Understanding the Output
Understanding the Output
Every contract scan produces a structured report. Here’s how to read it.
Output structure (always in this order)
Every scan follows the same structure. This consistency is intentional - it makes outputs easy to compare across contracts.
- ⚠️ Legal disclaimer
- 📋 Executive summary
- 📑 Contract metadata
- 🎯 Risk score and recommendation band
- 📊 Risk register
- ❌ Missing or ambiguous clauses
- ❓ Negotiation questions
- ✏️ Suggested redlines (Negotiation Prep mode only)
- 📧 Executive email draft (optional)
- 💾 Export artifacts
Section 1 - Legal disclaimer
Every output starts with:
⚠️ Not legal advice. This analysis is for issue-spotting and review preparation only. Consult qualified legal counsel before signing.
This is non-negotiable. Contract Risk Scanner accelerates review; it does not replace legal counsel.
Section 2 - Executive summary
A 3–5 sentence overview answering:
- What kind of contract is this?
- What’s the overall risk posture?
- What’s the top concern?
- Should this be escalated?
Section 3 - Contract metadata
A table of key facts extracted from the contract:
| Field | Example |
|---|---|
| Contract Type | SaaS MSA |
| Counterparty | Acme Software Inc. |
| Effective Date | 2026-01-15 |
| Term Length | 36 months |
| Renewal Type | Auto-renew, 60-day notice |
| Governing Law | Delaware |
| Total Contract Value | $180,000 |
“Not specified” means the field was missing or unclear in the source document. The agent never guesses.
Section 4 - Risk score and recommendation band
A 0–100 score with a recommendation:
| Score | Band | Recommendation |
|---|---|---|
| 0–25 | 🟢 Low | Safe to sign as-is |
| 26–50 | 🟡 Moderate | Negotiate flagged items |
| 51–75 | 🟠 High | Strongly recommend negotiation |
| 76–100 | 🔴 Critical | Escalate to legal |
The score is always shown with its breakdown so you can see what drove it:
Score breakdown: Liability (+25), Missing indemnification (+15), Auto-renewal (+10), Vague SLA (+8) = 58 (High)
📖 Read more: Custom Risk Registers
Section 5 - Risk register
The core of the output. A table with one row per identified risk:
| Column | Meaning |
|---|---|
| Area | Liability, IP, Data, Termination, Payment, SLA, etc. |
| Risk | One-sentence description |
| Severity | Low / Medium / High / Critical |
| Confidence | Low / Medium / High - how confident the agent is in this finding |
| Evidence | The contract quote (≤25 words) supporting this finding |
| Business Impact | Financial, operational, or reputational impact in plain language |
| Recommendation | Specific action to take |
💡 Pay attention to Confidence. A “High severity / Low confidence” finding may be a false positive worth a second look.
Section 6 - Missing or ambiguous clauses
Clauses the agent’s risk rules expected but couldn’t find in the contract:
- Missing: The clause is absent entirely
- Ambiguous: The clause is present but unclear, conflicting, or insufficient
If you uploaded a custom risk register, missing clauses reflect your custom required clauses.
Section 7 - Negotiation questions
5–10 questions to ask the counterparty, ranked by importance:
- Can liability be capped at 12 months of fees?
- Will you commit to 72-hour security incident notification?
- What is your subprocessor list?
These are designed to be ready to send.
Section 8 - Suggested redlines (Negotiation Prep mode only)
For each high-severity item:
Current language:
"Vendor shall have unlimited liability..."
Proposed redline:
"The total liability of either party shall not exceed
the fees paid in the twelve (12) months preceding..."
Rationale:
Caps financial exposure at predictable level; standard in market.
Section 9 - Executive email draft (optional)
If you ask, the agent will draft a 4–6 sentence email summarizing risk for leadership.
“Draft an exec email for this scan”
Section 10 - Export artifacts
Every scan offers downloadable files. See Exporting Results.
Confidence levels explained
| Confidence | What it means |
|---|---|
| High | Clear evidence in contract text matches a known risk pattern |
| Medium | Likely a risk, but language is ambiguous or context-dependent |
| Low | Possible risk, but worth human verification |
Always treat Low-confidence findings as prompts for human review, not conclusions.
Severity vs. Confidence
These are independent:
- High severity + High confidence = act on it
- High severity + Low confidence = verify before acting
- Low severity + High confidence = note for awareness
- Low severity + Low confidence = often ignorable