MCPFax Reliability MCP server
com.mcpscores/agent-reliability-scoresmcpscores.comstreamable-httpv1.1.5Scanned
Before you pay an endpoint you cannot vouch for: credit grade, drift and preflight.
7 tools: 7 change things.
- Reads private data · 0
- Reads text others write · 2
- Can send data out · 2
- 4 findings
Tools listed without sign-in. Remote posture
With Anlyon
7 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 MCPFax Reliability MCP server can do
The MCPFax Reliability MCP server is published by mcpscores.com, a domain the MCP Registry verified. It has 7 tools. It can create or change data (7 tools: mcpfax_endpoint_credit_score, mcpfax_integration_drift, mcpfax_wallet_authenticity, mcpfax_endpoint_revenue_estimate and 3 more), but cannot delete data. None of its tools only reads.
2 tools bring in text other people can write (mcpfax_payer_wallet_profile and mcpfax_preflight_check); 2 tools can send data out of your system (mcpfax_endpoint_credit_score and mcpfax_endpoint_revenue_estimate).
The fixed checks matched 4 findings (base64-blob). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 13, 2026, version 1.1.5.
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
Changes things 7Claude Code: ask
mcpfax_endpoint_credit_scoreCreditworthiness grade for one x402 endpoint or MCP server: letter grade A-E (or NR when unrated), a 0-1000 score, per-component subscores, a confidence level, flags, and the dated…Can send data outNo 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
- outboundThe input takes a URL (input.endpoint): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Creditworthiness grade for one x402 endpoint or MCP server: letter grade A-E (or NR when unrated), a 0-1000 score, per-component subscores, a confidence level, flags, and the dated evidence each figure rests on. Use this before trusting or paying an unfamiliar endpoint. The grade ranks observed risk; it is not a calibrated probability of failure, not validated over any forward horizon, and not a security audit. Coverage is young: the response carries archive_coverage (measured archive span), most listings grade NR, and payer-concentration risk is reported as "not rated" until a listing's own payment record clears the published coverage floor. Costs $0.008 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
mcpfax_endpoint_revenue_estimateObserved on-chain revenue signals for one listing: total USDC inflow, how many distinct payers, the modal transfer size, payment rate, 7- and 30-day trend windows, and where it sit…Can send data outNo 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
- outboundThe input takes a URL (input.endpoint): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Observed on-chain revenue signals for one listing: total USDC inflow, how many distinct payers, the modal transfer size, payment rate, 7- and 30-day trend windows, and where it sits in its cohort by percentile. Use this to size a seller's real traction. Inflow is observed on-chain and is not proven to be x402 sales revenue. Trend windows are only as wide as the wallet's real scanned span, which the response reports as trends.scanned_days with per-window backed-by-scan flags; a 30-day window is not yet backed by 30 days of scanning. Costs $0.005 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
mcpfax_integration_driftWhat changed under an x402 endpoint or MCP server you have ALREADY integrated: payTo rotations, price, asset and network changes, terms added or removed, and tool-schema changes.No 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.
Description, as the server wrote it What changed under an x402 endpoint or MCP server you have ALREADY integrated: payTo rotations, price, asset and network changes, terms added or removed, and tool-schema changes. This is not a discovery tool and will not help you choose a server; it tells you whether the one you wired in still matches what you wired in. Pass up to 50 subjects and the `since` watermark returned by your last call, and you receive only what changed after it plus a fresh watermark, so repeat checks stay cheap. A subject with no changes is reported as observed-unchanged with the date that is true as of — that is a real answer and is billed. A subject MCPFax has never observed is named in `not_observed` and given NO verdict, because reporting "no changes" about something we never watched is indistinguishable from a genuine all-clear. The watermark is the last date MCPFax actually observed, never the current clock, so no window is skipped between sweeps. Coverage is young and every response carries its archive span. Costs $0.008 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
mcpfax_payer_wallet_profileKnow-your-agent profile for a paying wallet: how many distinct services it pays, total observed spend, whether the address is a contract or an externally owned account, and coverag…Reads text others write1 finding
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.
Findings
- Warn
The input schema at properties.address.examples[0] contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.address.examples[0]: "0x9a07CeA7ed111c93a233d1d12BbB68a511B59Eea"
Description, as the server wrote it Know-your-agent profile for a paying wallet: how many distinct services it pays, total observed spend, whether the address is a contract or an externally owned account, and coverage-gated screening signals. Use this to vet a counterparty that is paying you. Screening signals are withheld with status "not_rated" until the observed record for the address clears the published coverage floor (14 days, 20 payments); the measured graph facts are always returned and carry no inference about intent. Costs $0.008 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
mcpfax_payto_change_feedChronological feed of rating actions since a date: grade upgrades and downgrades, payTo wallet changes (a seller redirecting payment), listing removals, and returns.No 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.
Description, as the server wrote it Chronological feed of rating actions since a date: grade upgrades and downgrades, payTo wallet changes (a seller redirecting payment), listing removals, and returns. Use this to monitor a portfolio of endpoints for adverse changes rather than re-scoring each one. The feed cannot return events from before the observatory archive started; the response reports coverage.archive_since, coverage.rating_actions_recorded, and whether the requested window predates the archive. Costs $0.005 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
mcpfax_preflight_checkCheck that an endpoint is live and that its terms have not changed, BEFORE you send it an expensive request.Reads text others write1 finding
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.
Findings
- Warn
The input schema at properties.expected_payto.examples[0] contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.expected_payto.examples[0]: "0x1831f336585a6C67B6A954d28f3E07F44C4EEBbd"
Description, as the server wrote it Check that an endpoint is live and that its terms have not changed, BEFORE you send it an expensive request. A $15 call with a hypothetical 2% loss rate carries $0.30 of expected loss; a reachability check cannot promise to prevent it. Compare the quoted fee with the value of the pending call and the failures this check can actually detect. For a cheap $0.001 lookup, the $0.008 fee is eight times the call price. Returns two halves. LIVE: one unpaid bounded probe (~2.2s cap) giving reachability, HTTP status, latency, whether an MCP server still answers initialize, and the price and payTo it is serving right now. HISTORICAL: reliability over our OBSERVED window (stated in days, never implied), payTo changes, price changes, tool-schema changes, first-seen date, and credit grade or an honest not-rated. Supply expected_price, expected_payto or expected_schema_hash and each is compared with old to new and the date we first observed the change. The verdict is ALLOW, WARN or UNKNOWN and is ADVISORY: it reports evidence, it is not a warranty, an audit, or a promise your call will succeed. UNKNOWN is a real answer and is returned whenever the evidence does not support the other two. If we have never observed the target AND cannot reach it, you get UNKNOWN and are charged nothing. Otherwise $0.008 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
mcpfax_wallet_authenticityPayer-concentration and wash-risk assessment for a receiving wallet: how concentrated its observed payers are, the measured coverage of that observation, an explicit wash-risk hair…2 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.
Findings
- Warn
The input schema at properties.wallet.description contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.wallet.description: "...x characters, e.g. '0x1831f336585a6C67B6A954d28f3E07F44C4EEBbd'."
- Warn
The input schema at properties.wallet.examples[0] contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.wallet.examples[0]: "0x1831f336585a6C67B6A954d28f3E07F44C4EEBbd"
Description, as the server wrote it Payer-concentration and wash-risk assessment for a receiving wallet: how concentrated its observed payers are, the measured coverage of that observation, an explicit wash-risk haircut ONLY when coverage is sufficient, how many listings declare the same address, and observed USDC inflow. Use this to judge whether a seller's apparent revenue reflects real independent demand. The haircut and the adjusted-inflow figure are withheld as null with status "not_rated" until the wallet's own record covers at least 14 days and 20 payments, because below that a thin payer count is a property of our observation window, not of the wallet. Concentration is not proof of common control. Costs $0.01 USDC per call via x402 on Base.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
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.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write2
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out2
- 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, 18:42 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
- 7
- 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 7 tools that change things (7 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcpfax-reliability.bowling-anthony.workers.dev/mcp: tools/list returned 7 tools to an anonymous client, including mcpfax_endpoint_credit_score (write), mcpfax_integration_drift (write), mcpfax_wallet_authenticity (write), mcpfax_endpoint_revenue_estimate (write), mcpfax_payer_wallet_profile (write) and 2 more
Findings
Matches of fixed checks, with the text they matched.
The input schema at properties.wallet.description contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.wallet.description: "...x characters, e.g. '0x1831f336585a6C67B6A954d28f3E07F44C4EEBbd'."
The input schema at properties.wallet.examples[0] contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.wallet.examples[0]: "0x1831f336585a6C67B6A954d28f3E07F44C4EEBbd"
The input schema at properties.address.examples[0] contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.address.examples[0]: "0x9a07CeA7ed111c93a233d1d12BbB68a511B59Eea"
The input schema at properties.expected_payto.examples[0] contains a 42-character base64 string.
What this check looks forMatched text
inputSchema.properties.expected_payto.examples[0]: "0x1831f336585a6C67B6A954d28f3E07F44C4EEBbd"
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.1.5. Values in angle brackets are yours to fill in.
claude mcp add --transport http agent-reliability-scores https://mcpfax-reliability.bowling-anthony.workers.dev/mcpclaude mcp add --transport http agent-reliability-scores https://mcpfax-reliability.bowling-anthony.workers.dev/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": [],
"ask": [
"mcp__agent-reliability-scores__mcpfax_endpoint_credit_score",
"mcp__agent-reliability-scores__mcpfax_integration_drift",
"mcp__agent-reliability-scores__mcpfax_wallet_authenticity",
"mcp__agent-reliability-scores__mcpfax_endpoint_revenue_estimate",
"mcp__agent-reliability-scores__mcpfax_payer_wallet_profile",
"mcp__agent-reliability-scores__mcpfax_payto_change_feed",
"mcp__agent-reliability-scores__mcpfax_preflight_check"
],
"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 agent-reliability-scores. Review it before you use it.
Capability tags behind 4 rules
- mcp__agent-reliability-scores__mcpfax_endpoint_credit_score outbound
- mcp__agent-reliability-scores__mcpfax_endpoint_revenue_estimate outbound
- mcp__agent-reliability-scores__mcpfax_payer_wallet_profile untrusted-input
- mcp__agent-reliability-scores__mcpfax_preflight_check 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/com.mcpscores/agent-reliability-scores/verify?hash=3dae746c49ed605a5b9529854026c44b120506f88b0d747fcab85cda6984604c" | jq -e '.changed == false'Pinned hash 3dae746c49ed, from Nov 13, 2026, 18:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.mcpscores/agent-reliability-scores/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.mcpscores/agent-reliability-scores)Source: MCP Registry. Synced Oct 21, 2026, 17:59 UTC.
MCPFax Reliability MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the MCPFax Reliability MCP server safe?
- No scan can promise that. What its definitions show: 7 tools, 7 of which can change, send or delete something, and 4 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the MCPFax Reliability MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the MCPFax Reliability 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 MCPFax Reliability MCP server have?
- 7 tools: 7 write. mcpfax_endpoint_credit_score, mcpfax_integration_drift, mcpfax_wallet_authenticity, mcpfax_endpoint_revenue_estimate, mcpfax_payer_wallet_profile, mcpfax_payto_change_feed and mcpfax_preflight_check.
Does the MCPFax Reliability MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the MCPFax Reliability MCP server send data to outside services?
- 2 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): mcpfax_endpoint_credit_score and mcpfax_endpoint_revenue_estimate.
Has the MCPFax Reliability MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 13, 2026.