agentleFS
nishant-Tiwari24nishant-Tiwari24/cracked/skills/strategy/offer-decode/SKILL.md

offer-decode

Skill5 starsChanged 8 days agoMITView on GitHub

Decode a tech job offer or CTC breakup into what the candidate actually takes home. It separates fixed, variable, one-time and deferred pay (ESOPs/RSUs), estimates monthly in-hand salary with stated assumptions, flags red flags (bonds, clawbacks, inflated CTC, notice period, non-competes), compares multiple offers side by side, and drafts a negotiation reply. Use when the user has an offer letter, a CTC breakup, wants to compare offers, asks "what is my in-hand salary", or wants to negotiate an offer.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call the Skill tool with "cracked-profile" (country.

    line 12
  2. Distinctive:

    If you're comparing offers, collect all of them before analysing.

    line 14
  3. Distinctive:

    Tax rules change every year: If you have a web tool, verify the current slabs for the relevant financial year before computing.

    line 39
  4. Distinctive:

    Run arithmetic with the calculator or the shell, never in your head, and show the result.

    line 41
  5. Distinctive:

    Check and report each one:

    line 46

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Offer decode

"₹18 LPA" can mean ₹1.05 lakh a month in hand, or ₹80,000 with a ₹2 lakh bond. Candidates routinely pick the worse offer because of how the CTC was written. Decode it properly.

0. Inputs

  1. Call the Skill tool with "cracked-profile" (country; current location and city matter for rent and professional tax).
  2. Get the offer(s): offer letter PDF, CTC annexure, a screenshot, or typed numbers. Read every page. Bonds, clawbacks and notice periods hide in annexures and the fine print.
  3. If you're comparing offers, collect all of them before analysing.

1. Break it down

For each offer, build this table (annual and monthly):

ComponentAnnualMonthlyCategory

Categories:

  • Fixed cash: basic, HRA, special allowance, other fixed allowances
  • Variable: performance bonus. Note the target % and whether it's guaranteed. Ask the user for the company's historical payout. If it's unknown, show 0%, 70% and 100% payout side by side and label it an assumption (the calculator does this).
  • One-time: joining/sign-on bonus, relocation. Check for clawbacks ("repayable if you leave within 12 months").
  • Deferred / equity: ESOPs, RSUs. Note the vesting schedule, cliff, strike price (ESOPs), and liquidity (private company ESOPs may never be sellable).
  • Benefits counted in CTC: employer PF, gratuity, insurance premiums, meal cards, NPS. These are real money but not in-hand cash.

Year-1 cash = fixed + expected variable + one-time. Recurring cash = fixed + expected variable. Show both. The headline CTC is neither.

2. Estimate in-hand

For India, use INDIA-CTC.md and run the calculator rather than doing the maths yourself: python3 scripts/ctc.py --fixed <annual fixed> --variable <annual bonus> --regime both [--rent <annual rent> --metro] [--pt <professional tax>] [--pf capped|full] [--wage-rule basic|labour-code]. Paste its table. If you change an assumption, rerun it. For other countries, estimate with the main tax and social-security components (e.g. US: federal + state + FICA, 401k if stated), and say which ones you included.

Rules:

  • Tax rules change every year. If you have a web tool, verify the current slabs for the relevant financial year before computing. If not, use the reference values, name the year they're from, and say so clearly.
  • Show the calculation step by step, and state every assumption (tax regime, PF on full basic vs capped, professional-tax state, rent).
  • Run arithmetic with the calculator or the shell, never in your head, and show the result.
  • End with a monthly in-hand estimate as a range (e.g. ₹1.02–1.06 L), not false precision.

3. Red flags and fine print

Check and report each one:

  • Service bond / training bond: amount, duration, enforceability concerns. State the facts and suggest they ask HR to clarify. You're not a lawyer.
  • Clawbacks on joining bonus or relocation.
  • Variable % that's high relative to fixed, or with vague payout conditions.
  • Notice period (60–90 days hurts the next switch), probation terms, non-compete / non-solicit.
  • Inflated CTC: insurance, gratuity, or one-time items padding the headline.
  • Joining date and location flexibility; relocation support; work mode.
  • ESOPs at early startups: high strike, short exercise window, or a 4-year vest with a 1-year cliff. Treat them as a lottery ticket, not salary.

4. Compare (if more than one offer)

Side-by-side table: recurring cash, year-1 cash, monthly in-hand, equity (with a "realistic value" note), bond, notice period, role/team, learning, location and cost of living (adjust in-hand for rent differences between cities), and brand/growth. Then a recommendation with reasoning, and the one factor that would flip it.

5. Negotiate (if they want to)

Draft a short, polite email or message: appreciation → a specific ask (a fixed-pay increase, joining bonus, waived bond, relocation, joining date) → justification (competing offer, market data the user provides, or their unique fit) → flexibility. Note that campus placement offers are usually non-negotiable; off-campus and lateral offers usually are.

6. Output

Save the analysis to $CRACKED_DIR/offers/<company>-<date>.md. No scorecard for this skill.

End with: "Estimates for planning only, not tax or legal advice. Confirm the final numbers with HR."


Skill calls: if your agent has no Skill tool, open the named skill's SKILL.md in the cracked skills folder (core/, interview/, profile/, strategy/) and follow it instead.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.