issue": "Overbroad definition with no marking requirement",
"risk": "H",
"preferred_redline": "Narrow to information marked Confidential or confirmed in writing within 10 days",
"fallback": "Add marking requirement for written; 10-day confirmation for oral",
"rationale": "Overbroad definition traps all shared information as confidential",
"owner": "legal",
"deadline": "pre-signing"
}
],
"summary": {
"total_issues": 5,
"high": 2,
"medium": 2,
"low": 1
}
}
Example Output (Text):
NDA CLAUSE REVIEW — ISSUE LOG
==============================
File: vendor_nda.txt
Perspective: Recipient
Issues Found: 5 (H:2 M:2 L:1)
# Risk Clause Issue
1 H Definition of Confidential Info Overbroad definition; no marking requirement
Preferred: Narrow to marked information with 10-day oral confirmation
Fallback: Add marking requirement for written; 10-day confirmation for oral
Rationale: Overbroad definition traps all shared information
Owner: legal | Deadline: pre-signing
2 H Standard Carveouts Missing independent development carveout
Preferred: Add standard independent development exception
Fallback: Add with documentary evidence requirement
Rationale: Missing carveout blocks internal R&D
Owner: legal | Deadline: pre-signing