Skip to content

YieldSignal MCP server

io.github.Stakemate369/yieldsignalStakemate369streamable-httpv1.2.0Scanned

Yield & risk for agents: signal, decision, durability, capacity, sensitivity, exposure, persistence

7 tools: 7 only read.

Tools: 0 delete, 0 irreversible, 0 write, 7 read.
  • Reads private data · 0
  • Reads text others write · 0
  • Can send data out · 0
  • No findings

Sign-in required, no OAuth metadata. Remote posture

Anlyon alternative

Run the operations that matter as governed actions.

Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.

Anlyon Radar does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.

Import an OpenAPI spec →

What the YieldSignal MCP server can do

The YieldSignal MCP server is published by the GitHub account Stakemate369. It has 7 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.

None of its tools reads private data, brings in text other people write, or sends data out.

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Last scanned Oct 21, 2026, version 1.2.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

7 tools

Only reads 7Show 7 read-only tools
  • get_exit_capacityExit capacity for a Base lending market: per-protocol utilization and withdrawable liquidity read from the protocol's own books (Aave, Compound), plus — if you pass amountUsd — whe…No findings

    Why it is in this group

    • The name "get_exit_capacity" says "get": a word for reading.
    Description, as the server wrote it
    Exit capacity for a Base lending market: per-protocol utilization and withdrawable liquidity read from the protocol's own books (Aave, Compound), plus — if you pass amountUsd — whether that size can be withdrawn right now and what share of the market it would be. High APY at high utilization means the market pays well and will not let you out; this tool separates the two. Protocols that do not expose borrowed-vs-supplied are marked unmeasured and are never recommended as executable. USDC only for USD figures; WETH returns utilization without USD (no price oracle in the paid path).
  • get_leadership_persistenceHow long this service's own leadership calls actually hold, measured from its hourly on-chain attestations (EAS on Base).No findings

    Why it is in this group

    • The name "get_leadership_persistence" says "get": a word for reading.
    Description, as the server wrote it
    How long this service's own leadership calls actually hold, measured from its hourly on-chain attestations (EAS on Base). Returns the median duration a protocol stayed the best pick, survival rates at 6/24/72h, how much of the rotation is the same two protocols trading places, and what chasing the leader is worth per $10k before gas. Measured over 24 days: the WETH leader never changed once, the ETH staking leader changed every ~26h, the USDC leader every ~2h with half the switches being a round trip between the same pair — three assets sold at one price with wildly different reliability. Also answers, empirically, whether a wider lead lasts longer: it does not (Spearman -0.04 over 177 completed spells), so gap size is not a usable proxy for confidence. Every input is a public attestation UID: recompute it yourself from base.easscan.org and get the same number. Observed history with sample sizes attached, never a forecast — assets whose lead has not changed in the window are reported as a floor with a censoring flag, never as a median.
  • get_rate_sensitivityHow close a Base lending market is to the kink where borrow rates explode.No findings

    Why it is in this group

    • The name "get_rate_sensitivity" says "get": a word for reading.
    Description, as the server wrote it
    How close a Base lending market is to the kink where borrow rates explode. Returns, per protocol, the current utilization, the kink read from the protocol's own interest rate curve, the headroom in bps, and the borrow APY at points around the kink — plus how many times the borrow cost multiplies just past it. Measured on a live reading: Compound USDC sat 0.17 points below its kink, where borrow cost goes from ~4% to ~16%. Aave and Compound only: Morpho's adaptive IRM has no static curve to read and DefiLlama-sourced protocols expose none, so they are marked unmeasured and never assumed stable. Describes the current state of the curve, not a prediction that utilization will move.
  • get_shared_exposureShared risk exposure across a declared portfolio.No findings

    Why it is in this group

    • The name "get_shared_exposure" says "get": a word for reading.
    Description, as the server wrote it
    Shared risk exposure across a declared portfolio. Give it your positions (protocol:usd pairs) and it returns, per factor, how much of your capital sits behind the same collateral, price oracle or vault curator — and through which venues it gets there. This is the question the market does not answer: depeg and hack alerts tell you an event happened, not whether you are two hops from it. In the Stream Finance collapse only 1 of ~320 MetaMorpho vaults held the broken asset directly, yet $93M of loss became $285M of contagion. Measured live: a Morpho USDC vault sits 93.7% behind cbBTC while Compound's USDC market is 43.1% — holding both is one risk in two wrappers. Morpho is attributed per isolated market and Compound by its real posted-collateral basket; Aave is reported unattributed, because a v3 supplier is exposed to the entire pool and splitting that across assets would imply diversification that does not exist. Structural shared exposure, not a correlation estimate.
  • get_yield_decisionBuyer-side MOVE/HOLD decision (Layer 1 premium — sells the decision, not the raw datapoint).No findings

    Why it is in this group

    • The name "get_yield_decision" says "get": a word for reading.
    Description, as the server wrote it
    Buyer-side MOVE/HOLD decision (Layer 1 premium — sells the decision, not the raw datapoint). Given your current position, size, move cost and horizon, returns whether moving your capital to the best risk-adjusted protocol pays for itself now — with expected net gain, break-even days and a confidence tier. Deterministic from the underlying signal, which is EIP-712 signed and returned in a sibling content block (re-run the decision locally to reproduce it). Priced above the plain signal tool.
  • get_yield_durabilityHow much of the current APY survives if incentives stop.No findings

    Why it is in this group

    • The name "get_yield_durability" says "get": a word for reading.
    Description, as the server wrote it
    How much of the current APY survives if incentives stop. Splits each protocol's yield into base interest vs reward/incentive, reports the post-incentive floor, and says whether the leader changes without incentives. Only protocols whose source itemizes the reward component are decomposed — the rest are listed as undecomposable and NEVER assumed incentive-free, and no ranking claim is made when the current leader is one of them. Also returns bestVerifiableFloor: the highest yield provably independent of incentives. A stress test of readings taken now, not a forecast of when a campaign ends.
  • get_yield_signalReal-time risk-weighted yield signal.No findings

    Why it is in this group

    • The name "get_yield_signal" says "get": a word for reading.
    Description, as the server wrote it
    Real-time risk-weighted yield signal. USDC/WETH: lending APY on Base (Aave/Compound/Morpho read onchain, Moonwell/Euler/Fluid via DefiLlama). ETH_STAKING: liquid staking APY on Ethereum mainnet (Lido/Rocket Pool/Coinbase Wrapped Staked ETH/Frax Ether/Binance Staked ETH, all via DefiLlama) — a different chain and category from the lending signals, not a Base lending market. Source tagged per reading (never estimated). Result is signed (EIP-712 typed data) by the payment-receiving address, returned as a sibling content block for offline verification. That same address is registered on-chain as an ERC-8004 agent identity (agent-card.json) and periodically publishes EAS attestations of past readings (Base mainnet) — a public track record independent of this server's uptime.

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Sign-in required, no OAuth metadata

Looked at Nov 20, 2026, 19:00 UTC, unchanged since Nov 6, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
402
tools/list served without credentials
no
Tools that change things, listed without credentials
0
Protocol version
not observed
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed

The probe stopped: initialize answered HTTP 402.

Posture changes

  1. Nov 6, 2026
    • HTTP status of an anonymous initialize: 200 to 402
    • tools/list served without credentials: yes to no
    • Protocol version: 2025-06-18 to none

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. PostureRemote posture changed. HTTP status of an anonymous initialize: 200 -> 402; tools/list served without credentials: yes -> no; Protocol version: 2025-06-18 -> none.View diff
    • HTTP status of an anonymous initialize: 200 to 402
    • tools/list served without credentials: yes to no
    • Protocol version: 2025-06-18 to none

Install

Pinned to version 1.2.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http yieldsignal https://yieldsignal.vercel.app/mcp
claude mcp add --transport http yieldsignal https://yieldsignal.vercel.app/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__yieldsignal__get_yield_signal",
      "mcp__yieldsignal__get_yield_decision",
      "mcp__yieldsignal__get_yield_durability",
      "mcp__yieldsignal__get_exit_capacity",
      "mcp__yieldsignal__get_rate_sensitivity",
      "mcp__yieldsignal__get_shared_exposure",
      "mcp__yieldsignal__get_leadership_persistence"
    ],
    "ask": [],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as yieldsignal. Review it before you use it.

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/io.github.Stakemate369/yieldsignal/verify?hash=c72c16032d1897cede8047a6ebeca10d274880a6adac866d3c9c50276b99cbc5" | jq -e '.changed == false'

Pinned hash c72c16032d18, from Oct 21, 2026, 19:05 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.Stakemate369/yieldsignal/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.Stakemate369/yieldsignal
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.Stakemate369/yieldsignal.svg)](https://radar.anlyon.com/servers/io.github.Stakemate369/yieldsignal)

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:21 UTC.

YieldSignal MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the YieldSignal MCP server safe?

No scan can promise that. What its definitions show: 7 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.

Can the YieldSignal MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the YieldSignal MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the YieldSignal MCP server have?

7 tools: 7 read. get_yield_signal, get_yield_decision, get_yield_durability, get_exit_capacity, get_rate_sensitivity, get_shared_exposure and get_leadership_persistence.

Does the YieldSignal MCP server need sign-in or OAuth?

Its remote endpoint did not answer the last anonymous look.

Does the YieldSignal MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the YieldSignal MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Oct 21, 2026.