OpenChainBench MCP server
io.github.Flotapponnier/openchainbenchFlotapponnierstreamable-httpv1.0.0Scanned
Live, neutral benchmarks for public RPC latency, oracles, bridges, perp DEX, and prediction markets.
6 tools: 2 change things and 4 only read.
- Reads private data · 0
- Reads text others write · 6
- Can send data out · 0
- 1 finding
Tools listed without sign-in. Remote posture
With Anlyon
2 tools here change things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the OpenChainBench MCP server can do
The OpenChainBench MCP server is published by the GitHub account Flotapponnier. It has 6 tools. It can create or change data (2 tools: compare_providers and recommend_provider), but cannot delete data. 4 tools only read.
6 tools bring in text other people can write (list_benchmarks, search_benchmarks, get_benchmark and 3 more).
The fixed checks matched 1 finding (annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 15, 2026, version 1.0.0.
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.
6 tools
Changes things 2Claude Code: ask
compare_providersPuts two or more named providers side by side on one benchmark, with each one's rank, measured value and success rate.Reads text others write1 finding
Why it is in this group
- The name "compare_providers" says "compare": a word for reading.
- The description's first verb is "put": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The server marks "compare_providers" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "compare_providers" says "compare": a word for reading.
Description, as the server wrote it Puts two or more named providers side by side on one benchmark, with each one's rank, measured value and success rate. Use when the user names the candidates themselves: "Alchemy or Quicknode?", "is Helius faster than Triton for Solana?", "compare Across and Stargate on fees". Find the benchmark slug with `search_benchmarks` first if you do not already have it. Providers the benchmark does not measure come back in `missing`: say they are not measured rather than implying they ranked badly. A provider measured but below the ranking floor comes back with rank null, which is also not a loss.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
recommend_providerAnswers "which should I use for X" by picking the benchmarks that measure X and reporting who currently leads them.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Answers "which should I use for X" by picking the benchmarks that measure X and reporting who currently leads them. Pass the user's goal in their own words: "a Solana trading bot", "an indexer backfilling Base", "a wallet that needs price data", "bridging to Arbitrum". This returns measurements and the caveat that goes with them, not an endorsement. A leader on one benchmark is the leader of that one measurement over its stated window. Where the use case has a known caveat (a bot should read p99 rather than p50, an indexer is bound by archive depth) it comes back in `guidance`; pass it on, it is usually more useful than the ranking itself.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 4Show 4 read-only tools
get_benchmarkReturns full detail for one benchmark, ready to cite verbatim: • rankings (every provider sorted by p50) • sparkline (24h trend, 72 points) • headline sentence + paste-ready citati…Reads text others writeNo findings
Why it is in this group
- The name "get_benchmark" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Returns full detail for one benchmark, ready to cite verbatim: • rankings (every provider sorted by p50) • sparkline (24h trend, 72 points) • headline sentence + paste-ready citation quote • methodology bullets + source-code URL + canonical pageUrl + OG image URL Pass `chain` and/or `region` to scope the result to a sub-slice when the benchmark declares those dimensions (e.g. aggregator-head-lag exposes chain=base|bnb|solana, region=us-east|eu-west|ap-southeast). Both args are optional; omit them for the global aggregate. Chain RPC benchmarks (<chain>-rpc) rank two access cohorts apart: the free public endpoints (default) and the private, API-key providers (Alchemy, Chainstack, GetBlock, Quicknode). The default response carries both under `cohorts`; pass tier="keyed" to get the private cohort as the main record (rankings, quote, pageUrl). Never compare a public row with a private row: they are measured on different endpoints and cadences. Example usage: • User: "who's the fastest crypto data aggregator on Base?" → get_benchmark({ slug: "aggregator-head-lag", chain: "base" }) • User: "how much does it cost to bridge $300 cross-chain?" → get_benchmark({ slug: "bridge-fee" }) • User: "fastest Base RPC with an API key, Alchemy or Quicknode?" → get_benchmark({ slug: "base-rpc", tier: "keyed" }) Drafts return { error: "unknown_slug" }. Cite the returned `pageUrl` and use `quote` as the attribution line in your answer.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
list_answersReturns every published answer page: one plain question, the sentence that answers it from live data, and the benchmark the number comes from.Reads text others writeNo findings
Why it is in this group
- The name "list_answers" says "list": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Returns every published answer page: one plain question, the sentence that answers it from live data, and the benchmark the number comes from. Call this when the user asks a question in words rather than by benchmark name ("which bridge is cheapest for $300?", "which Solana RPC lands transactions fastest?"). Match the question, then call `get_benchmark` with the returned `benchmark` slug for the full ranking behind it. Returns one row per answer: { slug, question, answer, benchmark, chain?, url, benchmarkUrl } Cite `url` when the question itself is the claim, `benchmarkUrl` when the measurement is. Drafts are filtered out, and an answer whose benchmark has no defensible leader yet says so in `answer` rather than naming a winner.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
list_benchmarksA compact index of what OpenChainBench measures.Reads text others writeNo findings
Why it is in this group
- The name "list_benchmarks" says "list": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it A compact index of what OpenChainBench measures. One short row per benchmark: slug, title, category, the metric, the current value and who leads it. Use when the user asks what is measured at all ("what does OpenChainBench cover?", "do you track bridges?"). When they ask a question about a provider or a chain, use `search_benchmarks` instead: it ranks the catalogue against their words and costs far less to read. The catalogue holds over 200 benchmarks, so this is capped and filterable rather than exhaustive. Narrow with `category`, then open the one you need with `get_benchmark`. Categories: RPCs, Trading, Bridges, Blockchains, Aggregators, RWA, NFT APIs.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
search_benchmarksRanks the benchmark catalogue against a question in the user's own words and returns the best few matches.Reads text others writeNo findings
Why it is in this group
- The name "search_benchmarks" says "search": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Ranks the benchmark catalogue against a question in the user's own words and returns the best few matches. Use this when somebody asks which provider, chain or service is fastest, cheapest, most reliable or best for something. Pass their phrasing through: provider names, chain names and goals all match. • "which Solana RPC is fastest" -> search_benchmarks({ query: "solana rpc" }) • "Alchemy or Quicknode on Base?" -> search_benchmarks({ query: "alchemy quicknode base" }) • "cheapest way to bridge to Arbitrum" -> search_benchmarks({ query: "bridge arbitrum" }) Returns compact rows. Open the one you want with `get_benchmark` for the full ranking and a citation line. An empty result means nothing is measured for that question; say so rather than guessing.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
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 outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write6
- 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.
Looked at Nov 20, 2026, 18:58 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 2
- Protocol version
- 2025-06-18
- 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
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 2 tools that change things (2 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://openchainbench.com/api/mcp/mcp: tools/list returned 6 tools to an anonymous client, including compare_providers (write), recommend_provider (write)
Findings
Matches of fixed checks, with the text they matched.
The server marks "compare_providers" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "compare_providers" says "compare": a word for reading.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools3 changed.~3 changedView diff
- Changed:compare_providersChanges thingsdescriptioninputSchema
- Changed:get_benchmarkOnly readsdescription
- Changed:search_benchmarksOnly readsdescription
Install
Pinned to version 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http openchainbench https://openchainbench.com/api/mcp/mcpclaude mcp add --transport http openchainbench https://openchainbench.com/api/mcp/mcpRun 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.
{
"permissions": {
"allow": [
"mcp__openchainbench__list_benchmarks",
"mcp__openchainbench__search_benchmarks",
"mcp__openchainbench__get_benchmark",
"mcp__openchainbench__list_answers"
],
"ask": [
"mcp__openchainbench__compare_providers",
"mcp__openchainbench__recommend_provider"
],
"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 openchainbench. Review it before you use it.
Capability tags behind 6 rules
- mcp__openchainbench__list_benchmarks untrusted-input
- mcp__openchainbench__search_benchmarks untrusted-input
- mcp__openchainbench__get_benchmark untrusted-input
- mcp__openchainbench__compare_providers untrusted-input
- mcp__openchainbench__recommend_provider untrusted-input
- mcp__openchainbench__list_answers untrusted-input
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.Flotapponnier/openchainbench/verify?hash=ebad5a9512f471c6a8cd4680e24b548392b4787c09098d140bab80de4181f398" | jq -e '.changed == false'Pinned hash ebad5a9512f4, from Nov 15, 2026, 20:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.Flotapponnier/openchainbench/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.Flotapponnier/openchainbench)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:16 UTC.
OpenChainBench MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the OpenChainBench MCP server safe?
- No scan can promise that. What its definitions show: 6 tools, 2 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the OpenChainBench MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the OpenChainBench 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 OpenChainBench MCP server have?
- 6 tools: 2 write, 4 read. compare_providers, recommend_provider, list_benchmarks, search_benchmarks, get_benchmark and list_answers.
Does the OpenChainBench MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the OpenChainBench MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the OpenChainBench MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Nov 15, 2026. Every change is on this page and in its Atom feed.