⚖️ Legal & Contract Ops Prompt Generator Free · No Signup

Indemnification Clause Comparison: Two Clauses, One Table, A Verdict

⚡ Quick Answer

Comparing indemnification clauses means lining up the same dimensions in both: who indemnifies whom, what triggers it, whether the other side's own negligence is carved out, whether the indemnity sits inside or outside the liability cap, who controls the defence, and how long it survives. This generator takes Clause A and Clause B and returns that comparison as a table — quoted positions, risk, and fallback language. Free · No signup · Runs in your browser.

⚡ Indemnification Clause Comparison: Two Clauses, One Table, A Verdict

Configure your facts below to generate a production-ready, constraint-locked prompt.

100% Client-Side

Why indemnity clauses need comparison, not review

An indemnification clause is meaningless in isolation. "Supplier shall indemnify Customer against third-party claims arising from the Services" sounds standard until you place it next to the alternative and see the differences that matter: the other version is mutual; it carves out claims caused by the indemnified party's own negligence; it caps the indemnity at twelve months of fees; it gives the indemnifying party control of the defence. None of those differences is visible in a solo read, because a solo read has nothing to measure against. That is why the working unit on this page is the comparison table, and why the generator takes two clauses — yours versus theirs, paper versus playbook, or draft versus the market-standard benchmark described below.

The dimensions every comparison must line up

  1. Who indemnifies whom. Mutual, or one-way? A unilateral obligation is the first one-sided pattern to flag.
  2. Trigger and scope. Third-party claims only, or direct losses too? Which claim types — IP infringement, personal injury, property damage, data breaches, tax? Vague "arising from" scope cuts both ways.
  3. The negligence carve-out. Does the indemnity cover claims caused by the indemnified party's own negligence? Language like "regardless of the cause" can make one side pay for the other's fault. The market formulation limits the duty "except to the extent caused by the indemnified party's gross negligence or willful misconduct" — or broader negligence.
  4. Interaction with the liability cap. Is the indemnity inside the contract's cap, or carved out of it? An indemnity carved out of the cap is, in practice, uncapped — the second one-sided pattern, and the one most often missed in a quick read.
  5. Defence and control. Who defends, who chooses counsel, who controls settlement — and can a settlement impose obligations on the indemnified party without its consent?
  6. Survival. How long does the indemnity live after termination or expiry?
  7. Government-authority and regulatory claims. Does the scope sweep in fines or claims by authorities? The fourth recurring one-sided pattern.
  8. IP-infringement scope. For technology contracts: is IP indemnity limited to the service as provided, or does it extend to combinations and modifications the indemnifier didn't supply?

Who this is for

Contract negotiators working both parties' paper in an MSA negotiation; in-house counsel maintaining a standard indemnity position and checking deviations from it; procurement and sales teams who need to know — before the legal call — which of two clauses is safer for their side and what the fallback ask is.

How to use it

  1. Paste Clause A and Clause B and label each (ours / theirs). No second clause? Paste your standard position as Clause B instead — the comparison works the same way.
  2. State your side and your standard position (cap basis, mutuality requirement, negligence carve-out). The verdict is given for your side; a clause that is "more favourable" is only meaningful relative to a side.
  3. Run the prompt and read the table row by row — positions are quoted, not paraphrased.
  4. Take the fallback column into the negotiation. Every high-risk row ships with language you can propose.

THE PROMPT TEMPLATE

PROMPT TEMPLATE
You are a senior commercial contracts reviewer. I represent the [SIDE: e.g. vendor / customer] in a [CONTRACT TYPE] governed by the law of [JURISDICTION].

MY STANDARD POSITION: [e.g. indemnities must be mutual; each party's indemnity is capped at the fees paid in the preceding 12 months and sits INSIDE the liability cap; no indemnity for claims caused by the indemnified party's own negligence; indemnifying party controls the defence and no settlement imposing obligations on us without our consent]

Compare the two indemnification clauses below across these dimensions, in this order: (1) who indemnifies whom; (2) trigger and scope of claims; (3) negligence carve-out — whether a party must pay for claims caused by the other party's own negligence ("regardless of the cause" language counts as NO carve-out); (4) interaction with the liability cap — inside, outside/carved out, or not determinable from the clause; (5) defence and settlement control; (6) survival period; (7) government-authority / regulatory claims; (8) IP-infringement scope.

RULES
- Quote the operative wording for each position. Do not paraphrase a position and present it as the clause's position.
- Where a dimension is absent from a clause, write "Absent" for that side. Where it cannot be determined from the text provided, write "Not determinable from the documents provided."
- Flag these one-sided patterns wherever they appear: unilateral obligation; scope covering the other party's own negligence; indemnity carved out of the liability cap (treat as effectively uncapped); government-authority claims included.
- Where you compare against "market standard," label it exactly: "Market-standard benchmark (model knowledge — not a cited source)". Never present the benchmark as a quotation or a statistic.

OUTPUT FORMAT
1. Comparison table: | Dimension | Clause A position (quoted) | Clause B position (quoted) | Risk to us | Fallback language |
2. One-sided pattern flags: list each pattern found, the clause it appears in, and the quoted words that create it.
3. Verdict: "Clause [A/B] is more favourable to [SIDE] because…" — 3 reasons maximum, each tied to a table row.

NEGATIVE CONSTRAINTS
- Do not state a position for a clause that the quoted text does not support.
- Do not invent a liability cap, survival period, or carve-out that is not in the text.
- Do not declare either clause "standard" or "market" as a conclusion — compare; let the table speak.

CLAUSE A ([LABEL: e.g. their MSA, v3])
[PASTE CLAUSE A]

CLAUSE B ([LABEL: e.g. our standard])
[PASTE CLAUSE B]

Best model: Claude Sonnet 5.5 for the closest reading of operative wording. For two full agreements rather than extracted clauses, extract the indemnity and liability sections first — or use Gemini 4 Argon (1M-context option) and instruct it to locate both sections itself.

Worked example — vendor paper vs customer paper

Filled inputs: Side represented: the vendor · Standard position: mutual indemnities, capped at 12 months of fees, inside the cap, negligence carve-out · Clause A (their MSA):

"Supplier shall indemnify, defend and hold harmless Customer from and against any and all claims, damages, losses and expenses arising out of or relating to the Services, regardless of the cause thereof. The obligations in this Section are not subject to the limitation of liability in Section 11."

Clause B (our standard):

"Each party shall indemnify the other against third-party claims to the extent caused by the indemnifying party's negligence or willful misconduct. Each party's indemnification obligations are subject to the cap in Section 11 (fees paid in the preceding 12 months) and the indemnifying party shall control the defence; no settlement imposing obligations on the indemnified party shall be entered without its prior written consent."

Output excerpt:

DimensionClause A position (quoted)Clause B position (quoted)Risk to usFallback language
Who indemnifies whom"Supplier shall indemnify… Customer""Each party shall indemnify the other"A is unilateral — we carry all indemnity risk in the dealAdopt B's mutual formulation
Negligence carve-out"regardless of the cause thereof""to the extent caused by the indemnifying party's negligence or willful misconduct"A makes us pay for claims caused by the Customer's own negligence"…except to the extent the claim arises from Customer's own negligence or willful misconduct"
Cap interaction"not subject to the limitation of liability in Section 11""subject to the cap in Section 11 (fees paid in the preceding 12 months)"A is carved out of the cap — effectively uncapped indemnity"Supplier's aggregate liability under this Section shall not exceed the fees paid in the preceding twelve (12) months"
Defence & controlAbsent"the indemnifying party shall control the defence" + consent for obligation-imposing settlementsA silent: we could be bound by a settlement we didn't controlInsert B's defence/consent sentence

One-sided pattern flags (excerpt): Unilateral obligation — Clause A ("Supplier shall indemnify… Customer"). Scope covering the other party's own negligence — Clause A ("regardless of the cause thereof"). Indemnity carved out of the liability cap — Clause A ("not subject to the limitation of liability in Section 11").

Verdict: Clause B is more favourable to the vendor because it is mutual, it limits our duty to claims we actually caused, and it keeps the indemnity inside a defined cap instead of leaving it unlimited.

GUARDRAILS

Quote-first. The comparison table runs on quoted positions for a reason: indemnity risk lives in single phrases — "regardless of the cause," "not subject to" — that paraphrase reliably smooths away. Check any row against the pasted clause before you negotiate from it.

"Not determinable from the documents provided." The cap interaction is the classic case: an extracted indemnity clause often does not say whether it sits inside the liability cap, because the cap lives in another section. The template forces "Not determinable" rather than a guess — and that answer is your instruction to pull Section 11 before concluding anything.

Benchmarks are labelled, not cited. Where the output compares a clause to "market standard," that is the model's general knowledge, and the template makes it say so. It is a negotiation aid, not evidence; do not quote it to the other side as authority.

Privilege and confidentiality. Two parties' contract paper plus your standard position is a map of your negotiating strategy — treat it accordingly. ABA Formal Opinion 512 (July 2024) applies confidentiality and competence duties to AI use, and United States v. Heppner (S.D.N.Y., February 2026) held that material shared with a consumer AI tool was not privileged. This generator runs 100% client-side — your clauses are never transmitted to PromptStaging — and anonymising party names before any external run costs you nothing in output quality.

Not legal advice. Whether to accept an uncapped or unilateral indemnity is a commercial-legal judgment for a qualified attorney in the governing jurisdiction, informed by insurance, deal value, and leverage — none of which a clause comparison can weigh for you.

Frequently Asked Questions

Practical answers covering model choice, workflow integration, and compliance rules.

Four recurring patterns: a unilateral obligation (only one party indemnifies); scope that covers the other party's own negligence (watch for "regardless of the cause"); an indemnity carved out of the liability cap, which makes it effectively uncapped; and scope that sweeps in government-authority or regulatory claims. Any one of them shifts risk one way; the example on this page shows a single clause carrying the first three.

From the indemnifying side, almost always — an uncapped indemnity is open-ended exposure that survives independently of the contract's fee value. A common playbook position caps the indemnity at the fees paid in the preceding 12 months and keeps it inside the overall liability cap. Carve-outs from the cap are sometimes negotiated deliberately (for example for IP infringement), but they should be a decision, not a drafting surprise.

Indemnification decides who pays for specified claims — one party covers the other's losses for defined events, typically third-party claims. Limitation of liability decides how much a party can be liable for overall, usually a cap plus excluded loss types. They interact: an indemnity carved out of the liability cap bypasses the "how much" limit entirely, which is why the comparison on this page always checks the two together.

Fix the dimensions first — who indemnifies whom, trigger and scope, negligence carve-out, cap interaction, defence control, survival, regulatory claims, IP scope — then extract each clause's position on every dimension, quoting the operative words. The verdict comes last and must trace back to the rows. The generator above runs exactly that sequence; doing it by hand in a spreadsheet works too, if you keep the quote-don't-paraphrase rule.

It usually means the indemnifying party pays even where the claim was caused by the indemnified party's own negligence — the indemnity applies "regardless of" whose fault it was. For the indemnifying side that is one of the most expensive phrases in contract drafting. The standard counter is a carve-out: the duty applies "except to the extent caused by the indemnified party's negligence or willful misconduct."

As a starting point for negotiation, yes; as evidence, no. When the output on this page refers to market standard, it is the model's general knowledge, labelled as such in the template — not a cited source or survey. Use it to decide what to ask for, and expect the other side's counsel to argue their own version of "market."

INTERNAL LINKS

--- ---