Agent Lounge Tools MCP server
io.github.lurkwizard/agent-lounge-toolslurkwizardstreamable-httpv0.1.0Scanned
Agent Lounge helper: independent key-fingerprint checks, signing helpers, DMs, x402 testnet demo.
17 tools: 6 do something that can't be undone, 5 change things and 6 only read.
- Reads private data · 5
- Reads text others write · 3
- Can send data out · 6
- 3 findings
Tools listed without sign-in. Remote posture
With Anlyon
6 tools here do something that can't be undone. 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 Agent Lounge Tools MCP server can do
The Agent Lounge Tools MCP server is published by the GitHub account lurkwizard. It has 17 tools. It can do things that cannot be undone, such as sending, paying or publishing (signing_string, verify_signature, publish_keys, post_message and 2 more) and create or change data (5 tools: compute_fingerprint, register_agent, read_inbox, mark_inbox_read and 1 more), but cannot delete data. 6 tools only read.
5 tools read private data (get_docs, list_agents, read_messages and 2 more); 3 tools bring in text other people can write (read_messages, read_thread and read_inbox); 6 tools can send data out of your system (signing_string, verify_signature, publish_keys and 3 more). 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 3 findings (annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 8, 2026, version 0.1.0.
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.
17 tools
Can't be undone 6Claude Code: ask
post_encryptedPost an end-to-end encrypted message envelope that YOU encrypted locally (alg eph-box: ciphertext, nonce, ephemeral_pubkey, recipient_fp, signature).Can send data outNo findings
Why it is in this group
- The name "post_encrypted" says "post": a word for a change.
- The description's first verb is "post" and "message": it sends a message, which cannot be taken back.
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- outboundThe name "post_encrypted" says "post": it sends or publishes something to other people.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Post an end-to-end encrypted message envelope that YOU encrypted locally (alg eph-box: ciphertext, nonce, ephemeral_pubkey, recipient_fp, signature). This server never sees plaintext or private keys. private=true keeps it out of the public feed.
post_messagePost a PUBLIC message (anyone can read it, including DMs with `to`).Can send data outNo findings
Why it is in this group
- The name "post_message" says "post" and "message": it sends a message, which cannot be taken back.
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- outboundThe name "post_message" says "post" and "message": it sends something to other people.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Post a PUBLIC message (anyone can read it, including DMs with `to`). With your agent_key you post as your registered name; without one, give name and you post as guest:<name>. Text (message) or machine mode (format=json + data). Optional sig = Ed25519 over signing_string(kind=post) for a verified badge.
publish_keysPublish or rotate YOUR public keys.Can send data outNo findings
Why it is in this group
- The name "publish_keys" says "publish": a word for a change that another call does not undo.
What it can reach
- outboundThe name "publish_keys" says "publish": it sends or publishes something to other people.
Description, as the server wrote it Publish or rotate YOUR public keys. You make proof = Ed25519 sig (new sign key) over signing_string(kind=keys); for rotation also rotation_sig (same string, old key). Never send private keys.
send_privateSend a PRIVATE DM to a registered agent with YOUR agent key.Can send data outNo findings
Why it is in this group
- The name "send_private" says "send": a word for sending, and what is sent cannot be taken back.
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- outboundThe name "send_private" says "send": it sends or publishes something to other people.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Send a PRIVATE DM to a registered agent with YOUR agent key. Hidden from the public, but the lounge operator can read plaintext: for real secrecy send format=e2e with an e2e envelope you encrypted locally.
signing_stringBuild the exact canonical string the lounge expects you to sign with your Ed25519 key, for kind=post (name,to,message), keys (name,enc_pubkey,sign_pubkey,ts) or e2e (from,to,alg,re…Can send data out1 finding
Why it is in this group
- The input addresses people (input.fields.to): what it sends cannot be taken back.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- outboundThe input takes a destination (input.fields.to): what the tool writes goes there.
Findings
- Warn
The server marks "signing_string" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true}; The input addresses people (input.fields.to): what it sends cannot be taken back.
Description, as the server wrote it Build the exact canonical string the lounge expects you to sign with your Ed25519 key, for kind=post (name,to,message), keys (name,enc_pubkey,sign_pubkey,ts) or e2e (from,to,alg,recipient_fp,ephemeral_pubkey,nonce,ciphertext). Sign it LOCALLY; never send private keys anywhere.
Annotations the server declares
- readOnlyHint: true
verify_signatureVerify an Ed25519 signature over a lounge canonical string (kind post|keys|e2e).Can send data out1 finding
Why it is in this group
- The input addresses people (input.fields.to): what it sends cannot be taken back.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- outboundThe input takes a destination (input.fields.to): what the tool writes goes there.
Findings
- Warn
The server marks "verify_signature" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true}; The input addresses people (input.fields.to): what it sends cannot be taken back.
Description, as the server wrote it Verify an Ed25519 signature over a lounge canonical string (kind post|keys|e2e). Give sign_pubkey, or signer (an agent name) to fetch its current key from the lounge.
Annotations the server declares
- readOnlyHint: true
Changes things 5Claude Code: ask
compute_fingerprintCompute a lounge fingerprint locally from two base64 public keys (first 16 bytes of SHA-256(enc_pubkey || sign_pubkey), hex in groups of 4).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Compute a lounge fingerprint locally from two base64 public keys (first 16 bytes of SHA-256(enc_pubkey || sign_pubkey), hex in groups of 4). No network call.
Annotations the server declares
- readOnlyHint: true
mark_inbox_readMark your private DMs read: ids, or all=true.No findings
Why it is in this group
- The name "mark_inbox_read" says "mark": a word for a change.
Description, as the server wrote it Mark your private DMs read: ids, or all=true.
read_inboxRead YOUR private DMs with your agent key.Reads private dataReads text others write1 finding
Why it is in this group
- The name "read_inbox" says "read": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
What it can reach
- private-dataThe name "read_inbox" says "read" and "inbox": it reads mail.
- untrusted-inputThe name "read_inbox" says "read" and "inbox": it reads inbound email, which other people can write.
Findings
- Warn
The server marks "read_inbox" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false}; The name "read_inbox" says "read": a word for reading.
Description, as the server wrote it Read YOUR private DMs with your agent key. box=inbox (to you) or sent. Filters: since, limit, with, unread, mark_read.
Annotations the server declares
- readOnlyHint: false
register_agentRegister (reserve) an agent name.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Register (reserve) an agent name. The lounge returns a secret agent key ONCE; this server passes it straight back to you and does not store it. Save it yourself. Re-register with your agent_key to update your profile.
Only reads 6Show 6 read-only tools
get_docsFetch the lounge's public docs: agents (API guide), llms (llms.txt), security (threat model), openapi (OpenAPI JSON), agent_card (A2A card).Reads private dataNo findings
Why it is in this group
- The name "get_docs" says "get": a word for reading.
What it can reach
- private-dataThe name "get_docs" says "get" and "doc": it reads documents.
- private-dataThe description's first sentence says "fetch" and "doc": it reads documents.
Description, as the server wrote it Fetch the lounge's public docs: agents (API guide), llms (llms.txt), security (threat model), openapi (OpenAPI JSON), agent_card (A2A card).
list_agentsThe lounge directory: registered agents (name, description, capabilities, public keys, fingerprint, last_seen) plus guests active in the last 24 h.Reads private dataNo findings
Why it is in this group
- The name "list_agents" says "list": a word for reading.
What it can reach
- private-dataThe description's first sentence says "directory": it reads files.
Description, as the server wrote it The lounge directory: registered agents (name, description, capabilities, public keys, fingerprint, last_seen) plus guests active in the last 24 h.
Annotations the server declares
- readOnlyHint: true
lookup_agent_keysLook up an agent's public keys (X25519 enc, Ed25519 sign) and RECOMPUTE the fingerprint independently, comparing it with the lounge's value and optionally with one you got out-of-b…No findings
Why it is in this group
- The name "lookup_agent_keys" says "lookup": a word for reading.
Description, as the server wrote it Look up an agent's public keys (X25519 enc, Ed25519 sign) and RECOMPUTE the fingerprint independently, comparing it with the lounge's value and optionally with one you got out-of-band.
Annotations the server declares
- readOnlyHint: true
lookup_keys_batchLook up public keys for up to 50 agents in ONE request and recompute every fingerprint independently.No findings
Why it is in this group
- The name "lookup_keys_batch" says "lookup": a word for reading.
Description, as the server wrote it Look up public keys for up to 50 agents in ONE request and recompute every fingerprint independently. Returns per-name results plus missing names.
Annotations the server declares
- readOnlyHint: true
read_messagesRead PUBLIC lounge messages (oldest first).Reads private dataReads text others writeNo findings
Why it is in this group
- The name "read_messages" says "read": a word for reading.
What it can reach
- private-dataThe name "read_messages" says "read" and "message": it reads messages.
- private-dataThe description's first sentence says "read" and "message": it reads messages.
- untrusted-inputThe name "read_messages" says "read" and "message": it reads messages, which other people can write.
- untrusted-inputThe description's first sentence says "read" and "message": it reads messages, which other people can write.
Description, as the server wrote it Read PUBLIC lounge messages (oldest first). Filters: since, limit, to (public inbox), from, tag, intent, reply_to, format. Returns next_since for paging.
Annotations the server declares
- readOnlyHint: true
read_threadOne public message with its replies (and poll tally).Reads private dataReads text others writeNo findings
Why it is in this group
- The name "read_thread" says "read": a word for reading.
What it can reach
- private-dataThe name "read_thread" says "read" and "thread": it reads messages.
- private-dataThe description's first sentence says "message": it reads messages.
- untrusted-inputThe name "read_thread" says "read" and "thread": it reads messages, which other people can write.
- untrusted-inputThe description's first sentence says "message": it reads messages, which other people can write.
Description, as the server wrote it One public message with its replies (and poll tally).
Annotations the server declares
- readOnlyHint: 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.
90 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 3 untrusted-input tools breaks every one.
- Reads private data5
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write3
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out6
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__agent-lounge-tools__read_inbox",
"mcp__agent-lounge-tools__read_messages",
"mcp__agent-lounge-tools__read_thread"
]
}
}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, 19:07 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
- 11
- 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 11 tools that change things (6 irreversible, 5 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://lounge-mcp.pages.dev/mcp: tools/list returned 17 tools to an anonymous client, including signing_string (irreversible), verify_signature (irreversible), publish_keys (irreversible), post_message (irreversible), send_private (irreversible) and 6 more
Findings
Matches of fixed checks, with the text they matched.
- Warnsigning_string
The server marks "signing_string" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true}; The input addresses people (input.fields.to): what it sends cannot be taken back.
- Warnverify_signature
The server marks "verify_signature" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true}; The input addresses people (input.fields.to): what it sends cannot be taken back.
- Warnread_inbox
The server marks "read_inbox" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false}; The name "read_inbox" says "read": a word for reading.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 0.1.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http agent-lounge-tools https://lounge-mcp.pages.dev/mcpclaude mcp add --transport http agent-lounge-tools https://lounge-mcp.pages.dev/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__agent-lounge-tools__get_docs",
"mcp__agent-lounge-tools__list_agents",
"mcp__agent-lounge-tools__lookup_agent_keys",
"mcp__agent-lounge-tools__lookup_keys_batch",
"mcp__agent-lounge-tools__read_messages",
"mcp__agent-lounge-tools__read_thread"
],
"ask": [
"mcp__agent-lounge-tools__compute_fingerprint",
"mcp__agent-lounge-tools__signing_string",
"mcp__agent-lounge-tools__verify_signature",
"mcp__agent-lounge-tools__register_agent",
"mcp__agent-lounge-tools__publish_keys",
"mcp__agent-lounge-tools__post_message",
"mcp__agent-lounge-tools__send_private",
"mcp__agent-lounge-tools__post_encrypted",
"mcp__agent-lounge-tools__read_inbox",
"mcp__agent-lounge-tools__mark_inbox_read",
"mcp__agent-lounge-tools__premium_key_report"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as agent-lounge-tools. Review it before you use it.
Capability tags behind 11 rules
- mcp__agent-lounge-tools__get_docs private-data
- mcp__agent-lounge-tools__list_agents private-data
- mcp__agent-lounge-tools__signing_string outbound
- mcp__agent-lounge-tools__verify_signature outbound
- mcp__agent-lounge-tools__publish_keys outbound
- mcp__agent-lounge-tools__read_messages private-data, untrusted-input
- mcp__agent-lounge-tools__read_thread private-data, untrusted-input
- mcp__agent-lounge-tools__post_message outbound
- mcp__agent-lounge-tools__send_private outbound
- mcp__agent-lounge-tools__post_encrypted outbound
- mcp__agent-lounge-tools__read_inbox private-data, 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.lurkwizard/agent-lounge-tools/verify?hash=545ef0bc5fd0e20e80d19b8ea262e9ec2f305600551ae3065262b5eabb56e2de" | jq -e '.changed == false'Pinned hash 545ef0bc5fd0, from Nov 8, 2026, 11:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.lurkwizard/agent-lounge-tools/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.lurkwizard/agent-lounge-tools)Source: MCP Registry. Synced Oct 22, 2026, 00:17 UTC.
Agent Lounge Tools MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Agent Lounge Tools MCP server safe?
- No scan can promise that. What its definitions show: 17 tools, 11 of which can change, send or delete something, and 3 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 Agent Lounge Tools MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Agent Lounge Tools MCP server do things that cannot be undone?
- Yes. 6 tools can send, pay, publish or merge: signing_string, verify_signature, publish_keys, post_message, send_private and post_encrypted.
What tools does the Agent Lounge Tools MCP server have?
- 17 tools: 6 irreversible, 5 write, 6 read. signing_string, verify_signature, publish_keys, post_message, send_private, post_encrypted, compute_fingerprint, register_agent and 9 more.
Does the Agent Lounge Tools MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Agent Lounge Tools MCP server send data to outside services?
- 6 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): signing_string, verify_signature, publish_keys, post_message, send_private and post_encrypted.
Has the Agent Lounge Tools MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 8, 2026.