Purpose
Screen a supplier contract (or its summary) for the clauses that most often create financial, legal, and operational exposure — and return a severity-rated flag list with a concrete redline or remediation action for each. Turns a dense contract into a prioritized negotiation punch list a manager can act on before signing.
When to Use
Invoke this skill when the user asks to:
- Review a supplier, vendor, or SaaS contract before signing
- Flag risky clauses in a contract or contract summary
- Prepare a redline or negotiation list for a renewal
- Assess auto-renewal, pricing escalation, or termination exposure
- Run a compliance risk check on procurement agreements
- Compare contract terms against safe/standard positions
Inputs Required
Before starting, collect:
- Contract text or summary: Full contract language, key excerpts, or a structured summary of terms
- Contract type: SaaS/subscription, professional services, goods supply, MSA, NDA, or other
- Contract value and term: Annual/total value and initial term length
- Party position: Whether the user is the buyer or the supplier
- Governing jurisdiction: State/country of governing law, if known
- Risk tolerance: Conservative / balanced / aggressive on negotiation posture
- Deal leverage: How much negotiating power the user has (sole-source vendor vs. competitive bid)
If jurisdiction or leverage is unknown, assume US buyer-side, balanced tolerance, and moderate leverage — and note the assumption.
Instructions
Step 1: Extract the Terms That Map to Each Risk Domain
Read the contract and locate the operative language for each of the seven domains below. If a domain's clause is absent, that absence is itself a flag (silence often defaults to the supplier's favor). For each domain, capture the exact clause reference or quote so the flag is traceable.
Step 2: Screen Each Risk Domain and Assign a Severity
Rate every flag on this scale:
- Critical — direct financial exposure, uncapped liability, or a term that could trap the buyer for years; do not sign without change
- High — meaningful cost or legal risk; negotiate before signing
- Medium — unfavorable but survivable; negotiate if leverage allows
- Low — minor or standard-market term; note for awareness
Domain 1 — Pricing Escalation
- Flag uncapped or index-linked increases (e.g., "CPI + X%"), silent renewal-price resets, and increases at the vendor's "sole discretion"
- Safe position: annual increase capped at a fixed percentage (commonly 3–5%) with written notice
- Critical trigger: no cap, or a cap above 7%, or price set at renewal without a ceiling
Domain 2 — Auto-Renewal Traps
- Flag auto-renewal paired with a long notice window (e.g., 90+ days), evergreen renewals, and notice periods that are hard to hit
- Safe position: auto-renewal with a ≤30-day opt-out notice, or renewal by affirmative agreement only
- Critical trigger: auto-renewal for the full original term with a >60-day non-renewal notice window
Domain 3 — Liability Caps and Indemnification
- Flag uncapped supplier liability exclusions, caps set below fees paid, one-way indemnification, and broad consequential-damages waivers
- Safe position: mutual liability caps at a meaningful multiple of annual fees, with carve-outs for data breach, IP infringement, and gross negligence
- Critical trigger: supplier liability capped at a nominal amount while buyer indemnifies broadly
Domain 4 — SLA Enforceability
- Flag SLAs with no financial remedy, credits capped so low they don't matter, "commercially reasonable efforts" language, and remedies the buyer must claim within an unrealistic window
- Safe position: defined uptime/response targets, service credits that scale with failure severity, and termination rights for chronic breach
- Critical trigger: SLA stated but with no remedy, or credits as the "sole and exclusive remedy" with no termination right
Domain 5 — Termination Provisions
- Flag termination-for-convenience granted only to the supplier, early-termination penalties, long cure periods for the vendor, and post-termination data-hostage or transition gaps
- Safe position: mutual termination for cause with a reasonable cure period, defined buyer exit rights, and a data-return/transition obligation
- Critical trigger: buyer cannot exit for material breach, or owes the full remaining contract value on termination
Domain 6 — Data, Security, and Compliance Obligations
- Flag missing data-processing terms, absent breach-notification timelines, unclear data ownership, subprocessor rights without notice, and non-compliance with applicable regimes (GDPR, CCPA, HIPAA, SOC 2)
- Safe position: defined breach-notification window, buyer data ownership, audit rights, and named compliance standards
- Critical trigger: no breach-notification obligation, or ambiguous ownership of buyer data
Domain 7 — Scope, Change, and Hidden-Cost Clauses
- Flag unilateral change-of-terms rights ("supplier may modify at any time"), overage/usage fees without caps, mandatory professional-services minimums, and fees buried outside the pricing exhibit
- Safe position: changes by mutual written agreement, capped overages, and all fees consolidated in one schedule
- Critical trigger: supplier can change material terms unilaterally with only posted notice
Step 3: Build the Ranked Flag Register
Order all flags by severity (Critical → High → Medium → Low). For each flag, write: the domain, the clause reference, the exposure in plain language (what could go wrong and roughly what it could cost), and a specific remediation.
Step 4: Write Remediation Actions
Each remediation must be usable at the negotiation table — a proposed redline, a fallback position, or a specific question to send the vendor. Not "review this clause." Calibrate aggressiveness to the stated risk tolerance and leverage.
Step 5: Produce a Deal-Level Verdict
Close with an overall risk rating (Acceptable / Negotiate Before Signing / Do Not Sign As-Is), the count of flags by severity, and the top 3 issues to resolve first.
Output Format
- Structure: Executive verdict at top, then a ranked flag register (table or structured blocks), then a prioritized top-3 negotiation list
- Length: 800–1,300 words depending on contract complexity
- Format: Structured document — verdict as short prose; flags as a table or labeled blocks with Domain / Severity / Clause Ref / Exposure / Remediation
- Sections: Deal-Level Verdict → Flag Register (ordered by severity) → Top 3 Priorities to Resolve → Assumptions Made
- Tone: Contract manager reviewing for a busy stakeholder — direct, specific, non-alarmist; states exposure without legalese padding
Quality Checklist
Before delivering output, verify:
- [ ] All seven risk domains were screened, and absent clauses were flagged as risks (not skipped)
- [ ] Every flag has a severity rating from the four-level scale
- [ ] Every flag cites a specific clause reference or quote — not "somewhere in the contract"
- [ ] Every remediation is a concrete action (redline/fallback/question) — none say only "review"
- [ ] Flags are ordered by severity, and the flag counts in the verdict match the register
- [ ] A deal-level verdict and a top-3 priority list are both present
Notes
- This is a screening tool, not legal advice — recommend that Critical and High flags be reviewed by qualified counsel before signing
- Absence of a clause is frequently the biggest risk: missing liability caps, breach-notification terms, or buyer exit rights default to the party who drafted the contract
- When leverage is low (sole-source vendor), reframe remediations as "must-have vs. nice-to-have" so the user knows which fights are worth having
- For renewals, pay special attention to Domains 1 and 2 — escalation and auto-renewal traps compound silently across cycles
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills