CanaryGuard 402 MCP server
io.github.ak10082247-max/canaryguard-402ak10082247-maxstreamable-httpv1.0.2Scanned
Deterministic prompt-injection detection & sanitization for AI agents via x402 on Base.
26 tools: 12 change things and 14 only read.
- Reads private data · 1
- Reads text others write · 26
- Can send data out · 2
- 2 findings
Tools listed without sign-in. Remote posture
Anlyon alternative
12 of these tools change something. Run those 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 CanaryGuard 402 MCP server can do
The CanaryGuard 402 MCP server is published by the GitHub account ak10082247-max. It has 26 tools. It can create or change data (12 tools: canary_sanitize, checkWalletDrainer, guardAgentMemory, checkSmartContractHoneypot and 8 more), but cannot delete data. 14 tools only read.
1 tool reads private data (getSafeYieldAllocation); 26 tools bring in text other people can write (canary_sanitize, inspectBaseToken, checkWalletDrainer and 23 more); 2 tools can send data out of your system (inspectPhishingUrl and auditUrlAgentReadiness). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
The fixed checks matched 2 findings (annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 16, 2026, version 1.0.2.
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.
26 tools
Changes things 12Claude Code: ask
auditEip712SignatureAudits EIP-712 and EIP-3009 typed-data signature payloads (Permit, Permit2, transferWithAuthorization) to block malicious spender approvals and phishing domains.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 Audits EIP-712 and EIP-3009 typed-data signature payloads (Permit, Permit2, transferWithAuthorization) to block malicious spender approvals and phishing domains. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
auditEip712SignaturePayloadAudits EIP-712 and EIP-3009 typed-data signature payloads (Permit, Permit2, transferWithAuthorization) to prevent unauthorized token approvals and replay attacks.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 Audits EIP-712 and EIP-3009 typed-data signature payloads (Permit, Permit2, transferWithAuthorization) to prevent unauthorized token approvals and replay attacks. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
auditUrlAgentReadinessAudits any domain or URL for AI agent readiness (llms.txt, .well-known/x402, openapi.json, MCP server card) and detects hidden CSS/DOM prompt-injection traps.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 Audits any domain or URL for AI agent readiness (llms.txt, .well-known/x402, openapi.json, MCP server card) and detects hidden CSS/DOM prompt-injection traps. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
calculateLlmTokenPricingCalculates exact input, cached-input, and output token costs across frontier LLMs (OpenAI, Anthropic, Gemini, DeepSeek, Meta) and recommends optimal model routing.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 Calculates exact input, cached-input, and output token costs across frontier LLMs (OpenAI, Anthropic, Gemini, DeepSeek, Meta) and recommends optimal model routing. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
canary_sanitizeDeterministic prompt-injection firewall and honeypot scrubber for AI agents.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 Deterministic prompt-injection firewall and honeypot scrubber for AI agents. Sanitizes untrusted text and blocks detected wallet-draining commands before context ingestion. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
checkSmartContractHoneypotAudits Base smart contracts for hidden transfer locks, owner blacklist modifiers, unverified proxy upgrades, and malicious fee-on-transfer hooks.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 Audits Base smart contracts for hidden transfer locks, owner blacklist modifiers, unverified proxy upgrades, and malicious fee-on-transfer hooks. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
checkWalletDrainerAudits EVM wallet and counterparty addresses on Base for Permit2 drainer exposure, approval phishing history, and OFAC sanctions flags.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 Audits EVM wallet and counterparty addresses on Base for Permit2 drainer exposure, approval phishing history, and OFAC sanctions flags. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
checkWalletSanctionsScreens EVM and Solana wallet addresses against OFAC SDN sanctions lists, Lazarus Group attributions, and known DeFi exploit/drainer blacklists.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 Screens EVM and Solana wallet addresses against OFAC SDN sanctions lists, Lazarus Group attributions, and known DeFi exploit/drainer blacklists. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
guardAgentMemoryScans RAG context chunks, vector DB retrievals, and persistent agent memory payloads for delayed sleeper-agent instructions and cross-session poisoning.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 Scans RAG context chunks, vector DB retrievals, and persistent agent memory payloads for delayed sleeper-agent instructions and cross-session poisoning. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
repairAndSanitizeJsonDeterministically repairs malformed or truncated LLM JSON payloads (missing quotes, trailing commas, markdown fences) while scrubbing embedded prompt-injection payloads.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 Deterministically repairs malformed or truncated LLM JSON payloads (missing quotes, trailing commas, markdown fences) while scrubbing embedded prompt-injection payloads. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
runFullEnterpriseRiskAuditEnterprise 360-degree security audit bundle combining Base bytecode/proxy inspection, DEX liquidity rug-risk analysis, OFAC/Lazarus sanctions screening, EIP-712 permit verification…Reads text others write1 finding
Why it is in this group
- The name "runFullEnterpriseRiskAudit" says "run": 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 "runFullEnterpriseRiskAudit" 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,"openWorldHint":true}; The name "runFullEnterpriseRiskAudit" says "run": a word for a change.
Description, as the server wrote it Enterprise 360-degree security audit bundle combining Base bytecode/proxy inspection, DEX liquidity rug-risk analysis, OFAC/Lazarus sanctions screening, EIP-712 permit verification, and deterministic prompt-injection firewalling. Price: 0.100 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
verifyAgentAttestationGenerates a cryptographic attestation report verifying an autonomous AI agent wallet, x402 EIP-3009 payment capability, counterparty reputation score, and zero-drainer history on B…Reads text others write1 finding
Why it is in this group
- The description's first verb is "generate": 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 "verifyAgentAttestation" 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,"openWorldHint":true}; The description's first verb is "generate": a word for a change.
Description, as the server wrote it Generates a cryptographic attestation report verifying an autonomous AI agent wallet, x402 EIP-3009 payment capability, counterparty reputation score, and zero-drainer history on Base. Price: 0.020 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
Only reads 14Show 14 read-only tools
analyzeEvmWalletComprehensive EVM wallet security profile on Base, analyzing transaction velocity, approval exposure, drainer interaction risk, and smart-account state.Reads text others writeNo findings
Why it is in this group
- The name "analyzeEvmWallet" says "analyze": 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 Comprehensive EVM wallet security profile on Base, analyzing transaction velocity, approval exposure, drainer interaction risk, and smart-account state. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
auditMcpToolPoisoningInspects Model Context Protocol (MCP) tool definitions, descriptions, and JSON schemas for hidden prompt overrides, shadowing attacks, and exfiltration sinks.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "inspect": 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 Inspects Model Context Protocol (MCP) tool definitions, descriptions, and JSON schemas for hidden prompt overrides, shadowing attacks, and exfiltration sinks. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
auditMcpToolSchemaInspects Model Context Protocol (MCP) tool definitions, JSON schemas, and server cards for hidden instructions, tool shadowing, and data exfiltration vectors.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "inspect": 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 Inspects Model Context Protocol (MCP) tool definitions, JSON schemas, and server cards for hidden instructions, tool shadowing, and data exfiltration vectors. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
decodeBaseTxStatusDecodes Base Mainnet transaction receipts, EIP-3009 transferWithAuthorization settlements, ERC-20 transfer logs, gas fees, and finality confirmations.Reads text others writeNo findings
Why it is in this group
- The name "decodeBaseTxStatus" says "status": 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 Decodes Base Mainnet transaction receipts, EIP-3009 transferWithAuthorization settlements, ERC-20 transfer logs, gas fees, and finality confirmations. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getCryptoPriceOnchainReturns tamper-resistant on-chain USD spot prices, 24h volatility, and Chainlink/TWAP oracle deviation metrics for major Base and EVM assets.Reads text others writeNo findings
Why it is in this group
- The name "getCryptoPriceOnchain" 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 tamper-resistant on-chain USD spot prices, 24h volatility, and Chainlink/TWAP oracle deviation metrics for major Base and EVM assets. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getOndemandMarketSnapshotOn-demand institutional market intelligence snapshot combining Base DEX spot & TWAP prices, Aerodrome/Uniswap V3 liquidity depth, perpetual funding rates, and Polymarket whale orde…Reads text others writeNo findings
Why it is in this group
- The name "getOndemandMarketSnapshot" 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 On-demand institutional market intelligence snapshot combining Base DEX spot & TWAP prices, Aerodrome/Uniswap V3 liquidity depth, perpetual funding rates, and Polymarket whale orderbook divergence. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getSafeBaseTrendingReturns top Base Mainnet tokens by 24h DEX volume and liquidity velocity that pass CanaryGuard zero-honeypot and prompt-injection safety gates.Reads text others writeNo findings
Why it is in this group
- The name "getSafeBaseTrending" 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 top Base Mainnet tokens by 24h DEX volume and liquidity velocity that pass CanaryGuard zero-honeypot and prompt-injection safety gates. Price: 0.02 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getSafeDefiYieldsCurates top risk-adjusted USDC, EURC, and WETH DeFi yield pools on Base (Morpho, Aave V3, Aerodrome, Moonwell), pre-screened for TVL depth and audit status.Reads text others writeNo findings
Why it is in this group
- The name "getSafeDefiYields" 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 Curates top risk-adjusted USDC, EURC, and WETH DeFi yield pools on Base (Morpho, Aave V3, Aerodrome, Moonwell), pre-screened for TVL depth and audit status. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getSafeYieldAllocationComputes an institutional risk-adjusted DeFi capital allocation matrix across multi-audited Base and Arbitrum liquidity vaults (Morpho Blue, Aave V3, Moonwell, Aerodrome) with smar…Reads private dataReads text others writeNo findings
Why it is in this group
- The name "getSafeYieldAllocation" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "vault": it reads secrets.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Computes an institutional risk-adjusted DeFi capital allocation matrix across multi-audited Base and Arbitrum liquidity vaults (Morpho Blue, Aave V3, Moonwell, Aerodrome) with smart-contract timelock and depeg stress scores. Price: 0.020 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getTokenHistoryAndSpreadReturns multi-timeframe Base token price history (1h, 24h, 7d, 30d OHLCV summary), Chainlink vs.Reads text others writeNo findings
Why it is in this group
- The name "getTokenHistoryAndSpread" 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 multi-timeframe Base token price history (1h, 24h, 7d, 30d OHLCV summary), Chainlink vs. Uniswap V3/Aerodrome bid-ask spread in bps, and realized volatility skew. Price: 0.020 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
inspectBaseTokenAudits Base ERC-20 token contracts for honeypot risks, buy/sell transfer taxes, liquidity depth, and prompt injection traps in token metadata.Reads text others writeNo findings
Why it is in this group
- The name "inspectBaseToken" says "inspect": 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 Audits Base ERC-20 token contracts for honeypot risks, buy/sell transfer taxes, liquidity depth, and prompt injection traps in token metadata. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
inspectContractInfoInspects Base EVM smart contract bytecode size, EIP-1967/UUPS proxy implementation slots, EIP-7702 delegation designators, and verified function selectors.Reads text others writeNo findings
Why it is in this group
- The name "inspectContractInfo" says "inspect": 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 Inspects Base EVM smart contract bytecode size, EIP-1967/UUPS proxy implementation slots, EIP-7702 delegation designators, and verified function selectors. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
inspectPhishingUrlAudits target URLs, API endpoints, and Web3 dApp domains for IDN homograph spoofing, Angel/Pink/Inferno drainer scripts, and hidden DOM prompt injections.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "inspectPhishingUrl" says "inspect": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe name "inspectPhishingUrl" says "inspect" and "url": it reads web pages, which anyone can write.
- untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Claude Code: ask, because it is tagged outbound.
Description, as the server wrote it Audits target URLs, API endpoints, and Web3 dApp domains for IDN homograph spoofing, Angel/Pink/Inferno drainer scripts, and hidden DOM prompt injections. Price: 0.050 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
inspectUniswapPoolStateAudits Uniswap V3/V4 and Aerodrome liquidity pools on Base, returning tick liquidity depth, fee tier, price impact bps, and LP concentration rug-risk scores.Reads text others writeNo findings
Why it is in this group
- The name "inspectUniswapPoolState" says "inspect": 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 Audits Uniswap V3/V4 and Aerodrome liquidity pools on Base, returning tick liquidity depth, fee tier, price impact bps, and LP concentration rug-risk scores. Price: 0.010 USDC on Base or Solana.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- 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.
52 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write26
- 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.
{
"permissions": {
"deny": [
"mcp__canaryguard-402__getSafeYieldAllocation"
]
}
}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, 19:01 UTC, unchanged since Oct 29, 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
- 12
- Protocol version
- 2024-11-05
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://canaryguard-402.onrender.com/.well-known/oauth-protected-resource/mcp
- 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 12 tools that change things (12 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://canaryguard-402.onrender.com/mcp: tools/list returned 26 tools to an anonymous client, including canary_sanitize (write), checkWalletDrainer (write), guardAgentMemory (write), checkSmartContractHoneypot (write), auditEip712SignaturePayload (write) and 7 more
Posture changes
- Oct 29, 2026
- Tools that change things, listed without credentials: 14 to 12
Findings
Matches of fixed checks, with the text they matched.
The server marks "runFullEnterpriseRiskAudit" 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,"openWorldHint":true}; The name "runFullEnterpriseRiskAudit" says "run": a word for a change.
The server marks "verifyAgentAttestation" 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,"openWorldHint":true}; The description's first verb is "generate": a word for a change.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools8 changed.~8 changedView diff
- Changed:auditEip712SignaturePayloadChanges thingsdescription
- Changed:auditMcpToolPoisoningOnly readsdescription
- Changed:checkSmartContractHoneypotChanges thingsdescription
- Changed:checkWalletDrainerChanges thingsdescription
- Changed:getOndemandMarketSnapshotOnly readsdescription
- Changed:guardAgentMemoryChanges thingsdescription
- Changed:inspectPhishingUrlOnly readsdescription
- Changed:runFullEnterpriseRiskAuditChanges thingsdescription
Tools2 removed, 26 changed.−2 removed~26 changedView diff
- Removed:getPolymarketOddsPulse
- Removed:getPolymarketVerifiedOdds
- Changed:analyzeEvmWalletOnly readsdescription
- Changed:auditEip712SignatureChanges thingsdescriptioninputSchema
- Changed:auditEip712SignaturePayloadChanges thingsdescriptioninputSchema
- Changed:auditMcpToolPoisoningOnly readsdescriptioninputSchema
- Changed:auditMcpToolSchemaOnly readsdescriptioninputSchema
- Changed:auditUrlAgentReadinessChanges thingsdescription
- Changed:calculateLlmTokenPricingChanges thingsdescription
- Changed:canary_sanitizeChanges thingsdescription
- Changed:checkSmartContractHoneypotChanges thingsdescription
- Changed:checkWalletDrainerChanges thingsdescription
- Changed:checkWalletSanctionsChanges thingsdescription
- Changed:decodeBaseTxStatusOnly readsdescription
- Changed:getCryptoPriceOnchainOnly readsdescription
- Changed:getOndemandMarketSnapshotOnly readsdescription
- Changed:getSafeBaseTrendingOnly readsdescription
- Changed:getSafeDefiYieldsOnly readsdescription
- Changed:getSafeYieldAllocationOnly readsdescription
- Changed:getTokenHistoryAndSpreadOnly readsdescription
- Changed:guardAgentMemoryChanges thingsdescription
- Changed:inspectBaseTokenOnly readsdescription
- Changed:inspectContractInfoOnly readsdescription
- Changed:inspectPhishingUrlOnly readsdescription
- Changed:inspectUniswapPoolStateOnly readsdescription
- Changed:repairAndSanitizeJsonChanges thingsdescription
- Changed:runFullEnterpriseRiskAuditChanges thingsdescription
- Changed:verifyAgentAttestationChanges thingsdescriptioninputSchema
PostureRemote posture changed. Tools that change things, listed without credentials: 14 -> 12.View diff
- Tools that change things, listed without credentials: 14 to 12
Install
Pinned to version 1.0.2. Values in angle brackets are yours to fill in.
claude mcp add --transport http canaryguard-402 https://canaryguard-402.onrender.com/mcpclaude mcp add --transport http canaryguard-402 https://canaryguard-402.onrender.com/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__canaryguard-402__inspectBaseToken",
"mcp__canaryguard-402__auditMcpToolPoisoning",
"mcp__canaryguard-402__getSafeBaseTrending",
"mcp__canaryguard-402__getOndemandMarketSnapshot",
"mcp__canaryguard-402__getTokenHistoryAndSpread",
"mcp__canaryguard-402__getSafeYieldAllocation",
"mcp__canaryguard-402__analyzeEvmWallet",
"mcp__canaryguard-402__inspectContractInfo",
"mcp__canaryguard-402__inspectUniswapPoolState",
"mcp__canaryguard-402__getCryptoPriceOnchain",
"mcp__canaryguard-402__decodeBaseTxStatus",
"mcp__canaryguard-402__getSafeDefiYields",
"mcp__canaryguard-402__auditMcpToolSchema"
],
"ask": [
"mcp__canaryguard-402__canary_sanitize",
"mcp__canaryguard-402__checkWalletDrainer",
"mcp__canaryguard-402__guardAgentMemory",
"mcp__canaryguard-402__inspectPhishingUrl",
"mcp__canaryguard-402__checkSmartContractHoneypot",
"mcp__canaryguard-402__auditEip712SignaturePayload",
"mcp__canaryguard-402__runFullEnterpriseRiskAudit",
"mcp__canaryguard-402__verifyAgentAttestation",
"mcp__canaryguard-402__checkWalletSanctions",
"mcp__canaryguard-402__repairAndSanitizeJson",
"mcp__canaryguard-402__auditUrlAgentReadiness",
"mcp__canaryguard-402__calculateLlmTokenPricing",
"mcp__canaryguard-402__auditEip712Signature"
],
"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 canaryguard-402. Review it before you use it.
Capability tags behind 26 rules
- mcp__canaryguard-402__canary_sanitize untrusted-input
- mcp__canaryguard-402__inspectBaseToken untrusted-input
- mcp__canaryguard-402__checkWalletDrainer untrusted-input
- mcp__canaryguard-402__auditMcpToolPoisoning untrusted-input
- mcp__canaryguard-402__guardAgentMemory untrusted-input
- mcp__canaryguard-402__inspectPhishingUrl untrusted-input, outbound
- mcp__canaryguard-402__checkSmartContractHoneypot untrusted-input
- mcp__canaryguard-402__auditEip712SignaturePayload untrusted-input
- mcp__canaryguard-402__getSafeBaseTrending untrusted-input
- mcp__canaryguard-402__runFullEnterpriseRiskAudit untrusted-input
- mcp__canaryguard-402__verifyAgentAttestation untrusted-input
- mcp__canaryguard-402__getOndemandMarketSnapshot untrusted-input
- mcp__canaryguard-402__getTokenHistoryAndSpread untrusted-input
- mcp__canaryguard-402__getSafeYieldAllocation private-data, untrusted-input
- mcp__canaryguard-402__analyzeEvmWallet untrusted-input
- mcp__canaryguard-402__inspectContractInfo untrusted-input
- mcp__canaryguard-402__checkWalletSanctions untrusted-input
- mcp__canaryguard-402__inspectUniswapPoolState untrusted-input
- mcp__canaryguard-402__getCryptoPriceOnchain untrusted-input
- mcp__canaryguard-402__repairAndSanitizeJson untrusted-input
- mcp__canaryguard-402__auditUrlAgentReadiness untrusted-input, outbound
- mcp__canaryguard-402__decodeBaseTxStatus untrusted-input
- mcp__canaryguard-402__getSafeDefiYields untrusted-input
- mcp__canaryguard-402__calculateLlmTokenPricing untrusted-input
- mcp__canaryguard-402__auditEip712Signature untrusted-input
- mcp__canaryguard-402__auditMcpToolSchema 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.ak10082247-max/canaryguard-402/verify?hash=886f758bc385c29529d52eaa76271ee5df75544c0b386db4a53db38c88ba7ce1" | jq -e '.changed == false'Pinned hash 886f758bc385, from Nov 16, 2026, 09:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.ak10082247-max/canaryguard-402/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.ak10082247-max/canaryguard-402)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:22 UTC.
CanaryGuard 402 MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the CanaryGuard 402 MCP server safe?
- No scan can promise that. What its definitions show: 26 tools, 12 of which can change, send or delete something, and 2 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 CanaryGuard 402 MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the CanaryGuard 402 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 CanaryGuard 402 MCP server have?
- 26 tools: 12 write, 14 read. canary_sanitize, checkWalletDrainer, guardAgentMemory, checkSmartContractHoneypot, auditEip712SignaturePayload, runFullEnterpriseRiskAudit, verifyAgentAttestation, checkWalletSanctions and 18 more.
Does the CanaryGuard 402 MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the CanaryGuard 402 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): inspectPhishingUrl and auditUrlAgentReadiness.
Has the CanaryGuard 402 MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 16, 2026. Every change is on this page and in its Atom feed.