legal-diligence
Legal Diligence
Legal due diligence is where hidden liabilities surface — or slip through. This skill automates the most labor-intensive legal DD tasks: NDA management, contract scanning, disclosure drafting, and red flag identification.
When to Use
- Redlining counterparty NDA markups in a sell-side auction
- Scanning material contracts for change-of-control triggers
- Drafting disclosure schedules for a purchase agreement
- Compiling a red flag due diligence report
The /redline-nda Protocol
How It Works
Relies on a strict, pre-programmed negotiation playbook:
- Ingests the counterparty's marked-up NDA
- Compares semantic differences against the firm's standard positions
- Accepts benign formatting changes
- Reverts substantive deviations requiring senior partner review
- Generates a redlined counter-draft with commentary
Common NDA Negotiation Points
| Clause | Standard Position | Common Counterparty Push |
|---|---|---|
| Confidentiality term | 24 months | Reduce to 12 months |
| Non-solicitation | 18-24 months | Eliminate or shorten |
| Permitted disclosure | Named advisors only | Expand to financing sources |
| Residuals clause | Exclude | Include |
| Standstill | Include | Remove or sunset |
Time saved: 1-2 hours per bidder in high-volume auction processes.
The /flag-coc Protocol
How It Works
Deploys a cumulative risk synthesis model — not simple keyword matching:
- Semantic understanding of assignment restrictions and transferability clauses
- Identifies clauses triggered by equity ownership changes
- Flags financial penalties, consent requirements, and termination rights
- Compiles an actionable risk register
Output Format
## Change-of-Control Risk Register
**Contracts Scanned**: [N]
**CoC Provisions Found**: [N]
**Critical (deal-threatening)**: [N]
**Moderate (consent required)**: [N]
**Low (notification only)**: [N]
### Critical Findings
| Contract | Counterparty | Provision | Impact | Required Action |
|----------|--------------|-----------|--------|-----------------|
| | | | Revenue at risk: $X | Third-party consent required |
### Consent Request Priority
1. [Highest-value contract requiring consent]
2. [Second priority]
...
Time saved: 20+ hours during legal DD. Identifies hidden liabilities that could threaten revenue continuity post-close.
The /schedule-disclosure Protocol
How It Works
- Extracts data from VDR to populate initial schedule drafts
- Maps disclosures to specific SPA/APA sections (e.g., Section 3.14 Material Contracts)
- Cross-references for internal consistency against the operative agreement
- Flags potential omissions
Common Disclosure Schedules
| Schedule | Source Data | Threshold |
|---|---|---|
| Material Contracts | All contracts | >$250K annual value |
| Capitalization | Equity grants, warrants, options | All outstanding |
| Litigation | Pending/threatened claims | All |
| Employee Benefits | Plan documents | All plans |
| Real Property | Leases, owned property | All |
| IP Registrations | Patents, trademarks, copyrights | All registered |
⚠️ AI Limitations: LLMs cannot independently verify the economic substance of related-party transactions or interpret subjective legal standards like "Material Adverse Effect." AI outputs in definitive legal agreements must remain strictly advisory, requiring rigorous human validation.
Time saved: 12+ hours of manual drafting per transaction.
Red Flag Due Diligence Report
Key Sections
| Section | Focus |
|---|---|
| Executive Risk Summary | Prioritized dashboard by severity and valuation impact |
| Corporate Governance & Capitalization | Unresolved equity claims, missing consents, invalid 409A valuations |
| Material Contracts | Onerous CoC provisions, exclusivity, single-supplier concentration |
| Intellectual Property | Unsecured chain of title, open-source contamination, infringement |
| Regulatory & Compliance | Pending investigations, data privacy exposure, environmental liabilities |
| Labor & Employment | Contractor misclassification, unfunded pensions, key-person flight risk |
AI-Assisted Workflow
The AI continually scans the VDR as documents are uploaded, comparing against a predefined database of critical transactional risks. When anomalies are detected — missing IP assignment clauses, undocumented equity issuances, broken cap tables — they immediately populate the Red Flag Report template.
Operating Guidelines
- NDA redlining: always escalate novel or bespoke clauses to senior counsel
- CoC scanning: go beyond keywords — use semantic understanding for bespoke legal phrasing
- Disclosure schedules: err on the side of over-disclosure to minimize post-close indemnification risk
- Red flag reports: focus on exceptions, not comprehensive cataloguing
- All legal outputs are advisory — always require qualified legal review
Troubleshooting
| Problem | Cause | Fix |
|---|---|---|
| Change-of-control clauses missed | Buried in schedules | Run keyword scan on full contract corpus, not just main body |
| NDA redline rejected by counterparty | Overly aggressive positions | Use tiered playbook: "must-haves" vs. "nice-to-haves"; escalate only on must-haves |
| Disclosure schedules incomplete | Seller's counsel rushed | Provide a disclosure schedule checklist at LOI stage; allow 2-week cure period |
| Material contracts not provided | VDR gaps | Cross-reference audited financials for material contract obligations; request missing items |
More from lauraflorentin/skills-marketplace
multi-agent-collaboration
A structural pattern where multiple specialized agents communicate and coordinate to solve a problem that is too complex for a single agent. Use when user asks to "build a multi-agent system", "agents working together", "agent collaboration", or mentions team of agents, distributed agents, or swarm.
21reflection
A recursive pattern where an agent evaluates and critiques its own output to iteratively improve quality and catch errors. Use when user asks to "add self-reflection", "agent introspection", "self-critique", or mentions self-evaluation, meta-cognition, or quality self-assessment.
18human-in-the-loop
A hybrid pattern where the system pauses execution to request human approval, input, or disambiguation before proceeding with critical actions. Use when user asks to "add human approval", "require human review", "human-in-the-loop", or mentions approval workflows, human oversight, or escalation.
16planning
A high-level cognitive pattern where an agent formulates a structured sequence of actions (a plan) before executing any of them, ensuring goal-directed behavior. Use when user asks to "add planning to my agent", "task planning", "agent planning", or mentions plan generation, plan execution, or step-by-step planning.
14parallelization
A concurrency pattern where multiple agent tasks are executed at the same time to speed up processing or gather diverse perspectives. Use when user asks to "run agents in parallel", "parallelize tasks", "concurrent execution", or mentions parallel processing, fan-out, or batch execution.
13routing
A control flow pattern where a central component classifies an input request and directs it to the most appropriate specialized agent or tool. Use when user asks to "route between agents", "agent routing", "task dispatch", or mentions classifier routing, intent detection, or agent selection.
12