Quidli Connect MCP server
io.github.Quidli/connect-mcpQuidlistreamable-httpv0.6.25Scanned
Resolve social handles to wallets, score onchain reputation, pay in any ERC-20, SPL or native.
14 tools: 4 can delete or cancel things, 4 change things and 6 only read.
- Reads private data · 0
- Reads text others write · 14
- Can send data out · 0
- 4 findings
Tools listed without sign-in. Remote posture
With Anlyon
4 tools here can delete or cancel 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 Quidli Connect MCP server can do
The Quidli Connect MCP server is published by the GitHub account Quidli. It has 14 tools. It can delete or cancel things (4 tools: connect_drop, connect_drop_balance, connect_trust_create and connect_trust_revoke) and create or change data (4 tools: connect_scores_batch, connect_scores_by_account, connect_scores_by_username and connect_trust_check). 6 tools only read.
14 tools bring in text other people can write (connect_get_price, connect_get_chains, connect_lookup and 11 more).
The fixed checks matched 4 findings (base64-blob, long-description and 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 21, 2026, version 0.6.25.
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.
14 tools
Deletes or cancels 4Claude Code: deny
connect_dropExecute a Smart Send from the API key owner Connect embedded wallet.Reads text others write2 findings
Why it is in this group
- The name "connect_drop" says "drop": a word for removing or ending something.
- The input addresses people (input.recipients): what it sends cannot be taken back.
- The server marks it destructive (destructiveHint: true).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The description contains a 44-character base64 string.
What this check looks forMatched text
description: "...orted; USDC mint is EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v. ERC-721: set token..."
- Info
The description is 2,403 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...ted or 202 when recipients still processing — retry with the same idempotencyKey."
Description, as the server wrote it Execute a Smart Send from the API key owner Connect embedded wallet. EVM: batch native or ERC-20 (need native gas plus the token). Solana (chainId 1399811149): SOL or SPL from the Solana embedded wallet; packs up to 20 native or 10 SPL recipients per transaction. Social recipient types: email, phone, telegram, discord, farcaster, twitter, github (id or username). linkedin and slack are not on /drop — use connect_lookup first, then type wallet. After connect_lookup, EVM payouts use ethWalletAddress (never solWalletAddress); Solana payouts use solWalletAddress (never ethWalletAddress). Social types on /drop resolve server-side; for type wallet, pass the resolved payout address for the target chain. Omit tokenContract or set it to null for the native token; pass an ERC-20 contract or SPL mint otherwise — do not use the zero address. Amounts are smallest-unit integer strings (ETH 18 decimals, SOL 9, USDC usually 6). Solana native (tokenContract null): no ATA. Sending to a recipient without an existing funded account requires amount ≥ 890880 lamports (rent-exempt minimum for a system account); that SOL stays with the recipient. Below that the tx fails. Sender also pays a ~5000-lamport fee. Solana SPL: tokens sit in Associated Token Accounts (ATA), not on the wallet pubkey. Recipients need not already hold the token — the API prepends CreateIdempotent. The sender (not the recipient) pays ~2039280 lamports (~0.002039 SOL) rent per newly created dest ATA, plus tx fees, on top of the token amount (which can be as small as 1 unit). A 400 "Insufficient funds" on SPL is often missing SOL for ATA rent, not missing USDC. Token-2022 is not supported; USDC mint is EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v. ERC-721: set tokenStandard to erc721, tokenContract to the collection, and tokenId on each recipient. Do not send an amount. One safeTransferFrom per NFT. ERC-1155: set tokenStandard to erc1155, tokenContract to the contract, and tokenId plus tokenAmount (quantity, at least 1) on each recipient. The same id can go to several recipients. One safeTransferFrom per recipient. An EIP-7702 delegation signed before the matching standard must be replaced. Smart Send wallets enrolled on the previous Privy policy must re-enable Smart Send. Always call connect_drop_balance first. Returns 201 when submitted or 202 when recipients still processing — retry with the same idempotencyKey.
Annotations the server declares
- title: Send Tokens (Smart Send)
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
connect_drop_balanceGet native and token balances for the API key owner embedded wallet.Reads text others write2 findings
Why it is in this group
- The name "connect_drop_balance" says "drop": a word for removing or ending something.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- The server marks it not destructive (destructiveHint: false), 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 "connect_drop_balance" read-only (readOnlyHint: true), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "connect_drop_balance" says "drop": a word for removing or ending something.
- Warn
The server marks "connect_drop_balance" not destructive (destructiveHint: false), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "connect_drop_balance" says "drop": a word for removing or ending something.
Description, as the server wrote it Get native and token balances for the API key owner embedded wallet. Accepts Smart Send chains and enabled delegation chains. Robinhood Chain (4663) returns ERC-20 balances from Blockscout. Monad (143) returns ERC-20 balances from Moralis; other delegation chains return native plus featured tokens. Always call before connect_drop. Zero balances are omitted, so a missing token means balance 0. EVM: confirm native gas plus the ERC-20 being sent. Solana (chainId 1399811149): SOL in this response is spendable lamports on the wallet pubkey (rent locked in existing token accounts is not included). Native SOL drop to a recipient without an existing funded account: amount itself must be ≥ 890880 lamports and sender SOL must cover amount + ~5000 lamports fee. SPL drop: token balance ≥ total amount, and SOL ≥ tx fee + ~2039280 lamports (~0.002039 SOL) per recipient that may need a new Associated Token Account — even when sending USDC. Insufficient SOL for ATA rent fails before the token transfer.
Annotations the server declares
- title: Check Smart Send Balance
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_trust_createCreate unidirectional trust attestations on Base (EAS) from the API key owner embedded wallet to one or more wallets or social identities, in a single transaction.Reads text others writeNo findings
Why it is in this group
- The name "connect_trust_create" says "create": a word for a change.
- The input addresses people (input.to): what it sends cannot be taken back.
- The server marks it destructive (destructiveHint: true).
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 Create unidirectional trust attestations on Base (EAS) from the API key owner embedded wallet to one or more wallets or social identities, in a single transaction. Identical active attestations are returned without a new transaction. Recipients still provisioning are reported as processing while the others are attested. Requires CONNECT_API_KEY, Smart Send attestation signer enrolled, and ETH on Base for gas when a new attestation is submitted.
Annotations the server declares
- title: Create Trust Attestation
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
connect_trust_revokeRevoke active trust attestations from the API key owner to a target.Reads text others writeNo findings
Why it is in this group
- The name "connect_trust_revoke" says "revoke": a word for removing or ending something.
- The input addresses people (input.to): what it sends cannot be taken back.
- The server marks it destructive (destructiveHint: true).
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 Revoke active trust attestations from the API key owner to a target. Optional context limits which attestations are revoked. Requires CONNECT_API_KEY and ETH on Base.
Annotations the server declares
- title: Revoke Trust Attestation
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
Changes things 4Claude Code: ask
connect_scores_batchBatch scores for linked accounts or Connect usernames.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 Batch scores for linked accounts or Connect usernames. One result per request user. Optional filter sets passedFilter (quidli 0–100, neynar/lens 0–1, ethos 0–2800).
Annotations the server declares
- title: Score Identities (Batch)
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_scores_by_accountScores for a linked social account or wallet.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 Scores for a linked social account or wallet.
Annotations the server declares
- title: Score a Linked Account
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_scores_by_usernameScores by Connect public username.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 Scores by Connect public username.
Annotations the server declares
- title: Score a Connect Username
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_trust_checkCheck whether identities sit in a trust graph at depth 1.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 Check whether identities sit in a trust graph at depth 1. Omit context to match any context.
Annotations the server declares
- title: Check Trust
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 6Show 6 read-only tools
connect_get_chainsList product chains with per-feature compatibility.Reads text others writeNo findings
Why it is in this group
- The name "connect_get_chains" 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 List product chains with per-feature compatibility. lookup is true for EVM catalog chains and Solana; drop is true for Smart Send chains, enabled delegation chains, and Solana (chainId 1399811149 for connect_drop / connect_drop_balance).
Annotations the server declares
- title: List Supported Chains
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_get_priceGet public x402 list prices for lookup and scores (reference only; live paywall amounts are in 402 responses).Reads text others writeNo findings
Why it is in this group
- The name "connect_get_price" 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 Get public x402 list prices for lookup and scores (reference only; live paywall amounts are in 402 responses).
Annotations the server declares
- title: Get x402 Prices
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_lookupResolve social identities to EVM and Solana wallet addresses.Reads text others writeNo findings
Why it is in this group
- The name "connect_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 Resolve social identities to EVM and Solana wallet addresses. Resolving a recipient who has no wallet provisions one for them, so this is read-only for the caller but not for the recipient. If status is processing, retry the same payload.
Annotations the server declares
- title: Resolve Handle to Wallet
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_lookup_exposedList platforms a recipient has exposed on Connect, with enriched profile, scores, and wallet addresses.Reads text others writeNo findings
Why it is in this group
- The name "connect_lookup_exposed" 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 List platforms a recipient has exposed on Connect, with enriched profile, scores, and wallet addresses. Recipient may be a social account, an exposed wallet (EVM/Solana/smart wallet), or a Connect username. May require x402 payment when the profile owner charges for lookups.
Annotations the server declares
- title: List Exposed Accounts
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_meGet the Connect profile, scores, and all linked accounts for the API key owner.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "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 Get the Connect profile, scores, and all linked accounts for the API key owner. Use to identify which user the key belongs to.
Annotations the server declares
- title: Get My Connect Profile
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
connect_trust_graphList outgoing or incoming trust edges for an identity at depth 1.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "list": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it List outgoing or incoming trust edges for an identity at depth 1.
Annotations the server declares
- title: List Trust Edges
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write14
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 19:00 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
- 8
- 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 8 tools that change things (4 destructive, 4 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcp.connect.quid.li: tools/list returned 14 tools to an anonymous client, including connect_drop (destructive), connect_drop_balance (destructive), connect_trust_create (destructive), connect_trust_revoke (destructive), connect_scores_batch (write) and 3 more
Findings
Matches of fixed checks, with the text they matched.
- Warnconnect_drop
The description contains a 44-character base64 string.
What this check looks forMatched text
description: "...orted; USDC mint is EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v. ERC-721: set token..."
The server marks "connect_drop_balance" read-only (readOnlyHint: true), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "connect_drop_balance" says "drop": a word for removing or ending something.
The server marks "connect_drop_balance" not destructive (destructiveHint: false), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "connect_drop_balance" says "drop": a word for removing or ending something.
- Infoconnect_drop
The description is 2,403 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...ted or 202 when recipients still processing — retry with the same idempotencyKey."
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 changed.~1 changedView diff
- Changed:connect_dropDeletes or cancelsdescriptioninputSchema
Tools1 changed.~1 changedView diff
- Changed:connect_trust_createDeletes or cancelsdescriptioninputSchema
Install
Pinned to version 0.6.25. Values in angle brackets are yours to fill in.
claude mcp add --transport http connect https://mcp.connect.quid.liclaude mcp add --transport http connect https://mcp.connect.quid.liRun 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__connect__connect_get_price",
"mcp__connect__connect_get_chains",
"mcp__connect__connect_lookup",
"mcp__connect__connect_lookup_exposed",
"mcp__connect__connect_me",
"mcp__connect__connect_trust_graph"
],
"ask": [
"mcp__connect__connect_scores_batch",
"mcp__connect__connect_scores_by_account",
"mcp__connect__connect_scores_by_username",
"mcp__connect__connect_trust_check"
],
"deny": [
"mcp__connect__connect_drop",
"mcp__connect__connect_drop_balance",
"mcp__connect__connect_trust_create",
"mcp__connect__connect_trust_revoke"
]
}
}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 connect. Review it before you use it.
Capability tags behind 14 rules
- mcp__connect__connect_get_price untrusted-input
- mcp__connect__connect_get_chains untrusted-input
- mcp__connect__connect_lookup untrusted-input
- mcp__connect__connect_lookup_exposed untrusted-input
- mcp__connect__connect_scores_batch untrusted-input
- mcp__connect__connect_scores_by_account untrusted-input
- mcp__connect__connect_scores_by_username untrusted-input
- mcp__connect__connect_me untrusted-input
- mcp__connect__connect_drop untrusted-input
- mcp__connect__connect_drop_balance untrusted-input
- mcp__connect__connect_trust_create untrusted-input
- mcp__connect__connect_trust_revoke untrusted-input
- mcp__connect__connect_trust_check untrusted-input
- mcp__connect__connect_trust_graph 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.Quidli/connect-mcp/verify?hash=5ad5cf4494643aa4f6df18c0e027950ac6fe7ebc82490987c1955a264b970fb6" | jq -e '.changed == false'Pinned hash 5ad5cf449464, from Nov 21, 2026, 14:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.Quidli/connect-mcp/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.Quidli/connect-mcp)Source: MCP Registry · repository. Synced Nov 21, 2026, 14:18 UTC.
Quidli Connect MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Quidli Connect MCP server safe?
- No scan can promise that. What its definitions show: 14 tools, 8 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 Quidli Connect MCP server delete data?
- Yes. 4 tools can delete, cancel or revoke something: connect_drop, connect_drop_balance, connect_trust_create and connect_trust_revoke.
Can the Quidli Connect 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 Quidli Connect MCP server have?
- 14 tools: 4 destructive, 4 write, 6 read. connect_drop, connect_drop_balance, connect_trust_create, connect_trust_revoke, connect_scores_batch, connect_scores_by_account, connect_scores_by_username, connect_trust_check and 6 more.
Does the Quidli Connect MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Quidli Connect MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Quidli Connect MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 6, 2026. Every change is on this page and in its Atom feed.