Prompt Pilotby Write.law
All prompts
Analyze

Assess bad-faith exposure on a claim

Builds a conduct timeline from the claim file and tests it against the bad-faith standard you supply (never one it invents) so you see which file entries create exposure and which gaps in the file create more.

About 30 minadvancedInsurance, Litigation

Checked against standard v2 on Sep 29, 2026 revised Sep 19, 2026

What to gather first

  • The claim file: log notes, correspondence, reserve history, internal referrals
  • The bad-faith standard for the governing state, with the case or statute you are relying on
  • Every settlement demand and its deadline, and what the carrier did inside that window
  • Policy limits, the demand, and a realistic verdict range
  • Whether the exposure is first-party, third-party failure to settle, or both

Your prompt5,860 characters

Still to fill in: Claim file and handling record, The governing standard, as you have confirmed it, Governing state and forum

RoleYou are a coverage litigator who has tried extra-contractual cases from both chairs. You know bad faith is not one cause of action but a family of them, and that the elements and damages change completely across state lines, so you refuse to grade conduct against a standard you have not been handed. You read the claim file for its silences, because that is where the exposure lives.What I needAssess the carrier's exposure on the claim below under Governing state and forum, applying only the standard I supply. I am Policyholder pursuing the carrier.InputsClaim file: Claim file and handling record Governing standard, as I confirmed it: The governing standard, as you have confirmed it Demands and deadlines: Demands and deadlines State and forum: Governing state and forum Who I represent: Policyholder pursuing the carrier Limits and exposure: Limits and exposureHow to work this1. Apply only The governing standard, as you have confirmed it. If it is blank or is not an actual statement of elements, do not substitute one: name what I must confirm in Governing state and forum: the duty, any statutory hook and whether it is privately enforceable, the genuine-dispute defense, the punitive threshold, and stop short of grading conduct. 2. Build a dated conduct timeline from Claim file and handling record: every date the carrier learned, decided, sent, or moved a reserve. Mark each entry [FILE] where a document supports it, [INFERRED] where you fill a gap. 3. Name the silences. An unexplained interval, a note with no corresponding action, a reserve that moved without a stated basis. Each is its own finding, and these carry more exposure than documented decisions. 4. One table row per element: entries that support it, entries that defeat it, a rating of strong, contested, or not supported. No rating without both columns filled or an express note that one is empty. 5. State the carrier's best defense in its strongest form, then name the file entry that undercuts it. 6. Split the money: exposure inside limits, above limits, fees, any multiplier, each tied to a figure from Limits and exposure and a stated assumption. Ranges, not adjectives. 7. List every discovery-exposure category in the file, marking each whose protection turns on Governing state and forum law I have not confirmed.Ask me firstBefore you analyze anything, ask me these questions, then stop and wait: 1. Is the exposure first-party (the insured against its own carrier), third-party failure to settle within limits, or both? The elements and the damages differ completely. 2. Confirm that The governing standard, as you have confirmed it is the standard you want applied, and give me your source. If you have not yet settled whether Governing state and forum recognizes a common-law duty, a privately enforceable statutory claim, or only contract remedies, say so. I will list what to look up rather than supply elements myself. 3. What exists on this claim that is not in the file you gave me: emails outside the claim system, a supervisor conversation nobody logged, a defense counsel evaluation held separately, an unwritten authority limit? The gaps in a claim file are what a bad-faith case is built on, and you know where they are. 4. What is this assessment for: setting a reserve, building a mediation position, or deciding whether to sue, and is it being created as privileged work product or as something that could end up produced? That changes how bluntly I state a conclusion the record will not carry. Do not begin until I answer. If I tell you to proceed anyway, state each assumption at the top of your output and mark it [ASSUMPTION - verify].Output formatTimeline table. Gaps section. Element table with support, defeat, and rating. The carrier's best defense and what undercuts it. Two exposure figures with their assumptions. Discovery-exposure list. Close with the assumptions the analysis rests on and the one fact worth confirming first. End with one line naming the two of my answers that most changed this assessment: say which element rating or which exposure figure would have come out differently without them. If an answer changed nothing, say so; it means the question did not belong in the list.Never do this- If the assessment would read the same for any carrier in any state, it is too generic. Every rating cites a dated entry from the file. - No hedging filler. Cut "arguably," "it should be noted," and "bad faith is fact-intensive" in place of naming which fact. Do not tell me to consult an attorney. I am the attorney. - Never invent or recall from memory Governing state and forum's elements, its genuine-dispute defense, its punitive threshold, or whether an unfair-claims statute is privately enforceable. Anything not in The governing standard, as you have confirmed it is marked [UNVERIFIED - confirm before relying]. - Where you do not know whether a file category is discoverable in Governing state and forum, say you do not know rather than smooth over it. - Do not pad. If the file shows a prompt, documented, reasoned denial, say so in three sentences and spend the analysis on the one exposed entry. Length is not value. - Never treat a wrong coverage decision as bad faith by itself, or a reserve figure as an admission of value.Before you answer- Did I apply the supplied standard rather than one of my own, and mark every timeline entry [FILE] or [INFERRED]? - Does every element row cite entries on both sides, and did I state the carrier's best defense in its strongest form before rebutting it? - Would this fit a different claim file? It should not.

Running it takes you through the three turns (the first prompt, the pressure test, and the follow-up) and a check of the answer, in order. The Cockpit lets you add tunes. Both keep the values you entered. You can change your own copy as you like. It is labeled as yours everywhere it appears.

2

Pressure-test it

Asks the AI to look for weaknesses in its own answer.

At the 30(b)(6) deposition of the claims department, the plaintiff's lawyer has this timeline printed and marked up on the table in front of him. He intends to walk the corporate representative through it entry by entry and get an admission at each silence. Write, as the plaintiff's lawyer, the five questions that hurt most, name the file entry each is built on, and say which one the carrier cannot answer well. Then tell me which conclusion in the analysis above would not survive those answers.
3

Go deeper

Asks for more detail and further work once the first answer is right.

Every gap in that timeline is a document sitting in the carrier's claim file. Turn the gaps into a discovery plan: the document requests and 30(b)(6) topics aimed at each unexplained silence in the timeline, ranked by which element they would prove or defeat, with a note on which categories are likely to draw a protective-order fight in this state.
4Check what came backPaste the AI's answer and check it, item by item, against this prompt's rules.

Watch for

  • Bad-faith law varies more by state than almost anything else in insurance: elements, whether a statutory claim carries a private right of action, the genuine-dispute defense, punitive thresholds, and damages all differ. Supply the standard yourself and verify it; do not let the model produce one.
  • Whether reserves, claims manuals, internal guidelines, and coverage-counsel opinions are discoverable turns on state law and on the posture of the case. Confirm before you concede or withhold anything.
  • A wrong coverage decision is not bad faith by itself in most states. Analysis that conflates the two will overstate exposure and mislead a reserve discussion.
  • This analysis is protected work product only if you create and route it that way. Label it, and do not paste an unproduced claim file into any tool your firm's AI policy does not cover.
  • Time-limited policy-limits demands generate the largest excess exposure and often are not in the file you were handed. Confirm the demand history independently.

What comes back

A conduct timeline with every entry dated and marked [FILE] or [INFERRED]. A gaps section naming each unexplained silence in the record. An element-by-element table: the element as supplied, the entries supporting it, the entries defeating it, and a rating of strong / contested / not supported. A candid statement of the carrier's best defense and what undercuts it. Two separate exposure figures (inside limits and above limits, plus fees and any multiplier), each tied to a stated assumption. A discovery-exposure list flagging every category whose protection turns on unconfirmed state law.

See an example answer
*(After you answer the four questions, say, third-party failure to settle; Stowers standard supplied and sourced; outside the file there is a 7/22 supervisor conversation nobody logged and an unwritten $60,000 authority limit; and this is a privileged pre-mediation assessment. The $100k limits, the 7/9 time-limited demand and the $900k-$2.4M verdict range came in on the form.)* Conduct timeline | Date | Entry | Source | |---|---|---| | 6/2/2025 | Adjuster logs "liability clear, med specials $88k, plaintiff 34yo w/ surgery recommendation" | [FILE] | | 6/14/2025 | Reserve set at $75,000; no stated basis in the note | [FILE] | | 7/9/2025 | Policy-limits demand received: $100,000, unconditional except full release, open 21 days | [FILE] | | 7/15/2025 | Defense counsel emails evaluation: "exposure exceeds limits, recommend tender" | [FILE] | | 7/22/2025 | Log entry: "awaiting additional med records"; no record of a request being sent | [FILE], gap noted | | 7/28/2025 | Counteroffer $60,000 | [FILE] | | 7/29/2025 | No entry. No extension requested | [INFERRED - from absence of any note] | | 7/30/2025 | Demand expires: 21 days from the 7/9 receipt | [FILE] | | 3/18/2026 | Verdict $2,140,000 | [FILE] | Gaps that matter. (1) Thirteen days between defense counsel's tender recommendation and any carrier response, with no supervisory referral in the file. (2) The 7/22 "awaiting records" note with no corresponding request. If no request went out, the note is worse than silence. (3) The reserve set at $75,000 on 6/14 against $88,000 in specials, with no documented reasoning. Element analysis: Stowers, as supplied | Element | Supports | Defeats | Rating | |---|---|---|---| | Demand within limits | 7/9 demand at $100,000 policy limits | none | Strong | | Unconditional, full release | Demand conditioned only on a full release | Carrier may argue the release did not cover the co-defendant | Contested: read the release | | Ordinarily prudent insurer would accept | 6/2 note conceding clear liability; $88k specials; 7/15 counsel recommendation to tender | 7/22 note asserting incomplete records | Strong against the carrier | The carrier's best defense. That the medical record was incomplete on 7/30 and the surgery recommendation had not been converted to a treatment plan, so a prudent insurer could still evaluate below limits. What undercuts it: defense counsel's own 7/15 evaluation, and the absence of any records request behind the 7/22 note. If no request exists, that defense will not survive the 30(b)(6). Exposure. Inside limits: $100,000. Above limits: the $2,140,000 judgment less limits equals $2,040,000 of excess exposure, plus prejudgment interest and the insured's fees, and a statutory claim under ch. 541 would need to be separately evaluated [UNVERIFIED - confirm current Texas treatment of chapter 541 damages and any multiplier]. Discovery exposure. Reserve history and the basis for the 6/14 figure; the claims manual section on limits demands; adjuster and supervisor compensation tied to closing ratios; the 7/15 defense counsel email, already in the file and likely not privileged as against the insured [UNVERIFIED - confirm Texas rule on the tripartite relationship]; any coverage-counsel opinion. Assumptions. That no records request exists behind the 7/22 note [verify first - this is the highest-value fact in the file]. That the 7/9 demand was unconditional as characterized [verify against the letter itself]. That the release covered only this insured [verify]. What your answers changed. Two of the four did real work. "Third-party failure to settle" is what produced the $2,040,000 excess figure and the entire element table. In a first-party frame there is no excess judgment to chase, Stowers is the wrong test, and the exposure section tops out at the $100,000 limit plus whatever a chapter 541 claim adds. The unlogged 7/22 supervisor conversation and the unwritten $60,000 authority limit are why the gap analysis leads the memo rather than closing it: a limit that exists but appears nowhere in the file is the single worst fact here, because the carrier cannot explain the 7/28 counteroffer without producing it and cannot produce it without conceding it drove the decision. Told the file was complete, that section is two lines about missing supervisor referrals and the memo's centre of gravity stays on the timeline. Your privilege answer changed nothing. Every rating here rests on a dated entry already in the file, so this reads the same as a mediation exhibit as it does as work product. I would not soften a line of it.
Why this prompt is written this way
## Framework 1. **The standard comes from the user, not from memory.** Elements, statutory hook, private right of action, punitive threshold, genuine-dispute defense. If it is not supplied, name what must be confirmed and stop short of grading conduct. 2. **Timeline from the file.** Every date the carrier learned something, decided something, wrote something, or moved a reserve. Each entry marked [FILE] or [INFERRED]. 3. **Read the silences.** The 40-day gap, the supervisor note that is not there, the reserve that moved without a stated reason. 4. **Elements against entries.** Each element gets the entries that support it and the entries that defeat it. No conclusion without both columns. 5. **Score the carrier's best defense honestly.** Usually a genuine coverage dispute. Say what in the file undercuts it. 6. **Two buckets of money.** Contract exposure inside limits; extra-contractual exposure above them, plus fees and any statutory multiplier. 7. **Discovery is the real risk.** Reserves, claims manuals, internal guidelines, coverage-counsel opinions, adjuster compensation. Flag each, and flag that discoverability itself varies by state.