Skills/Resume and hiring/Offer negotiation brief

Offer negotiation brief: the leverage is created before the conversation that spends it

Produces a written brief: the three numbers, the movable offer components split from the structurally fixed ones, one consolidated ask, and the walk-away condition recorded before the first call.

Not yet measured skill 4,295 words MIT by Locul Verified safe · 0 secrets Written 2026-08-20
/plugin marketplace add mkhalid1/locul-skills

Then /plugin to install Resume and hiring, which includes this skill.

Download just this file4,295 words
No account and nothing to sign up for. The marketplace command works in Claude Code today. Using something else? Copy the file or download it and put it where your assistant expects.
We have not measured this skill. There is no result on this page because we have not run one. It is written, it has been read for accuracy, and it is free to take. Nothing below claims it improves an output, because we have not shown that. This is different from a skill that failed our test: those are not published at all.
What it is, and what we are not claiming

Untested. The asset is the component taxonomy, which says why a sign-on moves when base cannot, plus the rule that all three numbers are written down and dated before the first call.

We have not measured this one. It is published untested, and it is the last of the seven in this category on purpose, because it is the only one where the useful work happened before the conversation it appears to be about.

What it knows, concretely. First, a component taxonomy with a structural reason attached to each entry, not a list of things to ask for. A bonus target is a percentage attached to a level rather than to a person, so asking to raise it spends an ask on something that was never available. A sign-on is the most movable component inside a company with published bands, precisely because it is one-off cash that does not raise the band, does not compound into future merit cycles, and creates no comparison problem with the peers already at that level. Base is the least movable in the same company and the most movable in one with no bands at all.

Second, three numbers written down and dated before the first call: the reservation you will actually decline below, the target the evidence supports, and the anchor you say out loud. The rule on the anchor is a constraint rather than a formula. If you cannot say in one sentence where the anchor came from, it is not an anchor, and the first question about it ends the conversation you were trying to have.

Third, a decision rule with a refusal in it. If you have no band and no comparator, you do not counter. You ask for the range for the level first, because a counter made blind becomes the ceiling.

Who it is not for. If the offer is above your target and near the top of a range the employer published themselves, this is overkill and the brief will tell you so. If you have no alternative at all, no other process and nothing to return to, it will not manufacture leverage, and it says that plainly rather than dressing a weak position up as a strong one.

When to reach for it

  • The moment a verbal offer is made on a call and before you say anything other than thank you, which is the last point your number is still unspoken.
  • When the written breakdown has just arrived by email and the recruiter has asked for an answer by a stated date.
  • Before replying to a recruiter who has asked what you are currently earning or what you are looking for.
  • When a second process is still live and the first employer has attached a deadline that lands before it concludes.
  • When an offer contains a private company equity grant and somebody has quoted it to you as a single cash figure.

Why there is no number on this page

Measuring one skill honestly costs about twenty model sessions: five runs with it, five without, on real material, each output graded alone by a session that is not told the other arm exists, against a rubric written by somebody who never saw the skill. We have not spent that on this one yet, so it ships labelled rather than ships silently.

How it would be measured. Tier A. Material: twelve invented offer packets, each with a full written breakdown, a stated jurisdiction, a posted or absent band, and an equity grant, where an author who never saw the file recorded in advance which components were movable and which branch of the decision rule applies. Objective spine: correct branch selected, the cannot-tell branch chosen when no band and no comparator exist, a single consolidated ask rather than a sequence, bonus target correctly identified as level-attached, private equity recorded as units and percentage rather than a cash figure, and the walk-away condition written before any counter.

The spine is unusually clean for a soft-skills topic, because most of the gradeable content is classification rather than persuasion. Which components are movable, which branch of the decision rule an offer falls into, whether the equity was recorded as units and a percentage or as a cash figure, and whether the ask arrived as one message or three: all countable from a written brief by somebody who never saw the file.

The awkward part is that the material has to be invented offer packets, and an author writing them can accidentally make the movable component obvious. They need writing by someone who does not know the taxonomy, or the exercise grades itself.

The honest prior is that a strong model already produces reasonable negotiation language unprompted. The narrower questions are whether it consolidates the ask instead of sequencing it, whether it refuses to counter when there is no band, and whether it declines to convert a private equity grant into a dollar number.

The rule that decides pass or fail was written down before any run was executed and it does not move afterwards. It is in the method note on the hub, along with the full results table including every skill that was tested and cut.

What it does not do

Stated plainly, because a skill that claims everything is useful for nothing.

  • It is not legal advice and it is not tax advice. Pay disclosure rules, salary history bans and equity taxation are jurisdiction specific and change, and equity taxation in particular needs a qualified professional rather than a document.
  • It cannot value a private company grant, because the inputs that would let anyone do so, the preference stack, the eventual exit price and the dilution between now and then, are not knowable at signing.
  • It cannot see the employer's band, their approval limits, or how many other candidates are at offer stage, so it works from what you can evidence and never from what they can afford.
  • It has no view on whether you should take the job. Everything here is about the terms of an offer you have already decided you want.
  • It does nothing for an offer where the level was set wrong, which is a screening problem that had to be fixed before the loop. A recruiter or a compensation professional inside the company can move a level, and no document you write afterwards can.
  • Comparator sites such as levels.fyi are better than this at the one thing it does not do, which is supplying market numbers. This ranks and frames evidence, it does not source it.

Install it

  1. Open Locul, go to Library, and choose Import. One-click import from this page lands shortly.
  2. Locul writes the file to the right folder for every assistant you have connected, so you do not have to know where each one keeps its skills.
  3. Environment variables and headers in any shared config are replaced with a placeholder before they reach you, so importing a stranger's setup cannot hand you their credentials or take yours.
  4. Locul is free to start, on Mac and Windows. Get it here.
  1. Download SKILL.md using the button above, or copy the file.
  2. Save it at .claude/skills/offer-negotiation-brief/SKILL.md in your project, or under ~/.claude/skills/offer-negotiation-brief/SKILL.md on Mac and Linux, or %USERPROFILE%\.claude\skills\offer-negotiation-brief\SKILL.md on Windows, to make it available everywhere.
  3. Start a new session. Claude Code picks up the skill from the name and description in the file's frontmatter, so you can also invoke it by name.
  1. Download or copy the file.
  2. For Claude Desktop, add it through the skills panel in settings, or drop the folder into your skills directory.
  3. For Cursor and other assistants that read plain instruction files, paste the body into your project rules file. The skill is plain markdown with no tool bindings, so it carries across.

Pairs well with

What else does this job

A recruiter who is on your side, meaning an external one paid on placement, does part of this better than any document, because they know the employer's approval limits and you never will. The cost is that their incentive is the deal closing, not the deal being good.

A compensation professional or an employment lawyer is the right call when the offer contains an equity grant large enough to matter, a restrictive covenant, or a clawback you do not understand. This file is explicit that it is neither, and that the tax treatment of options in particular is a professional question rather than a document question.

The comparator sites are genuinely the best source of the one input this does not supply, which is numbers. Use them for that and ignore their advice sections.

The model with no skill at all writes a perfectly serviceable counter-offer email. What it tends not to do is stop you: it will happily draft the third separate small ask, help you argue that your rent went up, and put a dollar value on a private company option grant because a recruiter put one there first.

Read the full source
---
name: offer-negotiation-brief
description: Builds a written negotiation position for a job offer before any conversation happens: the reservation, target and anchor numbers, the offer components separated into structurally movable and structurally fixed, a single consolidated ask ranked by evidence, the scripted sentences for the first exchange, and the walk-away condition recorded in advance. Carries the reasons a bonus target cannot move while a sign-on can, how to record a private company equity grant without pretending to value it, and a decision rule that refuses to counter when no band and no comparator exist. This skill should be used when an offer has been made or is imminent and the terms are still open.
---

# Offer negotiation brief

## The claim this skill is built on

The output is a brief: a short written document containing three numbers, a component list, one consolidated ask, four or five sentences you will actually say, and a walk-away condition. It is written before the call and it is dated. That date is doing real work, and the failure modes section explains why.

The obvious approach is to prepare for the conversation. Almost all published negotiation advice is tactics for the call: how to pause, how to phrase a counter, when to stay silent. By the time the call happens the outcome is largely fixed by three things that were decided earlier.

**Whether the employer knows your current number.** Once a number of yours is in the room it functions as the ceiling. Everything after that is an argument about how far above your own number you deserve to be, which is a worse argument than one about what the role pays.

**Whether another process is live.** Leverage is the existence of an alternative. It is not confidence, preparation, or how much the hiring manager likes you. An alternative you have to invent is not one, and experienced recruiters ask follow-up questions that expose an invented one within two exchanges.

**Whether the level was set correctly.** Level maps to band, and band sets the base range, the bonus percentage and the equity range at once. A level correction therefore moves all three and is worth more than any base negotiation. It is also nearly always settled before the offer, because raising a level afterwards usually means re-running a committee or an extra loop.

A brief that starts at the call is starting three moves late. So this one starts by writing down what is already true, and builds the ask out of that rather than out of technique.

## Step 1. Record the three conditions before anything else

Write these at the top of the brief, because they determine which of the later steps are available to you.

- **Has any number of mine been stated?** Note where, to whom, and whether it was a current salary or an expectation. If a number has been given, the anchor step below is mostly closed and you should say so rather than pretending otherwise.
- **What is my real alternative?** One of: another written offer, another process at a named stage with a date, my current job with its actual total, or nothing. Write the honest one. This becomes the reservation.
- **What level am I being offered, and does it match the scope in the job description?** If the description names ownership, on-call, or people management that the level does not carry, that is a level question and it goes in the ask first, ahead of base.

## Step 2. Decompose the offer, movable against fixed

This decomposition is the part of the brief that decides where the ask goes. Movability is structural, not a matter of how firmly anyone asks.

**Level.** The largest lever and usually already closed. Moving it changes base, bonus and equity together, and it is expensive to raise after an offer because it typically needs a committee or an extra interview. Raise it only where the scope evidence is unambiguous.

**Base salary.** In a company with published bands, the least movable component. Exceeding the band requires an exception that goes up one or two levels of approval and creates an internal comparison problem with every peer already at that level, which is the real reason it is refused. In a company with no bands, an early stage or small firm, base is the most movable component, because there is nobody to compare you against.

**Bonus target.** A percentage attached to the level, not to the person. Effectively never movable. What occasionally is available is a guaranteed first-year bonus, which is a sign-on wearing a different name.

**Sign-on bonus.** The most movable component in a banded company. It is one-off cash, it does not raise the band, it does not compound through future merit cycles, and it sets no precedent for peers. It almost always carries a clawback: repayment, often pro-rated, if you leave inside a stated period, commonly twelve months and sometimes twenty-four. Read whether repayment is on the gross figure, because tax was withheld from the amount you received and repaying gross out of net is a real loss.

**Equity grant.** Movable within a per-level range, and sometimes more movable than base because it does not appear in the salary comparison your future peers can see. At a private company the number of units is a fact and the value is a projection. Step 6 handles this.

**Start date.** Highly movable and cheap for the employer. Two to four extra weeks can be worth a great deal if it carries you past a vesting date or a bonus payment at your current employer, and it is the component people forget to ask for.

**Title.** In some companies the title is cosmetic and the level is real, so a title can move on its own. In others the title is the level. Establish which before spending an ask on it.

**Location and remote terms.** Frequently tied to a geographic pay tier, so moving between tiers after signing often triggers an adjustment. Get the written policy, not a summary of it.

**Relocation.** Usually a fixed tiered package, sometimes convertible to cash at a discount, often clawed back on the same schedule as a sign-on.

**Review timing.** An off-cycle review at six months, or inclusion in the next merit cycle rather than being pro-rated out of it for joining late in the year. Cheap, rarely requested, and it compounds.

**Notice buy-out or make-whole.** If you are forfeiting a deferred bonus or serving a long notice period, a payment covering the forfeited amount is easier to approve than base, because it is bounded and you can evidence it with a document.

## Step 3. Set the three numbers, and date them

Write all three in the brief before any conversation, then do not edit them during one.

**Reservation.** The number below which you decline, and you must actually decline. Compute it against the real alternative from step 1, in total cash rather than base alone. If your alternative is your current job, the reservation is at least its total, because moving for less money and more risk needs a written reason.

**Target.** What the evidence supports. Every target traces to something in step 5: a posted band, a competing written offer, a documented comparator, or a level correction. A target with no trace is a wish.

**Anchor.** What you actually say. It sits above the target, and the gap is governed by a constraint rather than a percentage. The anchor must rest on the same evidence as the target, so that the question "where did that number come from" has a one-sentence answer. If it needs evidence you do not have, lower it until it does not. Two sanity checks: if they accepted the anchor immediately you should not feel you undershot, and splitting the difference between the anchor and their offer should land at or above the target.

Dating the numbers matters because a reservation that moves after the offer arrives was never a reservation. See Reservation Drift below.

## Step 4. The legal surface, stated carefully

**This is not legal advice.** Pay disclosure requirements, salary history bans and pay secrecy rules are set at national, state, provincial and city level, they have changed repeatedly since 2021, and they will change again. Check the statute or the labour authority for the exact place the role is based, and take a qualified adviser for anything that matters. What follows is the shape of the mechanisms, with dated examples, so you know what to ask for.

There are four distinct mechanisms and they are often confused with one another.

1. **Posting disclosure.** The employer must publish a pay range in the advertisement itself. Verified example, checked 20 August 2026: California Labor Code section 432.3(c)(3) requires an employer with 15 or more employees to include the pay scale in any job posting, and section 432.3(m)(1) defines pay scale as a good faith estimate of the salary or hourly wage range the employer reasonably expects to pay for the position upon hire. The section was amended by Stats. 2025, Ch. 468, Sec. 1 (SB 642), effective 1 January 2026. That definition is worth reading twice, because a posted range is what they expect to pay on hire, which is not the same document as the full band for the level.
2. **Disclosure on request.** The employer must supply the range to an applicant who asks, sometimes at a defined point such as before an offer is made. California Labor Code section 432.3(c)(1) and (c)(2) require the pay scale to be given to an applicant on reasonable request, and to a current employee for their own current position.
3. **Salary history ban.** Two separate prohibitions that people routinely conflate. Asking: California Labor Code section 432.3(b) says an employer shall not, orally or in writing, personally or through an agent, seek salary history information, including compensation and benefits, about an applicant. Relying: section 432.3(a) says an employer shall not rely on that information as a factor in determining whether to offer employment or what salary to offer. Then section 432.3(h) says nothing in the section prohibits an applicant from voluntarily and without prompting disclosing salary history. That third clause is the one that matters to you: the California ban protects you from being asked, not from yourself. Not every jurisdiction draws the line in the same place. The Illinois Equal Pay Act of 2003, 820 ILCS 112, bars an employer from using wage history even where the applicant volunteered it, which is a materially different rule, and it is why the answer to "is this allowed where I am" is a statute rather than a habit.
4. **Pay secrecy bans.** Rules preventing an employer from forbidding employees to discuss their own pay with each other.

The detail differs enough between jurisdictions that you check rather than assume, and two more state examples show how much. Both are from the state agencies and the statute text, checked 20 August 2026. Washington's Equal Pay and Opportunities Act, RCW 49.58.110, requires an employer with 15 or more employees to include a wage scale or salary range in a job posting, together with a general description of all benefits and other compensation. New York State Labor Law section 194-b, effective 17 September 2023, applies from four or more employees, covers promotions and transfers as well as external advertisements, and requires the employer to state whether a role is commission-only. Same mechanism, different size thresholds, different scope, and neither one is what California says.

In the European Union, Directive (EU) 2023/970 on pay transparency was adopted in May 2023 and entered into force in June 2023, with a transposition deadline of 7 June 2026 according to the European Commission's equal pay pages, checked 20 August 2026. It creates a right to information on pay before employment, a prohibition on asking applicants about their pay history, and a prohibition on pay secrecy clauses. A directive binds member states rather than employers directly, so the text that governs you is your own country's implementing law. As of August 2026 the deadline has passed, and transposition timing and detail differ between member states, so read the national statute rather than the directive.

How this changes the brief. If a range was posted, it is evidence and it is theirs, which makes it the strongest thing you can quote back. If asking is banned where you are, the answer to the question is a redirect to the range rather than a number. If disclosure on request exists, ask before you counter, because that is the mechanism that gets you out of the cannot-tell branch in step 8. And nothing in any of these obliges an employer to pay at the top of a band, so a posted range spanning three levels is not information.

## Step 5. Rank the evidence, and drop what moves nothing

Order the ask by what actually shifts a number.

1. **A competing written offer with figures in it.** Strongest, because it changes the employer's risk of losing you and it is verifiable.
2. **The employer's own published band.** If the offer sits below a range they published, that is a fact about their document, not an opinion about you.
3. **A level correction backed by scope in the job description.** Moves three components at once, and is the only item here worth raising ahead of base.
4. **A documented internal comparator.** Rarely available from outside. Powerful when it is, and it must be a document rather than something a friend told you.
5. **A market data set, with its bias named.** Say which source and acknowledge its selection problem out loud, because a recruiter who has seen the same site will otherwise say it for you.

Moves nothing, and putting it in the brief weakens everything around it: cost of living, personal expenses, tenure at your previous employer, how much you want the job, effort already spent in the process, and an unnamed unwritten interest from a company you cannot identify.

## Step 6. Record the equity, do not value it

An equity grant is three facts: a quantity, a schedule, and a price if there is one. Write those three and refuse the fourth.

**Grant.** A number of options or restricted stock units. An option is a right to buy at a fixed strike price, so it is worth the difference between the eventual price and the strike and can be worth nothing. A unit converts to a share on vesting and is worth whatever that share is worth.

**Vesting schedule.** The period over which the grant is earned. A common shape is four years with a one-year cliff: nothing at all vests before the twelve-month mark, then a quarter vests at once, and the remainder vests monthly or quarterly. The cliff is what makes the first year binary.

**Refresh.** Additional grants made during employment, usually at annual review or promotion. This matters more than the initial grant, because compensation in years three and four is largely refresh. Ask for the refresh policy in writing. A large initial grant with no refresh produces a cliff at year four that people do not see coming.

**Why a private grant cannot be valued as cash.** The figure a recruiter quotes is usually the quantity multiplied by the most recent preferred share price from the last funding round, or by the spread between that price and your strike. Employees hold common stock. Investors hold preferred stock, which typically carries a liquidation preference and is paid first. In any outcome below the preference stack, common is worth substantially less than the quoted per-share figure and can be worth nothing. There is also normally no market: you cannot sell, and you cannot borrow against it safely.

**Exercise cost and the exercise window.** Exercising options costs cash: quantity multiplied by strike. Leaving usually starts a post-termination exercise window, commonly ninety days, after which vested but unexercised options expire, though some employers extend it to several years. Vested equity is therefore equity you keep only if you can write a cheque within three months of leaving.

**Tax.** The US Internal Revenue Service distinguishes statutory options, meaning incentive stock options and employee stock purchase plan options, from nonstatutory options. For a nonstatutory option with no readily determinable fair market value there is no taxable event at grant, and you include the fair market value of the stock less the amount paid when you exercise. Exercising an incentive stock option may trigger alternative minimum tax. That is the shape of it and it is not advice: the interaction of exercise timing, holding periods and alternative minimum tax is where the money actually is, and it needs a qualified tax professional.

The two questions that make a grant comparable at all: what percentage of fully diluted shares outstanding does this grant represent, and what is the current common valuation used for strike pricing. If the employer will not answer, record the refusal in the brief as a fact about the offer.

## Step 7. Script the exchange, and make one ask

Four sentences do most of the work.

**Receiving without accepting.** "Thank you, I am glad we got here. Could you send the full breakdown in writing: base, bonus target, the equity grant with the number of units, the strike and the vesting schedule, any sign-on, and the level. I will come back to you by [date]." This buys time and produces the document everything else depends on.

**Asking for the band.** "What is the range for this level, and where does this offer sit inside it?"

**The consolidated ask, made once.** "I have looked at the whole package. If you can do [base] and [sign-on], and move the start to [date], I will sign." One message, everything in it, and a commitment attached.

**Responding to a final offer.** "I understand. Is the base final, or the package final?" That single question separates a fixed component from a movable one, and it is the highest-yield sentence in the brief.

The consolidated rule is not politeness. Sequential asks reset goodwill each time, because each new one means the previous yes bought nothing, and the person who granted it has to return to the same approver twice. That is where the reputational cost lands, and it lands on the manager you are about to work for.

## Step 8. Deadlines, and the decision rule

Ask for the deadline in writing along with the reason for it. Some deadlines are real: a requisition that closes at quarter end, a start date tied to a project, a conversion cycle with a fixed calendar. If it is real, the honest ask is specific: "I have a process concluding on [date]. Can you hold until then." Give the date, never "a few weeks".

If they will not extend, you have a decision rather than a bluff to call, and you compare the offer to your reservation, not to the offer you hope the other process produces. Accepting and then reneging is legal in most at-will arrangements, burns a bridge in a small industry, and can trigger a short-tenure clawback.

**The decision rule.**

- **Below reservation.** Counter or decline. There is nothing to lose, so counter at target with your best evidence and state the position plainly.
- **Above target and near the top of a published band.** Take it. Spend the remaining goodwill on start date, review timing or level, not on base, because base is where the exception approval lives.
- **Between reservation and target.** Counter once, consolidated, with evidence, and accept the answer whatever it is.
- **You cannot tell, because there is no band and no comparator.** Do not counter. Ask for the range for the level and the fully diluted percentage of the grant first. A counter made blind becomes the ceiling, and this is the single most common way people anchor themselves down.

## Worked example

A mid-size logistics software company, privately held, hiring a senior backend engineer in a jurisdiction that requires a posted range. The posting says 145,000 to 185,000.

The brief, written before the call. Current position: base 144,000 with a 10 per cent bonus target, so 158,400 total cash, and vested equity with no liquidity. Reservation: 160,000 total cash, because below that the move is a pay cut for more risk. Target: 175,000 base, which is 192,500 total cash, traced to a posted midpoint of 165,000 and a second written offer at 172,000 base. Anchor: 182,000, traced to the same two facts plus on-call ownership in the description that the other role does not carry.

The offer arrives: base 152,000, bonus target 10 per cent, 12,000 share options at a strike of 4.10 over four years with a one-year cliff, no sign-on, start in three weeks. Total cash 167,200, which is above reservation and below target, so the rule says counter once.

The consolidated ask, sent as one message: base 175,000, a sign-on of 15,000, a start date four weeks later so a bonus at the current employer vests first, and the refresh policy in writing.

The outcome. Base moves to 168,000, and the recruiter says the band for this level tops out at 170,000 and going above needs a vice-president exception. The sign-on is granted at 15,000 with a pro-rated twelve-month clawback. The start date is granted. The refresh policy is disclosed as an annual grant at review.

One component turned out to be structurally fixed: the bonus target is attached to the level at 10 per cent and is not a per-person number, so an ask for 15 per cent was never available and would have consumed one of a small number of asks.

The equity is recorded, not valued. The recruiter quoted "about 195,000 of equity", which is 12,000 multiplied by the 16.25 spread between the last preferred price and the strike. The brief instead records: 12,000 options, strike 4.10, four-year vest with a one-year cliff, ninety-day post-termination exercise window, 0.04 per cent of fully diluted shares, exercise cost of 49,200 in cash to take all of it, and common stock sitting behind the preference stack.

**Verdict: accept, below target.** The final package is 184,800 total cash against a reservation of 160,000 and a target of 192,500, so the target was not met. It is accepted anyway, because the rule for an offer between reservation and target is one consolidated counter and then acceptance, and because the disclosed band ceiling of 170,000 puts the final base 2,000 under a hard edge. The remaining gap is a level question, not a base question, and level questions are settled at the screen.

The line recorded at the bottom of the brief: the posted range ran to 185,000 while the band for this level topped at 170,000, so part of the target had been traced to somebody else's level. A posted range that spans levels is not the band, and treating it as one sets a target that no approver could have granted.

## Failure modes

**Salami Asking.** Three separate small asks over four days. From the outside it reads as bad faith, because each yes turns out to have bought nothing, and the approver is asked twice for the same headroom.

**Reservation Drift.** The walk-away number quietly moves down after the offer arrives. It is recognisable because the new number is always just below the offer. A reservation that moves was a preference, and the dated note in the brief is the only defence.

**Anchoring Yourself Down.** A number given at the screen, months before the offer, in answer to a friendly question. Everything afterwards is an argument about how far above your own number you deserve to be.

**Component Blindness.** Four exchanges spent fighting base in a company with published bands, while the sign-on, the start date and the review timing, all of which were available, go unasked.

**Equity Illusion.** A private company grant entered into the comparison at the recruiter's quoted figure. Two offers get compared as though one pays more, when one of them is paying partly in an instrument that has no market, sits behind a preference stack, and may cost money to keep.

**Deadline Panic.** A stated expiry treated as a starting gun, so the written breakdown is never requested and the offer is accepted against a number nobody wrote down.

**Bridge Burning.** A negotiation that extracts two per cent and costs the relationship with the manager you are about to work for, who is also the person who will write your first calibration input.

**The Verbal Offer.** Negotiating hard against numbers that only ever existed in a phone call, where the level, the vesting schedule and the clawback terms were all assumed rather than read.

## What this skill does not do

- It is not legal advice and it is not tax advice. Pay transparency rules, salary history bans and equity taxation vary by jurisdiction and change, and equity taxation specifically needs a qualified professional.
- It cannot value private company equity and will not pretend to. It records the grant and prices it at zero in the decision.
- It cannot see the employer's band, their approval limits, or how many candidates are at offer stage. It works from evidence you hold.
- It has no view on whether you should take the job, only on the terms of one you have already decided you want.
- It does nothing for an offer where the level was set wrong, which is a screening problem that had to be fixed before the loop.
- It does not cover the screen or interview preparation, both of which are out of scope here and both of which decide more of this outcome than the call itself.
Why import instead of copy

A skill is only as good as what it can read.

These skills all ask your assistant to check things against your actual codebase, your actual schema, your actual design system. Locul keeps that context current on its own, from the files you already have, on your machine. Mac and Windows, free to start.

Start free