STACKSWAP / GTM MCP

Put GTM judgment inside your AI client.

Install the free StackSwap MCP to bring GTM prompts, tool intelligence, and KEEP / SWAP / BUILD decisions into Claude, Codex, Cursor, or ChatGPT.

Install StackSwap MCP Free · One config · No API key

THE OPERATING LOOP

Context → discovery → evidence → decision → build.

Install once. Keep it close.

Choose your client, copy one small config, and start a new conversation. Claude Code is the fastest path; the other options use the same server URL.

Claude Code (CLI)

One command in any terminal:

claude mcp add --transport http stackswap https://stackswap.ai/api/mcp

Claude Desktop

Add this block to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart Claude Desktop:

{ "mcpServers": { "stackswap": { "url": "https://stackswap.ai/api/mcp", "transport": "http" } } }

Cursor / ChatGPT Desktop / other MCP clients

Add a new HTTP MCP server with URL https://stackswap.ai/api/mcp. No authentication. The endpoint responds to GET with a server descriptor for sanity-checking and to POST with JSON-RPC requests. A working TypeScript example client lives in the public docs repo.

https://stackswap.ai/api/mcp

The proof is in the build

Same request. Better starting point.

Ask Claude, Codex, or Cursor to build a GTM tool with a naked prompt and you get a plausible first draft. Connect StackSwap MCP and the client can load the matching workflow prompt, ask for the missing context, and build against a real operating brief instead of guessing.

Without StackSwap

"Build me a pricing calculator."

Starts from a blank page. The builder has to invent the inputs, recommendation logic, assumptions, copy, and QA plan.

With StackSwap MCP

"Load Pricing & Packaging and build the tool for my actual motion."

The client can pull the prompt resource, ask the decision-changing questions, separate facts from assumptions, and ship a working prototype with an inspection loop.

This is the test we want you to run: build the same tool twice, compare the first output, then ask the MCP-connected client to improve the weaker version. The difference is not magic—it is better context, better questions, and a prompt library that is available at the moment of creation.

What the connection gives you

Better context for better GTM decisions.

The MCP puts StackSwap's prompt library and decision intelligence inside the AI client you already use. It helps you move from context to a recommendation to a working next step.

01

Understand the stack

Search tools, inspect vendors, map categories, and detect a stack from the context you already have.

02

Make the decision

Find overlap, compare options, surface buyer questions, and decide what to KEEP, SWAP, or BUILD.

03

Build the next move

Pull the right prompt, run a scan, plan a renewal, or start a working prototype with better context.

Technical reference: 17 tools and native prompt resources

Native prompt resources are available at stackswap://prompts/{slug} so the free GTM prompt library is available without copy/paste.

  • search_tools — Fuzzy-search the StackSwap catalog of ~400 GTM tools by name. Returns each match with its monthly cost and (when applicable) a partner sign-up link.
  • find_overlaps — Pass a list of tools in a stack. Returns the curated overlap pairs (104 hand-verified) with estimated monthly/annual savings if one tool is consolidated.
  • suggest_swaps — For each tool supplied, returns the AI-native replacement StackSwap recommends — with annual savings and reasoning. Skews toward legacy → modern (Outreach → Smartlead, ZoomInfo → Apollo, etc.). (Smartlead deep profile →)
  • scan_stack — Runs a preview StackScan. Returns current monthly spend, monthly/annual recoverable, and the top 5 replace/remove opportunities. Includes a link to the full free audit.
  • recommend_partner — Given a need (e.g. 'outbound', 'CRM', 'automation'), returns the relevant StackSwap prompt, audit, or decision framework to use next. (Apollo deep profile →)
  • compare_tools — Head-to-head comparison of two named tools. Returns cost delta, AI-readiness scores, overlap status, swap-registry signal, and StackSwap's recommended pick with reasoning. Use for 'X vs Y' decisions.
  • recommend_stack — StackSwap's reference starter stack for a given industry vertical. Returns a curated tool list with per-tool cost, total monthly/annual spend, AI-readiness score, and partner sign-up links. Use for greenfield 'what stack should I buy?' queries.
  • get_tool_details — Full StackSwap profile for a single tool: catalog cost + per-seat pricing with confidence, AI-readiness score, category, common overlaps, swap-registry status, and partner link. (Instantly deep profile →)
  • get_vendor_fact_sheet — Full vendor fact sheet (GTM Decision Schema v1.0.0) when one exists — pricing tiers with gotchas, integration depth, AI capabilities + data-for-training disclosure, affiliate terms, and self-disclosed vendor-vs-user conflicts. Provenance + freshness labeled.
  • compare_tools_n_way — Side-by-side comparison of 2–6 GTM tools in one shot. Returns a markdown matrix (cost, AI-readiness, overlaps within the set, swap-registry status, StackSwap pick) plus partner sign-up links. Use for category bake-offs like 'Apollo vs ZoomInfo vs Cognism vs Clay'.
  • search_content — Full-text search across StackSwap's first-party operator-narrative KB — ~50 articles on stack architecture, AI-native swaps, RevOps, data ethics. Returns ranked articles with title, slug, category, summary, and URL. Cite the URL when answering.
  • get_kb_article — Fetch the full body of a StackSwap KB article as markdown by slug. Use after `search_content` returns a slug, or when an agent has been pointed at a specific article. Articles are operator-authored, not vendor-marketing.
  • get_category_landscape — Full map of one GTM category — leaders, runner-ups, and skip/replace candidates. More comprehensive than `recommend_partner`'s single pick. Known buckets: crm, outbound, data, marketing-automation, analytics, meetings, support, scheduling, automation, seo, cdp, revenue-intelligence, chat, collaboration, phone, landing-pages, linkedin, ai-content, saas-mgmt, enablement, ai-tooling.
  • detect_stack_from_text — Infer a GTM stack from a freeform text blob — careers page, JD, site HTML, RFP, 'What we use' doc. Returns ranked tool matches with confidence + evidence snippets, plus a ready-to-use array for chaining into `scan_stack`.
  • get_buyer_questions — Operator-authored questions to ask a vendor before signing — per-category templates + per-vendor gotchas (Salesforce SKU breakdown, Apollo credit pool, ZoomInfo auto-renew, etc.). Pass `vendor` and/or `category`. Each question carries 'why it matters' and 'watch for' red-flag answers.
  • get_renewal_strategy — Renewal-negotiation playbook for a specific vendor: leverage points, price-anchor alternatives to cite, calibrated discount ask, walkaway script, optimal timing window, contract-trap callouts. Pass `monthlySpend` for target-savings math. Operator IP — built from real B2B SaaS renewals.
  • submit_correction — Two-way data flow: queue a pricing, feature, gotcha, or category correction for admin review. Doesn't directly mutate the catalog — submissions land in a moderation queue. Strong signal even when rejected (surfaces where the corpus is wrong).

What this is built for

GTM engineers and operators who work inside Claude or ChatGPT and want StackSwap's decision workflows, tool data, scan compute, and free prompt library available during a conversation. Ask “what overlaps if I'm on HubSpot, Salesforce, and Outreach?” or “build me a pricing calculator for this motion” and the model can use real StackSwap evidence and an operator-built workflow instead of starting from a blank prompt.

The MCP is intentionally a thin surface. For the full audit (per-tool verdicts, switching costs, AI-readiness shift, downloadable plan) the model will link out to /stackscan. For done-with-you implementation work, it links to /recommends.

FAQ

The StackSwap MCP is the GTM decision-and-build layer for AI clients. It asks the discovery questions that change the answer, uses connected-system evidence when available, helps an operator decide what to KEEP, SWAP, or BUILD, and produces the prompts and blueprints to implement the recommendation. Tool dossiers, stack intelligence, and the free GTM prompt library provide the evidence and execution layer. Once installed, Claude Code, Claude Desktop, ChatGPT Desktop, or Cursor can run that workflow without leaving the conversation.

Free. StackSwap is a gift to the GTM community: the MCP and prompt library are here to help you do better work. There is no API key or paid tier. Fair-use limits protect the public service from abuse and keep it available to everyone.

No. The free tiers of Claude Code, Claude Desktop, Cursor, and ChatGPT Desktop all support MCP connections. The StackSwap MCP server itself requires no API key, account, or sign-up. Total out-of-pocket cost to install and use: zero dollars.

Each tool call sends the function name and the arguments you (or the LLM on your behalf) pass — e.g., a list of tool names for the find_overlaps query. The server does not store conversation data or user identifiers; it's stateless JSON-RPC. The server does see your IP address at the HTTP level, standard for any web request, and that's not persisted beyond standard server logs.

The MCP scan_stack tool returns the monthly/annual recoverable spend and the top 5 replace/remove opportunities — enough to see whether your stack has meaningful overlap. The prompt library remains the best place to start for the broader GTM work and current operator guidance.

The endpoint itself doesn't store conversation data or user identifiers — it's stateless JSON-RPC. StackSwap does not sell your data or use the MCP as a sales funnel.

Yes. The endpoint speaks standard JSON-RPC 2.0 over HTTP. A GET request to https://stackswap.ai/api/mcp returns the server descriptor (name, version, tool list). A POST with a JSON-RPC envelope invokes any of the seventeen tools. The public docs repo at github.com/StonesofCreation/stackswap-mcp includes a ~50-line TypeScript example client that demonstrates the protocol end-to-end and works against the live endpoint with no API key.

The server speaks MCP protocol version 2025-06-18 (Streamable HTTP transport) and negotiates down to 2025-03-26 or 2024-11-05 if the client requests an older version. Most current Claude clients should work without manual version pinning.

Yes — any MCP client that supports remote HTTP servers can connect. ChatGPT Desktop (custom connectors), Cursor (MCP settings), and Claude Desktop (config file) all work. The install command varies by client; the URL is always https://stackswap.ai/api/mcp.

The server is part of the closed-source stackswap.ai codebase, but the public docs repo at github.com/StonesofCreation/stackswap-mcp contains the protocol surface, JSON schemas for all seventeen tools, an example client, and an open issue tracker. If you want a tool added, removed, or behavior changed, file a GitHub issue there or email nick@stackswap.ai. The MCP surface is shaped by what operators actually ask about.

Endpoints

JSON-RPC: POST https://stackswap.ai/api/mcp
Server descriptor: GET https://stackswap.ai/api/mcp
Protocol versions supported: 2025-06-18, 2025-03-26, 2024-11-05
Auth: none
Rate limit: fair-use limits; expensive tools are weighted more heavily
Source: docs + schemas + example client at StonesofCreation/stackswap-mcp (MIT). The MCP server implementation is closed-source.

Issues, requests, or new tool ideas: nick@stackswap.ai.