apex-safety MCP server
xyz.apexfaucet/apex-safetyapexfaucet.xyzstreamable-httpv1.4.0Scanned
AI Agent Trust Stack MCP: Veritas Proof-Chain Notary, Machine Credit Bureau, Firewall & Grader.
25 tools: 1 do something that can't be undone, 22 change things and 2 only read.
- Reads private data · 0
- Reads text others write · 23
- Can send data out · 4
- 1 finding
Tools listed without sign-in. Remote posture
With Anlyon
1 tool here does something that can't be undone. Govern it.
- 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 apex-safety MCP server can do
The apex-safety MCP server is published by apexfaucet.xyz, a domain the MCP Registry verified. It has 25 tools. It can do things that cannot be undone, such as sending, paying or publishing (tx_check) and create or change data (22 tools: agent_firewall, work_grade, veritas_notary, agent_credit and 18 more), but cannot delete data. 2 tools only read.
23 tools bring in text other people can write (agent_firewall, work_grade, veritas_notary and 20 more); 4 tools can send data out of your system (agent_firewall, tx_check, delivery_score and 1 more).
The fixed checks matched 1 finding (credentials-mention). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed 3 times since Anlyon Radar first saw them.
Last scanned Nov 19, 2026, version 1.4.0.
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.
25 tools
Can't be undone 1Claude Code: ask
tx_checkFree sample: https://apexfaucet.xyz/tx-check ($0.001).Can send data outNo findings
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Free sample: https://apexfaucet.xyz/tx-check ($0.001). Check one unsigned Base or Solana transaction and return SAFE or RISKY. Base: simulation, unlimited approval, approval-for-all, a hidden token move, and one exit check. Solana: simulation, a token drain, a delegate approval, and a rug mint. Nothing is signed or broadcast. $0.001 per call over x402 (USDC on Base or USDC on Solana only). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Changes things 22Claude Code: ask
agent_card_gradeAudit and grade any AI agent domain or A2A agent card (/.well-known/agent-card.json, agent.json, mcp-server-card.json, x402, /llms.txt) with a 0-100 score, A+-F grade, and concrete…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 Audit and grade any AI agent domain or A2A agent card (/.well-known/agent-card.json, agent.json, mcp-server-card.json, x402, /llms.txt) with a 0-100 score, A+-F grade, and concrete fixes. $0.002 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Agent card grade
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
agent_creditMachine Credit Bureau (300-950 autonomous AI agent credit & trust score): fuses Veritas Proof-of-Execution history and live domain hygiene to return an Ed25519-signed AgentCreditRe…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 Machine Credit Bureau (300-950 autonomous AI agent credit & trust score): fuses Veritas Proof-of-Execution history and live domain hygiene to return an Ed25519-signed AgentCreditReport with directly pluggable agent-firewall spend limits (maxTxUsd, dailyCapUsd, humanApproveOverUsd, escrowCollateralPct). Free explorer: https://apexfaucet.xyz/veritas/ . $0.005 per call over x402 or X-API-Key.
Annotations the server declares
- title: Agent credit
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
agent_firewallUnified orchestrator access-control & pre-payment firewall for autonomous AI agents: enforces per-tx/daily USD spend caps, human-approval threshold, allowlists/denylists, HTTPS/phi…Reads text others writeCan 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
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Unified orchestrator access-control & pre-payment firewall for autonomous AI agents: enforces per-tx/daily USD spend caps, human-approval threshold, allowlists/denylists, HTTPS/phishing URL check, prompt-injection scan, and secret/PII exfiltration guard, returning an Ed25519-signed ALLOW / HOLD_FOR_HUMAN / DENY attestation verifiable offline via /.well-known/jwks.json. Free interactive tester (10/day): https://apexfaucet.xyz/firewall/ . $0.005 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1) or X-API-Key. Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Agent firewall
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
agent_launch_auditComprehensive 1-call production launch readiness audit for any AI agent, MCP server, or x402 API domain: fuses A2A/MCP/x402/JWKS/llms.txt discovery grading, HTTPS/phishing threat c…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 Comprehensive 1-call production launch readiness audit for any AI agent, MCP server, or x402 API domain: fuses A2A/MCP/x402/JWKS/llms.txt discovery grading, HTTPS/phishing threat checks, tech-stack detection, and optional live x402 delivery probe, returning prioritized revenue/discovery fixes and an Ed25519-signed LaunchReadinessCertificate verifiable offline via /.well-known/jwks.json. $0.02 per call over x402 or X-API-Key. Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Agent launch audit
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
base_token_priceExecutable Base token USD price from simulated trade, FDV, liquidity and sellability.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 Executable Base token USD price from simulated trade, FDV, liquidity and sellability. $0.003 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
Annotations the server declares
- title: Base token price
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
base_verdictSimulated Base buy+sell honeypot & rug verdict (BASED/MID/COOKED/RADIOACTIVE) with round-trip cost.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 Simulated Base buy+sell honeypot & rug verdict (BASED/MID/COOKED/RADIOACTIVE) with round-trip cost. $0.01 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
Annotations the server declares
- title: Base verdict
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
content_safetyFast text content-safety & moderation check across CSAM, self-harm, violence/weapons, malware, phishing scams, and severe harassment ($0.001).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 Fast text content-safety & moderation check across CSAM, self-harm, violence/weapons, malware, phishing scams, and severe harassment ($0.001). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Content safety
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
contract_auditFull smart contract security audit across 8 EVM chains (ethereum, base, arbitrum, optimism, polygon, bsc, arc, robinhood): bytecode disassembly, EIP-1967/1167 proxy inspection, adm…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 Full smart contract security audit across 8 EVM chains (ethereum, base, arbitrum, optimism, polygon, bsc, arc, robinhood): bytecode disassembly, EIP-1967/1167 proxy inspection, admin powers, Sourcify source static analysis, and risk verdict. $0.05 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Contract audit
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
delivery_scoreFree sample: https://apexfaucet.xyz/delivery-score?url=https://apexfaucet.xyz/api/x402/page-markdown ($0.003).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.
What it can reach
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Free sample: https://apexfaucet.xyz/delivery-score?url=https://apexfaucet.xyz/api/x402/page-markdown ($0.003). Score one https x402 URL from 0 to 100. Five parts of 0 to 20: watchtower 7-day uptime for that exact URL, a valid unpaid 402, the USDC price, how fast that probe was, and test buys we already recorded. This call does not pay the URL. A URL we have never bought scores 0 on test buys and says so. $0.003 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
html_sanitizeSanitize untrusted HTML before feeding to an LLM or UI: strips XSS scripts/iframes/on* handlers/javascript: URIs, detects & removes hidden CSS prompt-injection blocks, and returns …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 Sanitize untrusted HTML before feeding to an LLM or UI: strips XSS scripts/iframes/on* handlers/javascript: URIs, detects & removes hidden CSS prompt-injection blocks, and returns safeHtml, clean Markdown, plainText & securityReport. $0.001 per call over x402 or X-API-Key. Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Html sanitize
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
package_auditnpm and PyPI package security and supply-chain audit: checks registry existence (catches hallucinated packages), deprecation, install scripts, age, and Google OSV.dev CVEs/GHSAs wi…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 npm and PyPI package security and supply-chain audit: checks registry existence (catches hallucinated packages), deprecation, install scripts, age, and Google OSV.dev CVEs/GHSAs with SAFE/CAUTION/RISKY verdict. $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Package audit
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
pii_redactDetect and redact private keys, API tokens, JWTs, emails, IBANs, SSNs, phones, IPs, and Luhn-validated payment cards before LLM logging or sharing ($0.001).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 Detect and redact private keys, API tokens, JWTs, emails, IBANs, SSNs, phones, IPs, and Luhn-validated payment cards before LLM logging or sharing ($0.001). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Pii redact
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
prompt_guardScan untrusted text for prompt injection, jailbreak attempts, delimiter spoofing, wallet exfiltration, and encoded payloads ($0.001).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 Scan untrusted text for prompt injection, jailbreak attempts, delimiter spoofing, wallet exfiltration, and encoded payloads ($0.001). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Prompt guard
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
robinhood_exit_checkReal simulated buy-and-sell honeypot and exit check on Robinhood Chain (chain 4663) across up to 4 Uniswap v4 pools for a token address.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 Real simulated buy-and-sell honeypot and exit check on Robinhood Chain (chain 4663) across up to 4 Uniswap v4 pools for a token address. $0.01 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
Annotations the server declares
- title: Robinhood exit check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
solana_token_reportSolana token rug & safety report: mint/freeze authorities, top holders, Jupiter exit ladder and verdict.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 Solana token rug & safety report: mint/freeze authorities, top holders, Jupiter exit ladder and verdict. $0.01 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
Annotations the server declares
- title: Solana token report
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
threat_intel_bulkFull X1 machine-readable scam blacklist and related-wallet cluster graph.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 Full X1 machine-readable scam blacklist and related-wallet cluster graph. $0.025 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Threat intel bulk
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
token_exit_matrixMulti-size DEX token liquidity, sell slippage, AMM price impact curve, and honeypot/exit risk matrix across USD position tiers before trading.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 Multi-size DEX token liquidity, sell slippage, AMM price impact curve, and honeypot/exit risk matrix across USD position tiers before trading. Not advice. $0.004 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Token exit matrix
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
token_safety_checkIs this token safe to buy?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 Is this token safe to buy? Give a token contract address and a chain (base, arc, robinhood, solana or x1) and get a verdict: BASED, MID, COOKED or RADIOACTIVE, from a real simulated buy and full sell (Base, Robinhood Chain) or that chain's measured exit (Arc, Solana, X1), with the round-trip cost and a plain one-liner. Free, read-only, no account. It measures the exit right now; it is not advice and not a guarantee, and a read we could not make comes back as an error, never as clean.
Annotations the server declares
- title: Token safety check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
url_threat_checkCheck any URL for phishing, brand impersonation, Punycode/IDN homographs, credential tricks, executable downloads, DNS resolution, and redirect chain risk (SAFE/CAUTION/RISKY).Reads text others writeCan send data out1 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.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Findings
- Info
The description mentions "credential".
What this check looks forMatched text
description: "...impersonation, Punycode/IDN homographs, credential tricks, executable downloads, DNS resol..."
Description, as the server wrote it Check any URL for phishing, brand impersonation, Punycode/IDN homographs, credential tricks, executable downloads, DNS resolution, and redirect chain risk (SAFE/CAUTION/RISKY). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Url threat check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
veritas_notaryHash-chained Proof-of-Execution Notary for autonomous AI agents: links any workflow step to an optional parentReceiptHash, runs deterministic security/integrity checks (prompt inje…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 Hash-chained Proof-of-Execution Notary for autonomous AI agents: links any workflow step to an optional parentReceiptHash, runs deterministic security/integrity checks (prompt injection, secret/PII leak, JSON validity), returns an Ed25519-signed VeritasReceipt with a free public verification URL (/veritas/?r=<hash>), and compounds the agent Machine Credit Score. Free 10/day demo: https://apexfaucet.xyz/veritas/ . $0.01 per call over x402 or X-API-Key.
Annotations the server declares
- title: Veritas notary
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
wallet_investigateMultichain counterparty & wallet due diligence for EVM (0x...) or Solana base58 addresses: OFAC SDN list, on-chain USDC/USDT freeze status, native & USDC balances plus nonces acros…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 Multichain counterparty & wallet due diligence for EVM (0x...) or Solana base58 addresses: OFAC SDN list, on-chain USDC/USDT freeze status, native & USDC balances plus nonces across Ethereum, Base, Arbitrum, Optimism, Polygon & Arc, contract vs EOA detection, ERC-8004 reputation, and Arc Testnet ApexServiceEscrow stats. $0.05 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Wallet investigate
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
work_gradeGrade any AI agent work deliverable before releasing escrow (ERC-8183 / G2 Jobs / MCP): checks JSON/schema validity, anti-refusal/stub quality, embedded indirect prompt injection, …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 Grade any AI agent work deliverable before releasing escrow (ERC-8183 / G2 Jobs / MCP): checks JSON/schema validity, anti-refusal/stub quality, embedded indirect prompt injection, secret/PII leakage, and task alignment, returning an Ed25519-signed PASS_RELEASE / REVISE / REJECT_REFUND attestation verifiable offline via /.well-known/jwks.json. Free interactive tester (10/day): https://apexfaucet.xyz/grade/ . $0.004 per call over x402 or X-API-Key. Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Work grade
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 2Show 2 read-only tools
cve_lookupCVE vulnerability lookup by CVE ID or keyword combining official NVD 2.0, OSV.dev affected package versions, and CISA Known Exploited Vulnerabilities (KEV).Reads text others writeNo findings
Why it is in this group
- The name "cve_lookup" says "lookup": 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 CVE vulnerability lookup by CVE ID or keyword combining official NVD 2.0, OSV.dev affected package versions, and CISA Known Exploited Vulnerabilities (KEV). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, then call again with `payment`.
Annotations the server declares
- title: Cve lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
evm_log_decodeDecode all EVM transaction event logs (ERC-20/721/1155, Uniswap V2/V3 swaps, approvals, CCTP, EIP-3009) + net asset flows, OR preflight-simulate unsigned calldata across 8 EVM chai…Reads text others writeNo findings
Why it is in this group
- The name "evm_log_decode" says "log": 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 Decode all EVM transaction event logs (ERC-20/721/1155, Uniswap V2/V3 swaps, approvals, CCTP, EIP-3009) + net asset flows, OR preflight-simulate unsigned calldata across 8 EVM chains (ethereum, base, arbitrum, optimism, polygon, bsc, arc, robinhood). $0.004 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
Annotations the server declares
- title: Evm log decode
- 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.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write23
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out4
- 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 19, 2026, 17:24 UTC, unchanged since Nov 19, 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
- 23
- 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
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 23 tools that change things (1 irreversible, 22 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://apexfaucet.xyz/api/mcp/safety: tools/list returned 25 tools to an anonymous client, including tx_check (irreversible), agent_firewall (write), work_grade (write), veritas_notary (write), agent_credit (write) and 18 more
Findings
Matches of fixed checks, with the text they matched.
- Infourl_threat_check
The description mentions "credential".
What this check looks forMatched text
description: "...impersonation, Punycode/IDN homographs, credential tricks, executable downloads, DNS resol..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools5 added.+5 addedView diff
- Added:agent_creditChanges things
- Added:agent_launch_auditChanges things
- Added:html_sanitizeChanges things
- Added:veritas_notaryChanges things
- Added:work_gradeChanges things
Tools3 added.+3 addedView diff
- Added:agent_firewallChanges things
- Added:package_auditChanges things
- Added:url_threat_checkChanges things
Tools2 added.+2 addedView diff
- Added:agent_card_gradeChanges things
- Added:cve_lookupOnly reads
Install
Pinned to version 1.4.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http apex-safety https://apexfaucet.xyz/api/mcp/safetyclaude mcp add --transport http apex-safety https://apexfaucet.xyz/api/mcp/safetyRun 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__apex-safety__evm_log_decode",
"mcp__apex-safety__cve_lookup"
],
"ask": [
"mcp__apex-safety__agent_firewall",
"mcp__apex-safety__work_grade",
"mcp__apex-safety__veritas_notary",
"mcp__apex-safety__agent_credit",
"mcp__apex-safety__agent_launch_audit",
"mcp__apex-safety__html_sanitize",
"mcp__apex-safety__token_safety_check",
"mcp__apex-safety__tx_check",
"mcp__apex-safety__delivery_score",
"mcp__apex-safety__agent_card_grade",
"mcp__apex-safety__url_threat_check",
"mcp__apex-safety__package_audit",
"mcp__apex-safety__contract_audit",
"mcp__apex-safety__wallet_investigate",
"mcp__apex-safety__base_verdict",
"mcp__apex-safety__base_token_price",
"mcp__apex-safety__solana_token_report",
"mcp__apex-safety__threat_intel_bulk",
"mcp__apex-safety__token_exit_matrix",
"mcp__apex-safety__robinhood_exit_check",
"mcp__apex-safety__prompt_guard",
"mcp__apex-safety__pii_redact",
"mcp__apex-safety__content_safety"
],
"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 apex-safety. Review it before you use it.
Capability tags behind 25 rules
- mcp__apex-safety__agent_firewall untrusted-input, outbound
- mcp__apex-safety__work_grade untrusted-input
- mcp__apex-safety__veritas_notary untrusted-input
- mcp__apex-safety__agent_credit untrusted-input
- mcp__apex-safety__agent_launch_audit untrusted-input
- mcp__apex-safety__html_sanitize untrusted-input
- mcp__apex-safety__token_safety_check untrusted-input
- mcp__apex-safety__tx_check outbound
- mcp__apex-safety__delivery_score outbound
- mcp__apex-safety__agent_card_grade untrusted-input
- mcp__apex-safety__url_threat_check untrusted-input, outbound
- mcp__apex-safety__package_audit untrusted-input
- mcp__apex-safety__contract_audit untrusted-input
- mcp__apex-safety__evm_log_decode untrusted-input
- mcp__apex-safety__wallet_investigate untrusted-input
- mcp__apex-safety__base_verdict untrusted-input
- mcp__apex-safety__base_token_price untrusted-input
- mcp__apex-safety__solana_token_report untrusted-input
- mcp__apex-safety__threat_intel_bulk untrusted-input
- mcp__apex-safety__token_exit_matrix untrusted-input
- mcp__apex-safety__robinhood_exit_check untrusted-input
- mcp__apex-safety__prompt_guard untrusted-input
- mcp__apex-safety__pii_redact untrusted-input
- mcp__apex-safety__content_safety untrusted-input
- mcp__apex-safety__cve_lookup 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/xyz.apexfaucet/apex-safety/verify?hash=1cdc8d50e3354fd1738308c8d092a856460582f6808393d0976157921524c990" | jq -e '.changed == false'Pinned hash 1cdc8d50e335, from Nov 19, 2026, 12:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/xyz.apexfaucet/apex-safety/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/xyz.apexfaucet/apex-safety)Source: MCP Registry. Synced Nov 19, 2026, 12:18 UTC.
apex-safety MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the apex-safety MCP server safe?
- No scan can promise that. What its definitions show: 25 tools, 23 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 apex-safety MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the apex-safety MCP server do things that cannot be undone?
- Yes. 1 tool can send, pay, publish or merge: tx_check.
What tools does the apex-safety MCP server have?
- 25 tools: 1 irreversible, 22 write, 2 read. tx_check, agent_firewall, work_grade, veritas_notary, agent_credit, agent_launch_audit, html_sanitize, token_safety_check and 17 more.
Does the apex-safety MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the apex-safety MCP server send data to outside services?
- 4 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): agent_firewall, tx_check, delivery_score and url_threat_check.
Has the apex-safety MCP server changed its tools?
- Yes, 3 times since Anlyon Radar started tracking it, most recently on Nov 19, 2026. Every change is on this page and in its Atom feed.