agentleFS
Hrutin-Suriya-KotniHrutin-Suriya-Kotni/until-hired/skills/offer-coach/SKILL.md

offer-coach

Skill2 starsChanged 4 days agoMITView on GitHub

Helps the candidate decide on offers and close the search well: tracks deadlines, compares offers side by side on pay, growth, team and risk, prepares honest negotiation talking points, and runs the wrap-up once they accept. Factual support, not financial or legal advice. Use when an offer or verbal offer arrives, when the candidate says 'I got an offer', 'compare offers', 'how do I negotiate', or 'I accepted'.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If one offer has a deadline and other processes are still running, help the candidate tell the other companies truthfully ("I have an offer …

    line 26

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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 coach

Say once, plainly: you are not a financial or legal adviser; you help the candidate see the facts and prepare, and they decide.

1. Capture the offer

  • File {"next": "decide on the offer"} (agent-setup references/safe-commands.md), then tracker.py update --id N --status offer --next-on <deadline> --json <path>; notify with notify.py (kind offer, priority high).
  • Write <folder>/offer.md (the job's folder from tracker.py folder --id N): fixed pay, variable (and how it's paid), equity (type, vesting, cliff), joining bonus, benefits, location and work mode, title and level, start date, notice buy-out, probation, deadline, and what is only verbal. Ask the candidate to get verbal parts in writing.

2. Compare (when there is more than one option, including "keep searching")

A table with the same rows for each option: guaranteed first-year pay, realistic total, cost of living or relocation, growth and learning, team and manager, product and stability, work mode and hours, commute, risk. Then ask the candidate to weight what matters most to them (from PROFILE.md), and show which option wins under their weights. Their decision, not yours.

3. Negotiate honestly

  • Anchor in the candidate's range from PREFERENCES.md and market data with sources (knowledge/MARKET.md), never the private floor.
  • Talking points: their strongest evidence for this role (from FACTS.md and the interview debriefs), the specific ask (a number or a review date), and a gracious fallback.
  • Only true leverage: mention another offer only if it exists, in the terms it actually has. Never invent offers, deadlines or numbers.
  • Draft the email or call notes; the candidate sends or speaks them.

4. Deadlines across processes

If one offer has a deadline and other processes are still running, help the candidate tell the other companies truthfully ("I have an offer with a deadline of <date>; you're my preference; is a faster decision possible?"). Drafts only.

5. After accepting

  1. File {"evidence": "<signed offer / acceptance email subject and date>"}, then tracker.py update --id N --status accepted --json <path>.
  2. Active processes elsewhere: a short, polite withdrawal draft for each; the candidate sends; mark withdrawn.
  3. Resignation and notice: a checklist (written resignation, handover notes, last day, documents to collect). Don't advise on their contract's legal terms.
  4. Retro: propose it with knowledge.py lesson --kind other --json <path> (what worked in this search, what didn't, the funnel numbers) and let the candidate approve it into knowledge/LEARNINGS.md.
  5. Celebrate it. Then stop the daily routine (tell until-hired).

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.