Routescore

io.routescore由 routescore.io 發布官方發布者v0.3.5更新於 Oct 24, 2026

Read-only pre-sign evidence for onchain agents: route quality, MEV exposure, clear/caution verdict.

已驗證Streamable HTTP可網頁執行金融安全與監控

概覽

AI 產生的概覽

Routescore 為助理提供鏈上兌換簽名前唯讀證據：建模的路徑品質、MEV 曝險以及 clear/caution 判定。

可以做什麼
在簽名前檢查兌換：以 clear/caution/unsupported 判定提供建模的路徑品質、價格衝擊區間、MEV 曝險與代幣辨識結果，並保存可雜湊驗證的紀錄，可用 record_id 取回；另可取得偵測器清單並透過 whoami 確認金鑰與方案。記錄並複核交易後的執行證據，支援冪等批次寫入與依帳號限定的已存紀錄複核。取得 MEV 三明治曝險、跨鏈橋執行失敗與 LRT 罰沒風險的建模保費估算。執行假設性蒙地卡羅情境，比較預期保費與預期損失或退款。僅在明確啟用的非正式環境伺服器上複核任意公開以太坊交易。
適用情境
適合在以太坊或 Robinhood Chain 上簽署鏈上兌換前，讓代理或開發者取得建模的時點性第二意見，或需要可離線複核的決策時點證據。也適合探索建模的風險保費與情境數值，但不用于執行或承保交易。
執行需求
需要 ROUTESCORE_API_KEY（或 Authorization bearer 標頭）中的 Routescore API 金鑰，任何方案皆可免費產生，啟動時會檢查金鑰格式。可作為本機 npm 套件透過 npx 執行，或連線遠端 streamable HTTP 端點。選用 ROUTESCORE_API_URL 可覆寫 API 位址。需能存取 Routescore API 網路；每日配額依方案而定。
安裝前請注意
API 金鑰屬機密且只顯示一次，請勿放入共用設定。所有輸出皆為建模的決策支援，不構成保險、承保、投資建議或執行保證。check_swap 為唯讀，不會簽名或轉移資金，但 record_execution 會將不可變紀錄寫入帳號且屬 Pro 方案。紀錄取回與報價工具屬 Power 方案，並依金鑰限流。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Routescore，將其新增到工作區。
  2. 連接 Authorization token 憑證。它們會被私密儲存，絕不會在公開頁面上展示。
  3. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.routescore%2Fmcp/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "routescore": {
      "type": "http",
      "url": "https://www.routescore.io/api/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

@routescore/mcp

The read-only pre-sign evidence layer — a "pre-sign journal" — for onchain agents.

Before an agent (or you) signs a swap onchain, check_swap returns a modeled read of route quality, MEV/execution exposure, and token-registry recognition as a clear / caution / unsupported verdict with its caveats — and persists a hash-verifiable record of exactly what was known before signing, re-verifiable offline. That record is the pre-sign journal. It is read-only: it never signs, executes, routes funds, or custodies assets.

An MCP server (and a keyed REST API) for Claude, Codex, Cursor, and any MCP-capable agent — a thin, stateless wrapper around the Routescore public API, storing no request or response data locally.

check_swap is free. A free API key runs 100 pre-sign checks/day (Pro 1,000/day, Power 10,000/day); the modeled premium-estimate tools, simulate_scenario, and persisted-record retrieval are Power-tier. Generate a key at Account → Developer → API & MCP access (/account) — free to mint on any plan. Keys look like rs_live_… and are shown once.

See how these modeled reads have tracked measured on-chain outcomes on the public calibration surface — Routescore publishes its own accuracy (Brier score, ECE, coverage), re-derivable from a bound source manifest.

Setup

Add the server to your MCP client config and set your key in the env block.

Claude Desktop (claude_desktop_config.json) / Cursor (.cursor/mcp.json):

json
{  "mcpServers": {    "routescore": {      "command": "npx",      "args": ["-y", "@routescore/mcp"],      "env": {        "ROUTESCORE_API_KEY": "rs_live_your_key_here"      }    }  }}

Claude Code (CLI):

bash
claude mcp add routescore --env ROUTESCORE_API_KEY=rs_live_... -- npx -y @routescore/mcp

Optional env:

  • ROUTESCORE_API_URL — override the API base (default https://www.routescore.io). Useful for local development: http://localhost:3000.

The server checks the key at startup: if ROUTESCORE_API_KEY is missing or does not match the minted key shape (rs_live_ followed by 64 lowercase hex characters), it exits immediately with an actionable error instead of failing on the first tool call. The configured value is never echoed. This is a shape check only — real key verification stays server-side (run the whoami tool).

Tools

ToolWhat it does
quote_mev_coverModeled premium estimate for MEV-sandwich exposure on a swap (modeled premium, expected/CVaR loss).
quote_bridge_refundModeled premium estimate for cross-chain bridge execution failure vs a modeled SLA expectation.
quote_lrt_slashingModeled premium estimate for slashing risk on an LRT position given AVS exposure.
simulate_scenarioWhat-if Monte Carlo: modeled expected premium vs refund/loss over a horizon.
check_swapPre-trade check an agent runs before it signs: modeled route quality, price-impact / slippage band, modeled MEV/execution exposure where observable, and a token registry read (recognized vs unverified), as a clear / caution / unsupported verdict. Supports Ethereum (1) and Robinhood Chain (4663). Recognition is not safety, sellability, rights, redemption, liquidity, or investment-quality verification. Decision support, not execution. Every keyed call also attempts to persist a hash-verifiable evidence record and returns its record_id, evidence_bundle_id, and record_output_hash — null, with a record_persistence_failed caveat, if the record store is unavailable.
get_preflight_recordFetch one persisted preflight evidence record by record_id (owner-scoped to the configured key's account). The record embeds the original check response verbatim plus a canonical-JSON SHA-256 integrity hash so the evidence can be re-verified offline. Read-only evidence: record creation rejects known execution-material keys (calldata, transaction payloads, signing material) and drops identity-shaped labels from the actor context.
get_detector_manifestLatest public MEV-detector run manifest (version hash + universe).
whoamiConfirm the key works and report its plan tier.

All quote_* tool results are modeled, point-in-time premium estimates — decision support only, not a live cover, insurance, refund, or premium-acceptance offer. Routescore does not underwrite risk.

Output contract

Routescore MCP is decision support, not execution infrastructure. check_swap, quote, and scenario tool results preserve the same trust envelope as the REST API, and the wrapper marks output as degraded if the upstream API ever omits required trust fields.

check_swap answers verdict: unsupported as an HTTP 422 with a full evaluated body — an answer, not an error. The wrapper relays those evaluated 422 bodies as normal structured tool results (gap-state fields, caveats, and record linkage included) so agents receive "not evaluated" as first-class evidence; true errors (400/401/403/404/429/5xx and non-evaluated 422 error envelopes) still surface as tool errors.

Downstream agents and dashboards should render the trust-envelope fields by default (abbreviated example):

json
{  "score_state": "partial",  "source_freshness": {    "state": "partial",    "checked_at": "2026-06-21T00:00:00.000Z",    "sources": [      { "name": "routescore_backend", "freshness_state": "fresh" },      { "name": "bridge_risk_labels", "freshness_state": "unknown" }    ]  },  "methodology_version": "routescore.public_api.v1",  "confidence_band": { "low": null, "high": null, "unit": "bps" },  "caveats": [    "Modeled, point-in-time decision support. Not an execution guarantee.",    "Unsupported or stale inputs widen uncertainty instead of hiding risk."  ],  "commercial_disclosure": {    "paid_placement": false,    "score_influenced_by_partner": false  }}

Local development

bash
npm installnpm run buildROUTESCORE_API_KEY=rs_live_... ROUTESCORE_API_URL=http://localhost:3000 node dist/index.js

Then point MCP Inspector at the command, or wire it into a client config as above.

Public Reddit research utility

The retail-research workflow includes a standard-library parser for a saved, publicly rendered old Reddit thread. It extracts public handles, comment counts, IDs, and permalinks without logging in or attempting to recover deleted authors:

bash
python3 scripts/scrape_reddit_public_handles.py /path/to/thread.html \  --csv work/reddit-public-handles.csv \  --json work/reddit-public-handles.json

The input must be HTML already obtained from a public old.reddit.com thread. The parser does not fetch pages or bypass Reddit access controls.

Notes

  • Scenario and quote outputs are modeled decision-support, not investment advice.
  • MCP does not custody assets, execute transactions, route funds, or guarantee outcomes.
  • Rate limits apply per key; responses carry X-RateLimit-* headers.

來源：README.md，提交 e3d5bbf

工具

11

check_swap

Check swap before signing (pre-sign evidence)

CALL THIS BEFORE SIGNING AN ONCHAIN SWAP. Read-only pre-sign evidence check: given notional, chain, and an optional route and token, it returns a modeled route-quality grade, price-impact / slippage band, modeled MEV/execution exposure where observable, and a token registry read (recognized vs unverified), composed into a clear / caution / unsupported verdict with machine-readable reasons and caveats. Point-in-time decision support: it does not execute, route funds, sign, or promise an outcome — the agent still decides. Recognition is not safety, sellability, redemption, rights, liquidity, or investment-quality verification; unverified tokens, unknown order flow, and uncalibrated sequencer ordering downgrade the verdict rather than reading as safe. Relay the verdict, reasons, and caveats verbatim.

輸入參數routechain_idtoken_intoken_outnotional_usd必填slippage_allowance_bps

get_preflight_record

Retrieve pre-sign evidence record

Retrieve the durable evidence record a prior check_swap call persisted — the "record" leg of plan → preflight → execute → record. Fetch one record by the record_id that check_swap returned. Owner-scoped: the configured key only ever sees records belonging to its own account. The record embeds the evaluated check_swap response payload (including its evidence_bundle_id, before the additive record_id / record_output_hash linkage fields) plus a canonical-JSON SHA-256 integrity hash (record_id and recorded_at excluded) so the evidence can be re-verified offline, and includes the stored evidence bundle. Read-only evidence: record creation rejects known execution-material keys (calldata, transaction payloads, signing material) and drops identity-shaped labels from the actor context.

輸入參數record_id必填

record_execution

Record immutable execution evidence

Persist 1-100 already-built execution_record.v1 documents as immutable, personal-account post-trade evidence. Requires a stable idempotency_key; an exact replay returns the original result and reuse with different content fails closed. Pro tier, owner-scoped, and bounded by the MCP transport byte limit. This tool never accepts calldata, signing material, routes funds, executes a transaction, applies reviewer overrides, or exposes another tenant. Invalid rows are recorded only as redacted rejection reasons and are never echoed.

輸入參數records必填idempotency_key必填

review_execution

Review stored execution evidence

Review one stored execution_record.v1 by record_id. Authenticated and personal-account owner-scoped: another account’s id is indistinguishable from a missing id. Returns the deterministic review_execution.v1 evidence coverage, exceptions, unavailable-evidence register, caveats, and canonical hash. Read-only retrospective decision support only: it does not query arbitrary public transactions, accept reviewer overrides, certify best execution or compliance, give advice, or expose admin actions.

輸入參數record_id必填

review_public_transaction

Review public Ethereum transaction

DEV-ONLY CONTRACT DISCOVERY; DISABLED AGAINST PRODUCTION. On an explicitly enabled non-production server, review one arbitrary public Ethereum transaction by transaction hash. This is separate from review_execution: it does not read an owner-scoped stored execution_record.v1 and never exposes tenant, reviewer, override, or administrative fields. Returns the deterministic tenant-free public_transaction_review.v1 coverage register, conclusions, explicit gap register, source provenance, caveats, and canonical hash. Receipt, block, index, and gas facts are measured; gas cost and point-in-time confirmation depth are calculated. Venue is unavailable in this endpoint until a server-owned verified registry adapter exists. Route, amounts, protocol fee, arrival price, feasible-route benchmark, dropped status, and historical reorg remain unavailable unless the public contract can support them. Read-only retrospective decision support only; it does not certify best execution or compliance and does not give advice.

輸入參數tx_hash必填chain_id必填

whoami

Verify API key & tier

Confirm the configured API key works and report which plan tier it carries. Returns the owning account email and tier.

無需輸入參數。

get_detector_manifest

Get MEV-detector manifest

Fetch the latest public MEV-detector run manifest: the detector version hash and the eligible pool/token universe it was committed to. Use to verify what the detector covers.

無需輸入參數。

quote_mev_cover

Quote: MEV-sandwich premium (modeled)

Modeled premium estimate for MEV-sandwich exposure on a swap. Returns a modeled premium (bps + USD), expected/CVaR loss, and conditions. Point-in-time decision support before routing a trade — not a live cover, insurance, or premium-acceptance offer.

輸入參數routechain_idasset_pairnotional_usd必填refund_threshold_bpsslippage_allowance_bpshistorical_sandwich_freq

quote_bridge_refund

Quote: bridge-failure premium (modeled)

Modeled premium estimate for cross-chain bridge execution failure against a modeled SLA expectation. Returns a modeled premium and conditions — not a live cover, refund, or premium-acceptance offer.

輸入參數bridgesla_secondsnotional_usd必填dest_chain_idsource_chain_idbridge_risk_score

quote_lrt_slashing

Quote: LRT-slashing premium (modeled)

Modeled premium estimate for slashing risk on a liquid-restaking-token (LRT) position given its AVS exposure. Returns a modeled premium and slashing bounds — not a live cover offer.

輸入參數lrt_tokencovered_eth_amount必填cover_duration_weekscorrelation_max_overlapunderlying_avs_exposure

simulate_scenario

Simulate premium-vs-loss scenario (modeled)

Run a what-if Monte Carlo: model expected premium vs expected refund/loss over a horizon, given assumptions about sandwich frequency, average loss, and deductible. Returns a narrative + distribution.

輸入參數avg_loss_bpshorizon_daysnotional_usd必填deductible_bpssandwich_freq_pct

版本歷史

1
  1. v0.3.5最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分