slack-mcp
Multi-workspace Slack integration supporting read and write operations with draft+confirm safety, vault auto-export, and triple-mode authentication (xoxc, xoxp, xoxb). Provides tools for listing and searching channels and users, reading channels and threads, searching messages, sending and replying to messages, updating or deleting own messages, adding reactions, and marking read.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@slack-mcp list my workspaces"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
slack-mcp
Multi-workspace Slack MCP server with draft+confirm safety, vault auto-export, and triple-mode auth.
One process, N workspaces, every write goes through draft → confirm. Built because the existing Slack MCP servers are single-workspace per instance and the Anthropic reference impl was deprecated.
Why use this over the alternatives
This | Anthropic connector | korotovsky/slack-mcp-server | |
Multi-workspace per server | Yes | No | No |
Draft + confirm on writes | Yes | No | No |
Vault auto-export of reads | Yes | No | No |
Prompt-injection scrubber | Yes | No | No |
Audit log | Yes | No | No |
xoxc / xoxp / xoxb auth | Yes | n/a | Yes |
Status | Active | Active (limited) | Active |
Related MCP server: mcp-notify
Install
Open Claude Code, paste:
/plugin marketplace add adelaidasofia/slack-mcp
/plugin install slack-mcp@slack-mcpThen fill in tokens in .env (see SETUP.md for cookie extraction walkthrough).
git clone https://github.com/adelaidasofia/slack-mcp.git ~/.claude/slack-mcp
cd ~/.claude/slack-mcp
pip3 install --break-system-packages -r requirements.txt
cp .env.example .env
# fill in tokens — see SETUP.md for cookie extraction walkthroughRegister in your project .mcp.json (or via claude mcp add -s user):
{
"mcpServers": {
"slack": {
"command": "python3",
"args": ["-m", "slack_mcp.server"],
"cwd": "/Users/YOU/.claude/slack-mcp"
}
}
}Restart Claude Code, then run claude mcp list to verify slack shows up.
Tools (v0.1.1)
Meta: list_workspaces, healthcheck
Read: list_channels, search_channels, read_channel, read_thread, list_users, search_users, get_user_profile, search_messages
Write (draft+confirm): send_message → confirm_send, send_reply_quote → confirm_send, update_own_message → confirm_send, cancel_draft
Write (low-consequence, immediate): add_reaction, mark_read, delete_own_message
Auth modes
Mode | Token shape | Capability | When to use |
|
| Full (search, internal APIs) | Default. Best for personal use across multiple workspaces. |
|
| Permanent, OAuth, search restricted on free plans | If you don't want occasional cookie re-extraction. |
|
| Bot-only, no search, invited channels only | Rare. Use only when you want bot semantics. |
Cookie extraction takes ~3 minutes per workspace. See SETUP.md.
Vault auto-export
Every read_channel call mirrors the channel to <vault>/🤖 AI Chats/Slack/<workspace>/<channel>.md. Idempotent (same channel + day overwrites cleanly). Disable via SLACK_MCP_VAULT_EXPORT=false.
Configuration
All config via env vars (loaded from .env at process start):
SLACK_WORKSPACES— comma-separated list of aliases (e.g.onde,mycelium)SLACK_PRIMARY_WORKSPACE— default workspace when tool calls omitworkspacePer-workspace:
SLACK_WORKSPACE_<ALIAS>_TYPE,_TOKEN,_COOKIE(xoxc only),_TEAM_ID,_LABELSLACK_MCP_VAULT_PATH— vault root for the auto-export mirror (default~/Documents/Vault)SLACK_MCP_VAULT_EXPORT—true/false(defaulttrue)SLACK_MCP_AUDIT_LOG_PATH— JSONL audit log pathSLACK_MCP_DRAFT_TTL_SECONDS— draft expiration (default 3600)SLACK_MCP_SCRUB_PROMPT_INJECTION—true/false(defaulttrue)
Safety patterns
Draft + confirm. Every send (
send_message,send_reply_quote,update_own_message) returns adraft_id. Nothing posts untilconfirm_send(draft_id)is called. Drafts expire after 1 hour. One-time confirm.update_own_messagedrafts include a before/after preview so the operator can diff before confirming the edit.Workspace required on writes. No global default for sends — every write tool requires
workspaceto prevent wrong-workspace posts when channel names collide.Audit log. Every tool call appends a JSONL record. Tokens are redacted.
Prompt-injection scrubber. Incoming message text is sanitized for known prompt-injection patterns (zalgo, role-spoof headers, fake fences). Hits are wrapped not deleted so the operator sees the attempt.
Token redaction.
list_workspacesreturns redacted profiles only. Tokens never appear in tool responses.
Related MCPs
Same author, same architecture pattern (FastMCP, draft+confirm on writes where applicable, vault auto-export, MIT):
imessage-mcp - macOS iMessage
whatsapp-mcp - WhatsApp via whatsmeow
google-workspace-mcp - Gmail / Calendar / Drive / Docs / Sheets
apollo-mcp - Apollo.io CRM + sequences
substack-mcp - Substack writing + analytics
luma-mcp - lu.ma events
parse-mcp - markitdown / Docling / LlamaParse router
rescuetime-mcp - RescueTime productivity data
graph-query-mcp - vault knowledge graph queries
investor-relations-mcp - seed-raise pipeline tracker
vault-sync-mcp - bidirectional vault sync
Telemetry
This plugin sends a single anonymous install signal to myceliumai.co the first time it loads in a Claude Code session on a given machine.
What is sent:
Plugin name (e.g.
slack-mcp)Plugin version (e.g.
0.1.0)
What is NOT sent:
No user identifiers, names, emails, tokens, or API keys
No file paths, message content, or anything from your work
No IP address is stored after dedup processing
Why: Helps the maintainer know which plugins people actually install, so attention goes to the ones that get used.
Opt out: Set the environment variable MYCELIUM_NO_PING=1 before launching Claude Code. The hook will skip the network call entirely. Already-pinged installs leave a sentinel at ~/.mycelium/onboarded-<plugin> — delete it if you want to reset state.
License
MIT — see LICENSE.
Available Tools
18 toolsadd_reactionA
Add an emoji reaction to a message.
emoji is the colon-stripped name (e.g. "thumbsup", "fire", "white_check_mark"). Low-consequence write — no draft+confirm. Idempotent (Slack returns already_reacted as a soft success).
Inputs: ts (required), emoji (required), channel (required), workspace. Returns: object
cancel_draftB
Cancel an unconfirmed draft. Idempotent.
Inputs: draft_id (required). Returns: object
confirm_sendA
Commit a previously-drafted send. Dispatches by draft.kind:
send_message / send_reply_quote -> chat.postMessage (new message)
update_message -> chat.update (in-place edit of target_ts)
Inputs: draft_id (required). Returns: object
delete_own_messageA
Delete one of your own messages.
Calls chat.delete with (channel, ts). xoxc user tokens can delete their OWN messages without special scopes. Returns ok=True on success. message_not_found is treated as a soft success (already-deleted or wrong ts).
Low-consequence (only your own messages, no destructive cascade) so no draft+confirm. The audit log records every call.
Inputs: ts (required), channel (required), workspace. Returns: object
get_user_profileC
Get full profile for a user_id (U... or W...).
Inputs: user_id (required), workspace. Returns: object
healthcheckA
Verify token validity per workspace + report draft store + audit log path.
For each workspace, calls auth.test. Returns user_id and team domain on success, or error code on failure. Does not retry. A failed auth.test means the token has been invalidated (logout, password change, app revoke) and needs re-extraction per SETUP.md.
Inputs: no named parameters. Returns: object
list_channelsA
List channels in the workspace.
types: comma-separated subset of public_channel, private_channel, mpim, im. Default omits DMs to keep responses short. limit caps results at the API page size.
Inputs: limit, types, workspace. Returns: object
list_usersC
List users in the workspace. Excludes deleted by default.
Inputs: limit, workspace, include_deleted. Returns: object
list_workspacesA
List configured Slack workspaces with auth type and primary status.
Returns a dict with workspaces (list of redacted profiles), primary
(alias of the default workspace), and errors (config errors found at
load time, e.g. malformed env values). Tokens never appear in the
response.
Inputs: no named parameters. Returns: object
mark_readA
Mark a channel read up to a given message ts.
Affects unread state across all your linked devices. Use the ts of the most recent message you've read.
Inputs: ts (required), channel (required), workspace. Returns: object
read_channelA
Read recent messages from a channel.
channel accepts ID (C12345) or name (#general). Newest first. Auto-exports the result to the vault at 🤖 AI Chats/Slack//.md (best-effort).
Inputs: channel (required), limit, latest, oldest, +1 more. Returns: object
read_threadA
Read a thread's parent message and replies.
ts is the parent message timestamp (Slack message ID). channel accepts ID or name. The returned messages list includes the parent as element 0.
Inputs: ts (required), channel (required), limit, workspace. Returns: object
search_channelsB
Substring match against channel names + topics. Case-insensitive.
Inputs: query, types, workspace. Returns: object
search_messagesA
Search messages with Slack's full-text search.
Slack search modifiers work in query: in:#channel, from:@user,
before:YYYY-MM-DD, after:YYYY-MM-DD, "exact phrase". The
explicit channel param is sugar — when provided, prepends
in:<channel> to the query.
Requires xoxc or xoxp auth (xoxb cannot search). Errors loudly with the workspace's actual auth_type if search isn't available.
Inputs: query (required), sort, limit, channel, +2 more. Returns: object
search_usersA
Substring match against user real_name/display_name/email/name.
Inputs: query, workspace. Returns: object
send_messageA
Stage a message send. Does NOT post until confirm_send is called.
Returns a draft_id, the resolved workspace + channel, the text preview, and an expires_at unix timestamp. Drafts expire after 1 hour and can only be confirmed once.
For thread replies, set thread_ts to the parent message's ts. For replies that quote (reply with quote-block in Slack UI), use send_reply_quote instead.
Inputs: text (required), channel (required), thread_ts, workspace. Returns: object
send_reply_quoteA
Stage a reply that threads under target_ts.
Same draft+confirm semantics as send_message. The recipient sees the reply nested under the target message in their Slack UI.
Inputs: text (required), channel (required), target_ts (required), workspace. Returns: object
update_own_messageA
Stage an edit to one of your own messages. Does NOT update until confirm_send is called.
Fetches the current message text up front so the draft preview shows BOTH the BEFORE and AFTER strings — a visual diff in plain Python. After confirm_send, the message is edited in place via chat.update.
Returns a draft_id, the resolved workspace + channel + target_ts,
and a preview dict with before, after, target_ts. Drafts
expire after 1 hour and can only be confirmed once.
Inputs: ts (required), text (required), channel (required), workspace. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 18 tools
Each tool has a clearly distinct purpose: listing, searching, reading, sending (with draft/confirm), reactions, etc. No two tools appear to do the same thing; even send_message and send_reply_quote differ in threading behavior.
All tools follow a consistent verb_noun pattern in snake_case (e.g., list_channels, search_users, delete_own_message). The verbs are varied but appropriate, and there is no mixing of conventions.
At 18 tools, the server is slightly above the ideal 3-15 range but still well-scoped. Each tool has a clear role, covering workspace, channel, user, and message operations without bloat.
The tool surface covers core Slack interactions: messaging (with drafts), searching, reactions, and user profiles. Minor gaps like channel creation or file uploads exist, but the set is complete for common read/write workflows.
Maintenance
Related MCP Connectors
Hosted Slack MCP server for Claude, ChatGPT and Cursor, with each action switched per account.
One MCP server for all your MCP servers: code mode, per-tool permissions, approvals and audit.
MCP server for mandates, delegation, policy-gated execution, credential grants, and audit.
One OAuth connector for your team's remote MCP servers, with Tool Packs, guardrails and logs.
Related MCP Servers
- AlicenseBqualityDmaintenanceThe most powerful MCP server for Slack Workspaces. This integration supports both Stdio and SSE transports, proxy settings and does not require any permissions or bots being created or approved by Workspace admins 😏.21,901MIT
- AlicenseAqualityFmaintenanceA minimal MCP server for sending Slack messages, supporting both bot and user tokens with concurrent safety and an AI bot signature in user-token mode.116 npm1MIT
- AlicenseNot gradedqualityCmaintenanceSelf-hosted MCP server for Slack using User OAuth Token, enabling message management, channel operations, and user interactions via natural language.156 npmMIT
- AlicenseAqualityDmaintenanceMCP server that reads Slack through your own browser session, authenticating as you rather than a bot. Provides tools to search messages, retrieve channel/DM/thread history, resolve users, and post messages (write disabled by default).10MIT