Context7
Up-to-date docs for AI agents
Beschreibung vom Entwickler (englisches Original)
AI-optimized MCP server for iMessage on macOS. Read, search, and send messages.
Diese drei werden nie zu einer Punktzahl zusammengefasst: Meinungen, Messungen und Beliebtheit sind verschiedene Dinge.
Für diesen Server haben wir noch keine Zusammenfassung in einfacher Sprache geschrieben. Unten stehen die Fakten, die wir aus der Registry gelesen haben.
Wähle die KI-App, die du nutzt, und kopiere die Einstellungen.
Du brauchst
Welche App nutzt du?
Wähle oben deine App. Wir merken sie uns, damit der Button beim nächsten Mal „In … installieren“ anzeigt.
Öffne Claude Desktop → Settings → Developer → Edit Config und füge dies in claude_desktop_config.json ein
{
"mcpServers": {
"imessage-max": {
"command": "uvx",
"args": [
"imessage-max"
]
}
}
}Führe das in deinem Terminal aus
claude mcp add imessage-max -- uvx imessage-maxIn .cursor/mcp.json in deinem Projekt oder in ~/.cursor/mcp.json einfügen
{
"mcpServers": {
"imessage-max": {
"command": "uvx",
"args": [
"imessage-max"
]
}
}
}Dein Browser fragt, ob er die App öffnen darf, dann zeigt die App die Einstellungen und bittet dich um Bestätigung. Schlüsselwerte bleiben leer, damit du sie selbst einträgst.
In .vscode/mcp.json in deinem Projekt einfügen
{
"servers": {
"imessage-max": {
"type": "stdio",
"command": "uvx",
"args": [
"imessage-max"
]
}
}
}Dein Browser fragt, ob er die App öffnen darf, dann zeigt die App die Einstellungen und bittet dich um Bestätigung. Schlüsselwerte bleiben leer, damit du sie selbst einträgst.
Dieser Server läuft auf deinem Computer, und ChatGPT kann sich nur mit Servern im Internet verbinden. Wähle eine andere App.
Deine Auswahl bleibt in einem Cookie auf diesem Gerät; kein Konto nötig.
Diese Einstellungen werden aus dem Registry-Eintrag des Entwicklers erzeugt. Wenn in der README etwas anderes steht, folge der README.
Läuft auf deinem Computer; die Tool-Liste erscheint nach der Installation in deiner KI-App.
Wir prüfen Informationen, die jeder sehen kann; den Code haben wir nicht gelesen. Checks, die nicht zutreffen, werden nicht mitgezählt. · Geprüft am 7. Nov. 2026
Der Registry-Name wurde mit dem GitHub-Konto cyberpapiii bestätigt, dem das Repo gehört
Dieses Repo haben wir noch nicht von GitHub gelesen
Dieses Repo haben wir noch nicht von GitHub gelesen
Dieser Server läuft auf deinem Computer; es gibt keinen Endpunkt im Internet
Läuft auf deinem Computer und nutzt stattdessen lokale Einstellungen
Läuft auf deinem Computer; Tool-Listen können wir nur von Endpunkten im Internet lesen
Verglichen mit allen aktiven Servern dieser Kategorie in der Registry. Sterne zählen nur, wenn dem Herausgeber das Repo gehört.
Er hat 1 von 1 zutreffenden Basischecks bestanden. Den Code haben wir nicht gelesen. Bevor du einen Schlüssel oder Token hinzufügst, sieh nach, welche Zugriffe der Server verlangt, und nutze einen Schlüssel, der nur die Berechtigungen hat, die du brauchst.
Die Registry nennt keinen erforderlichen Schlüssel, manche Server bitten dich aber bei der ersten Nutzung, dich anzumelden.
Er läuft auf deinem Computer. Deine KI-App startet ihn bei Bedarf, und die Daten laufen nicht über die Server des Entwicklers (außer der Server ruft selbst einen Onlinedienst auf).
Das GitHub-Repo wurde zuletzt – geändert, und die neueste Version (0.4.4) kam vor 10 Monaten in die Registry.
Apps, die MCP unterstützen, etwa Claude Desktop, Claude Code, Cursor und VS Code. Außerdem brauchst du Python (uv) auf deinem Computer.
In derselben Kategorie: Context7, Codebase Memory, Exa
Server derselben Kategorie von anderen Entwicklern, sortiert nach Sternen.
So liest du eine Karte:★ GitHub-SterneiDrei Säulen: Bewertungen · Checks · NutzungiAktualisiert = neueste Version in der Registry veröffentlichti
Up-to-date docs for AI agents
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
Unverändert aus dem Repo des Entwicklers übernommen. PopNx MCP hat den Inhalt weder geprüft noch empfiehlt es ihn. HTML, Bilder und Skripte wurden entfernt. Links führen von dieser Website weg.
Want a simple Mac control panel? The native menu bar companion checks setup, manages the local service, and shares assistant connection details.
An MCP (Model Context Protocol) server for iMessage. It lets AI agents read, search, and send your messages, resolving phone numbers to contact names along the way.
Written in Swift. One binary, no runtime dependencies.
One Swift implementation, under swift/, shipped as:
cyberpapiii/tap).codex-plugin/)mcpb/)Most iMessage tools expose raw database structures, requiring 3-5 tool calls per user intent. This MCP provides intent-aligned tools:
"What did Contact A and I talk about yesterday?"
→ find_chat(participants=["Contact A"]) + get_messages(since="yesterday")
"Show me the exact details for this thread before I reply" → get_chat_details(chat_id="chat123")
"Show me photos from the group chat" → list_attachments(chat_id="chat123", type="image")
"Find where we discussed the launch timeline" → search(query="launch timeline")
The tools work best when an agent uses them as short workflows instead of isolated one-off calls.
Agents should treat chat_id values like chat123 as internal handles for tool calls and exact sends. When explaining results to a person, use the returned chat name, group name, or participant-derived label instead of saying "Chat 123."
find_chat(participants=["Contact A"])
get_chat_details(chat_id="chat123") get_messages(chat_id="chat123", since="yesterday", limit=50)
Use this when the person matters more than the exact thread id.
search(query="launch timeline", limit=10)
get_context(message_id="msg_456", before=5, after=10)
Use this when you know the topic but not where it was discussed.
get_unread()
get_active_conversations(hours=24, min_exchanges=2)
Use this to surface unread threads and active conversations after a broad chat-list sweep.
list_attachments(chat_id="chat123", type="image", since="30d")
get_attachment(attachment_id="att123", variant="vision")
Use list_attachments to find the message where files were shared. It returns exact attachment ids and says whether each file is on disk, so you know before you fetch.
find_chat(participants=["Contact A", "Contact B"])
send(chat_id="chat456", text="Please use the latest draft")
For sensitive sends, prefer resolving the exact chat first and then using chat_id so the message lands in the intended thread.
brew tap cyberpapiii/tap
brew install imessage-max
git clone https://github.com/cyberpapiii/imessage-max.git
cd imessage-max/swift make build
swift build omits the embedded Info.plist, so Contacts can't be granted)
For local development, advanced setup, and the signed install workflow, see:
iMessage Max is a dual-era MCP server. Both transports (stdio and HTTP) serve both eras concurrently, selected per request:
| Era | Revisions | Lifecycle | Selected by |
|---|---|---|---|
| Modern | 2026-07-28 | Stateless, per-request _meta | io.modelcontextprotocol/protocolVersion in the request _meta (or server/discover) |
| Legacy | 2025-03-26 … 2025-11-25 | initialize + session | initialize request / Mcp-Session-Id |
Modern clients probe with server/discover and send the required MCP-Protocol-Version, Mcp-Method, and (for tools/call) Mcp-Name headers over HTTP. Legacy clients keep working unchanged. No client migration is required, and legacy support stays as long as real clients depend on it.
The server implements tools only. It has no prompts, resources, completion, subscriptions, tasks, or MRTR flows, on purpose. The official conformance suite runs against both eras with the documented baseline in docs/conformance-baseline.yml.
iMessage Max ships icons for the main MCP protocol surface and the client packaging surfaces that use their own metadata:
2025-11-25 initialize responses include PNG serverInfo.icons..codex-plugin/plugin.json and uses assets/codex/icon.png plus assets/codex/logo.png.mcpb/manifest.json and uses PNG assets under mcpb/assets/.The committed PNG source set is under assets/icons/ at 16x16, 32x32, 64x64, 128x128, 256x256, and 512x512.
Required to read ~/Library/Messages/chat.db:
For Homebrew installs, the binary is at /opt/homebrew/Cellar/imessage-max/VERSION/bin/imessage-max (not the symlink at /opt/homebrew/bin/). Find it with:
# Open the folder containing the actual binary
open $(dirname $(readlink -f $(which imessage-max)))
For source builds, add .build/release/imessage-max from your clone directory. After changing the grant, relaunch the server; macOS applies Full Disk Access only to processes started after the change.
In the file picker, press ⌘+Shift+G and paste the path to go straight there.
Required to resolve phone numbers to names. The server only asks for access when it is started from a terminal; launchd and MCP clients start it headless, and a headless process never prompts. Grant access once with imessage-max --request-contacts-access from a terminal, then restart the service. --contacts-policy request|skip (or IMESSAGE_MAX_CONTACTS_POLICY) overrides the terminal detection.
System Settings → Privacy & Security → Contacts → add imessage-max is the manual alternative.
Add imessage-max to your MCP client's server configuration.
Many MCP clients use a JSON structure like this:
For Homebrew:
{
"mcpServers": { "imessage": { "command": "/opt/homebrew/Cellar/imessage-max/VERSION/bin/imessage-max" } } }
For source builds:
{
"mcpServers": { "imessage": { "command": "/path/to/imessage-max/swift/.build/release/imessage-max" } } }
If your client uses a different config format, point it at the same binary path.
After saving the config, reconnect or restart your MCP client. The server should appear in the available tools, and you can verify the connection with diagnose.
Find chats by participants, name, or recent content.
By default, chats Messages.app has filtered into Unknown Senders or Junk are hidden. The response carries filtered_hidden, the number of chats the filter removed from this view; pass include_filtered=True to see them.
find_chat(participants=["Contact A"]) # Find a direct chat
find_chat(participants=["Contact A", "Contact B"]) # Find a group with both find_chat(name="Project Group") # Find by chat name find_chat(contains_recent="latest draft") # Find by recent content find_chat(name="Project Group", include_filtered=True) # Also search junk / unknown-sender chats
Inspect a known thread without opening the full conversation.
get_chat_details(chat_id="chat123") # Participants, handles, state, last message
get_chat_details(chat_id="chat123", include_shared_summary=false) # Skip recent shared summary
Retrieve messages with flexible filtering. Returns metadata for media. Explicit chat_id lookups are never filtered.
get_messages(chat_id="chat123", limit=50) # Recent messages
get_messages(chat_id="chat123", since="24h") # Last 24 hours get_messages(chat_id="chat123", from_person="Contact A") # From specific person get_messages(chat_id="chat123", has="links") # Messages that contain a URL
has filters by content type (links, attachments, images). links includes link messages Messages stores as URL preview balloons (the common case on macOS 26).
Group system messages (renames, members added or removed, someone leaving) come back with text: null and an event object: {"type": "rename", "title": "Trip"}, {"type": "participant_added", "participant": "Alice"}, {"type": "participant_removed", ...}, {"type": "left"}, or {"type": "other", "item_type": N} for event kinds the server does not name. Chat previews describe the same events in words ("renamed the group to Trip").
Reactions, replies, edits: each message may include reactions (["❤️ alice"] for standard tapbacks; custom emoji show the emoji itself; sticker reactions use the token 🩵 sticker), reply_to (msg_ of the originator), reply_count, and edited: true. Removed tapbacks are omitted. All four fields are optional and omitted when empty. search and get_context carry the same four fields. These are read-only; diagnose still reports tapbacks and edit_unsend as unsupported because the server cannot send them.
New messages across all chats after a ROWID cursor, in arrival order. Pass the returned next_rowid back as since_rowid to page or poll. next_rowid may be larger than the last returned message's rowid because consumed rows (reactions, filtered chats, orphans) advance it. Omit since_rowid to get only the current cursor. Cursors are only valid against this Mac's chat.db.
get_messages_since() # Current cursor only
get_messages_since(since_rowid=234000) # Messages after that row get_messages_since(since_rowid=234000, limit=50) get_messages_since(since_rowid=234000, chat_id="chat123") get_messages_since(since_rowid=234000, include_filtered=True)
| Parameter | Type | Default | Meaning |
|---|---|---|---|
since_rowid | integer | omitted | Exclusive ROWID cursor. Omit or pass -1 for the current cursor only. |
chat_id | string | omitted | Restrict to one chat (chat123 or 123) |
limit | integer | 100 | Maximum messages to return (1–500) |
include_filtered | boolean | false | Include junk / unknown-sender chats |
include_reactions | boolean | true | Attach reaction strings to returned messages |
Example response:
{
"since_rowid": 234000, "messages": [ { "id": "msg_234001", "rowid": 234001, "chat": {"id": "chat12", "name": "Alice Smith"}, "from": "Alice Smith", "text": "on my way", "ts": "2026-09-02T15:00:00Z" }, { "id": "msg_234010", "rowid": 234010, "chat": {"id": "chat40", "name": "Weekend"}, "from": "me", "text": "see you there", "ts": "2026-09-02T15:01:00Z" } ], "next_rowid": 234050, "has_more": false, "curren
Nur der Anfang wird gezeigt; den Rest findest du im Repo