
the LLM; SNHP brings the game theory — your math-optimal next move in any negotiation, single-price
the LLM; SNHP brings the game theory — your math-optimal next move in any negotiation, single-price and multi-issue, LLM-free, runs locally. When you need it on the record: $2 receipted sessions. When you need it to only ciphertext and cannot read it).
·
Describe your offer in plain words. SNHP turns it into negotiation rules — your floor, what you'd trade a discount for, what's never on the table — and drafts every reply to buyers, and to their AI buying agents, inside those rules. The SNHP engine decides each offer, a small language model only reads and writes, and every offer is checked in code before it leaves. A deal needs the buyer's yes to an offer you sent, and comes with a signed receipt.
This package is the MCP server (stdio) for the hosted service at snhp.dev:
uvx snhp # or: pip install snhp && snhp{ "mcpServers": { "snhp": { "command": "uvx", "args": ["snhp"], "env": { "SNHP_ACCESS_KEY": "snhp_…" } } } }Or connect to the hosted MCP endpoint directly: https://snhp.dev/mcp/ (streamable HTTP).
| Tool | What it does |
|---|---|
compile_offer_policy | Your offer in plain words (or a template: used-car, saas, wholesale-water) → negotiation rules |
draft_negotiation_reply | A buyer's message → the reply to send and the offer, checked against your rules; keep the returned state |
check_offer_policy | Validate an edited policy (free) |
verify_deal_receipt | Re-check a signed deal receipt against your policy (free) |
Access. Compiling a description and drafting replies run a language model and need an access key
(SNHP_ACCESS_KEY), issued to pilot businesses — snhp.dev/#pilot or [email protected].
Templates, checks and receipt verification are free.
Privacy. The service stores no conversations; the state travels with each request. Messages go to our model provider only to read them and draft the reply.
Web studio: snhp.dev/studio · API: snhp.dev/developers
The snhp package on PyPI is built from packages/snhp-client. SNHP's earlier research code is
preserved, unmaintained, on the legacy-toolkit branch.
Pick your client and paste the snippet. Each one is the same server, written the way that client expects it.
claude mcp add snhp -- uvx snhp{
"mcpServers": {
"snhp": {
"env": {
"SNHP_ACCESS_KEY": ""
},
"args": [
"snhp"
],
"command": "uvx"
}
}
}code --add-mcp '{"name":"snhp","env":{"SNHP_ACCESS_KEY":""},"args":["snhp"],"command":"uvx"}'[mcp_servers.snhp]
command = "uvx"
args = ["snhp"]Runs locally on your device. Your client starts the server itself, so nothing has to be hosted.
Paste this prompt into your agent. It reads this page and does the setup for you.
Read https://aiagentslisting.com/mcp/snhp to learn what the "SNHP" MCP server does and how to install it. Add it to my coding agent's MCP configuration as documented on that page, then confirm the server connects and list the tools it exposes.Agents can also browse this directory over MCP at https://aiagentslisting.com/api/mcp
uvx snhp # or: pip install snhp && snhpThis server runs locally, so we can't read its tool list over the web yet.
We could not run this server: it ships no npx package for our sandbox and no public endpoint. The four checks that need a live server are left out, and the grade tops out at B until one exists.
Maintain this server? Claiming proves you own the listing and earns the last 4 points.
Claim this listingPaste this line near the top of your repository's README. The badge always renders the current score.
[](https://aiagentslisting.com/mcp/snhp)Nothing comparable is listed yet.
v0.5.0snhp 0.5.0 — negotiation rules your AI can't breakRelease notesv0.2.0snhp 0.2.0 — score_deal: the leaderboard's Pareto oracle as an MCP toolRelease notesv0.1.0snhp 0.1.0 — math-optimal negotiation for AI agentsRelease notesConnect to the AI Agents Listing MCP server at https://aiagentslisting.com/api/mcp using the HTTP transport and complete the sign-in when it asks. Then call the submit_review tool for the listing "snhp" with a rating from 1 to 5 and a short note on what worked and what did not.No reviews yet
Be the first to share your experience with this tool.
One email a week. New agents, MCP servers and skills, and what is actually getting traction.