Maps venue exit costs, market depth, liquidity tiers, and stress indicators through read-only MCP tools.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent β or use 1-click editor setup below.
Clients with native remote MCP support can connect directly to this URL instead of the install method above.
π‘ Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Undertow β market liquidity map.
agent_access_statusfree
board_fullsubscriber
corporate_transmissionsubscriber
crypto_exit_checkfree
crypto_funding_costfree
crypto_portfolio_stressfree
Undertow provides market-liquidity and exit-cost evidence through MCP. Its tools cover per-venue selling costs, quoted-depth concentration, observed depth-collapse episodes, market-segment liquidity tiers, and selected funding or exposure scenarios. Results are read-only and are intended for analysis rather than order execution.
The public surface includes tools such as exit_cost, crypto_exit_check, venue_concentration, depth_episodes, liquidity_tiers, unwind_watch, trade_safety_exit_context, sealed_record, and agent_access_status. Subscriber tools add broader views including exit_desk_full, exit_schedule, board_full, corporate_transmission, household_credit, divergence_status, tide_clock, and unwind_stress. Access rights are checked for subscriber requests.
The hosted endpoint is https://api.seiche.info/undertow/mcp and uses streamable HTTP. Public requests do not require an API key or wallet. The service publishes observation times, venue inputs, size rungs, coverage gaps, and method details with relevant results.
Exit estimates are based on published size rungs. exit_cost reports a nearest-rung per-venue cost in basis points, approximate dollar impact, the cheapest and most expensive observed venues, and the spread between them. crypto_exit_check compares up to four requested BTC sell sizes against published rungs and does not interpolate between them. These figures are estimates, not executable quotes, and venue rankings can change during the day.
Other tools calculate caller-supplied funding costs or portfolio shocks, summarize venue concentration, identify realized depth-collapse episodes, and assign liquidity tiers to segments including U.S. Treasuries, investment-grade and high-yield credit, equities, ETFs, FX, China-related markets, and crypto. The tier response can include measure counts, replay eligibility, and a stated reason when a score is withheld.
The simplest connection uses the hosted URL directly. Claude Code can add it with:
Claude Desktop can use the repositoryβs stdio adapter. Local setup requires Python 3.12 or newer and uv:
Docker is also supported through the repositoryβs root Dockerfile. The adapter requires outbound HTTPS access to api.seiche.info; it does not need a key, bearer token, wallet, port, volume, or environment setting. The hosted service is the route to subscriber tools, while the stdio adapter exposes the anonymous public surface and guided prompts.
Undertow β market liquidity map does not provide executable prices, trading, wallet access, or order placement. Published estimates can be stale or affected by depth and coverage gaps, so the observation clock and venue inputs should be reviewed before using a result.
The adapter disables redirects, environment proxies, and arbitrary upstream URLs. Requests are limited to 64 KiB, responses to 2 MiB, and each upstream operation to 10 seconds including queue wait. It does not automatically retry requests or cache responses. Upstream version or catalog drift fails closed, while tool results, errors, and entitlement refusals are preserved.
The hosted protocol supports several MCP revisions, including the 2025-03-26, 2025-06-18, 2025-11-25, and 2026-07-28 behavior described by the project. The service exposes no MCP resources or resource templates. Settlement histories and source receipts are provided through a separate source-data MCP rather than this listing.
Factual signals from GitHub, npm, and our automated checks β not a rating.
No reviews yet β be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/undertow-market-liquidity-map-2)<a href="https://allmcps.com/mcp/undertow-market-liquidity-map-2"><img src="https://allmcps.com/api/badge/undertow-market-liquidity-map-2?style=directory" alt="Undertow β market liquidity map on AllMCPs" /></a>