agentleFS
Sign inStart free
getedgehqgetedgehq/skills/negotiation-prep/SKILL.md

negotiation-prep

Skill5 starsChanged 12 days agoApache-2.0View on GitHub

Prepare for and run a negotiation with structure: BATNA and ZOPA analysis, counterparty analysis, a 6-dimension score for any draft message, a pre-send checklist, and a concession ledger across rounds. Use for salary offers, vendor and supplier contracts, billing disputes, partnership terms, and any deal where the next message matters. Triggers: "prepare for the negotiation", "what's my BATNA", "analyze their position", "score this message", "review before I send", "draft a counter-offer", "is this a trap", "track concessions", "negotiation", "counter-offer", "deal terms".

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 10

    Distinctive

    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:

    Never state a fact that has not been verified: If a number, date, clause or claim is not in the user's material, mark it as an assumption …

    line 23
  2. Distinctive:

    Before any analysis, establish:

    line 29
  3. Distinctive:

    Ask for the actual documents if they exist.

    line 36
  4. Distinctive:

    Read the counterparty's message or position

    line 43
  5. Distinctive:

    Run SCORE on your own draft before presenting it

    line 61
  6. Distinctive:

    Before any message goes out, run the checklist:

    line 80
  7. Distinctive:

    Never open at your final number: Leave room to concede to where you actually want to land.

    line 93
  8. Distinctive:

    What is not said matters: If a clause that favors you is missing from their draft, you do not have to point it out.

    line 95
  9. Distinctive:

    Every concession needs a trade: Never give something for nothing.

    line 96
  10. Distinctive:

    "Standard" is not an argument: Ask for specifics, or argue the substance instead.

    line 99

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “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

Negotiation Prep

Structured preparation and message review for negotiations where the next move matters.

Inspired by the negotiation preparation approach taught in the CEMS Negotiation Strategy course at Nova SBE.

Hard rule

  • Never reveal the user's BATNA, walk-away point or reservation price in any draft, directly or by implication.
  • Never state a fact that has not been verified. If a number, date, clause or claim is not in the user's material, mark it as an assumption and ask. Do not invent market data, precedents or "industry standard" figures.

Core Workflow

1. Context Loading

Before any analysis, establish:

  • Who are the parties, and what is their relationship (one-off or repeated)?
  • What has been agreed so far? Look for written evidence: messages, emails, contracts, offer letters.
  • What is in dispute, and what is not yet on the table?
  • Which channel is this on? (chat = concise, email = structured, call = flexible)
  • What is the timeline? Deadlines, meetings, renewal dates, external events.

Ask for the actual documents if they exist. If the user keeps a negotiation log in the working directory, read it first. State any assumption you make.

2. Mode Selection

Pick the mode from the request. If unclear, ask, or default to PREPARE.

ANALYZE - "analyze their position", "what do you make of this"

  • Read the counterparty's message or position
  • Detect contradictions across rounds
  • Separate position (what they say) from interest (what they need)
  • Assess power dynamics
  • Flag emotional vs. rational arguments
  • Output: structured analysis with actionable insights

SCORE - "score this", "review before I send"

  • Apply the 6-dimension rubric (see references/framework.md): Information Leakage, Anchoring, Tone, Logic, Leverage, Trap Detection
  • Score 1-10 per dimension plus overall
  • Quote the specific phrases that are problematic
  • Suggest a concrete rewrite for each flagged phrase
  • Output: score card + fixes + a rewritten version if the overall score is below 8/10

DRAFT - "draft a reply", "write a counter-offer"

  • Structure: opening (sets tone) + substantive points + close
  • Concede from your own anchor, never from theirs
  • Run SCORE on your own draft before presenting it
  • Output: ready-to-send message + score + rationale for the key choices

TRACK - "what have I conceded", "where do we stand"

  • Build or update the concession ledger across all rounds
  • Net position: given vs. received
  • Flag concession fatigue (giving more per round)
  • Identify remaining leverage and room to move
  • Output: concession table + strategic assessment

PREPARE - "prepare for the call", "what's my strategy"

  • BATNA and ZOPA analysis for both sides (see references/framework.md)
  • Points to raise vs. points to hold back
  • Likely counterarguments and prepared responses
  • Walk-away line and fallback positions
  • Output: one-page prep doc with talking points and red lines

3. Pre-Send Gate

Before any message goes out, run the checklist:

  1. INFORMATION: Am I revealing anything usable against me?
  2. ANCHORING: Am I reinforcing my frame or theirs?
  3. ROOM: Do I have space to move if they push back?
  4. TONE: Would this read well if a neutral third party saw it?
  5. OMISSIONS: Is anything conspicuously absent that I want to stay absent?
  6. CONTRADICTIONS: Does this contradict my position elsewhere?
  7. INVITATION: Am I inviting a response I don't want?
  8. TIMING: Is now the right time to send it?

Key Principles

  • Never reveal your BATNA. The moment they know your alternative, they price it in.
  • Never open at your final number. Leave room to concede to where you actually want to land.
  • Separate the people from the problem. Firm on substance, warm on the relationship.
  • What is not said matters. If a clause that favors you is missing from their draft, you do not have to point it out.
  • Every concession needs a trade. Never give something for nothing.
  • Detect false urgency. "We need to close today" usually means they need to close today.
  • Watch for position shifts between rounds. A sudden change in tone often means they consulted someone.
  • "Standard" is not an argument. Ask for specifics, or argue the substance instead.

References

  • Scoring rubric, BATNA/ZOPA templates, concession ledger, pre-send checklist: references/framework.md
  • Worked examples and anti-patterns: references/examples.md

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.

Check your setup, free