Skills/npm/slack-mcp

slack-mcp@5.2.1

CAUTION

Publisher claims: "Catch up on Slack without reading it. Unreads, threads, search. Browser-session or hosted OAuth." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher jtalk22 · publishing for 315 daysInstalls 6,045Last detonated 3d ago
What would fix thisWatchDispute verdict

Coverage

Server started and answered tools/list.
Tested
MCP initialize and tools/list · every tool description, for hidden instructions · every file read, file written and host contacted by what ran
Not tested
what the tools do when called with real credentials

Before you give it your token

It will ask for
API_KEYSLACK_API_KEYSLACK_MCP_AUTH_TEST_URLSLACK_MCP_HTTP_AUTH_TOKENSLACK_MCP_TOKEN_STORAGESLACK_TOKEN or --oauth-lifeboat / --refresh-tokensConfiguration, not credentials: SLACK_COOKIE, SLACK_MCP_CHROME_PROFILE, SLACK_MCP_CHROME_USER_DATA_DIR, SLACK_MCP_EXTRACTION_MODE, SLACK_MCP_HTTP_ALLOWED_ORIGINS, SLACK_MCP_HTTP_INSECURE, SLACK_MCP_KEYCHAIN_TIMEOUT_MS, SLACK_MCP_NO_UPSELL, SLACK_MCP_PROFILE, SLACK_MCP_PROVENANCE, SLACK_MCP_READ_ONLY, SLACK_MCP_SKILLS, SLACK_MCP_SKILLS_DIR, SLACK_MCP_TOOLS, WRITE_PATH_TOOLS
Where it talks to
revasserlabs.com (its own)Also slack.com — not the publisher's domain; check what for before connecting.
What the honeypot saw
no credential files read · no network
What we could not test
what the tools do when called with real credentials

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names slack.com as well as its own vendor. The guard injects API_KEY, SLACK_API_KEY, SLACK_MCP_AUTH_TEST_URL, SLACK_MCP_HTTP_AUTH_TOKEN, SLACK_MCP_TOKEN_STORAGE, SLACK_TOKEN only into requests to revasserlabs.com (and its subdomains) and refuses the rest.

$vet-skills run --secrets API_KEY=revasserlabs.com,SLACK_API_KEY=revasserlabs.com,SLACK_MCP_AUTH_TEST_URL=revasserlabs.com,SLACK_MCP_HTTP_AUTH_TOKEN=revasserlabs.com,SLACK_MCP_TOKEN_STORAGE=revasserlabs.com,SLACK_TOKEN=revasserlabs.com -- npx -y @jtalk22/slack-mcp

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios) on 2026-11-11 and scored it 64/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Prompting user to enter sensitive credentials; document.cookie access; Prompting user to enter sensitive credentials. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 24 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
3.7sprocessdependency install exit 0
24sclaimed20 tools: slack_token_status, slack_health_check, slack_refresh_tokens, slack_list_conversations, slack_conversations_history, slack_get_full_conversation, slack_search_messages, slack_users_info, slack_send_message, slack_get_thread, slack_list_users, slack_add_reaction, slack_remove_reaction, slack_conversations_mark, slack_conversations_unreads, slack_users_search, slack_workflow_save, slack_workflows, slack_catch_me_up, slack_session_report
24sprocessscript backend runs setup only
24sprocessno install scripts to re-run

Static findings

CRITICAL
Prompting user to enter sensitive credentialslib/token-store.js:930 · Prompting user to enter sensitive credentials — social engineering"Enter the token manually, or set SLACK_TOKEN and SLACK_COOKIE in the environment.",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST03
MEDIUM
npx invocation of non-well-known packagelib/lifeboat.js:33 · npx invocation of non-well-known package — may execute untrusted codeconst SETUP_CMD = "npx -y @jtalk22/slack-mcp --setup";Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packageskills/slack-workflow-profiles/SKILL.md:55 · npx invocation of non-well-known package — may execute untrusted codenpx -y @jtalk22/slack-mcp --apply-template oncall-handoff --channels C012345,C067890 --priority-people U0PRIMEFix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
document.cookie accessscripts/setup-wizard.js:311 · document.cookie access — can steal session cookies// Token-only one-liner (cookie is HttpOnly and cannot be read via document.cookie)Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.
AST07
CRITICAL
Prompting user to enter sensitive credentialsscripts/token-cli.js:153 · Prompting user to enter sensitive credentials — social engineeringconsole.log("Or enter the token by hand: npx -y @jtalk22/slack-mcp --setup");Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST03
MEDIUM
npx invocation of non-well-known packagescripts/setup-wizard.js:578 · npx invocation of non-well-known package — may execute untrusted codeIS_MACOS ? "SLACK_MCP_TOKEN_STORAGE=keychain-only npx -y @jtalk22/slack-mcp --setup"Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandlib/workflow-store.js:11 · File permission modification command — agent changes file ownership/permissions* File: ~/.slack-mcp-workflows.json (chmod 600)Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
HIGH
Collapsed <details> block carrying instructionsREADME.md:259 · Collapsed <details> block carrying instructions<summary><strong>Author labels on every message</strong></summary> Each message carries `origin` (`self`, `internal`, `external`, `bot`, `unknown`) and `author_Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
npx invocation of non-well-known packagelib/handlers.js:312 · npx invocation of non-well-known package — may execute untrusted codenext_action: "Run npx -y @jtalk22/slack-mcp --setup"Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Collapsed <details> block carrying instructionsREADME.md:281 · Collapsed <details> block carrying instructions<summary><strong>Read-only mode, for when a label is not enough</strong></summary> ```bash npx -y @jtalk22/slack-mcp --read-only # or SLACK_MCP_READ_ONLY=1 ``` Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
File permission modification commandscripts/setup-wizard.js:68 · File permission modification command — agent changes file ownership/permissions: 'Tokens saved with chmod 600';Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
npx invocation of non-well-known packagescripts/setup-wizard.js:624 · npx invocation of non-well-known package — may execute untrusted code"npx -y @jtalk22/slack-mcp --read-only # or SLACK_MCP_READ_ONLY=1, if this agent only needs to read");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandscripts/setup-wizard.js:587 · File permission modification command — agent changes file ownership/permissions`At ${TOKEN_FILE}. Restrict it with the file's own permissions; chmod does not apply here.`);Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
npx invocation of non-well-known packagescripts/token-cli.js:134 · npx invocation of non-well-known package — may execute untrusted code// Reachable from an npx install as `npx -y @jtalk22/slack-mcpFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandlib/workflow-store.js:12 · File permission modification command — agent changes file ownership/permissions* Atomic write pattern (temp → chmod → rename) prevents corruption fromFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
npx invocation of non-well-known packagelib/slack-client.js:342 · npx invocation of non-well-known package — may execute untrusted codethrow new Error("No credentials available. Run npx -y @jtalk22/slack-mcp --setup or open Slack in Chrome.");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/setup-wizard.js:469 · npx invocation of non-well-known package — may execute untrusted codeprint("Run setup wizard to refresh: npx -y @jtalk22/slack-mcp --setup");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File type combined with upload actionskills/slack-workflow-profiles/SKILL.md:14 · File type combined with upload action — agent may exfiltrate local data files`~/.slack-mcp-workflows.json`. Saving one sends nothing to Slack.Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
npx invocation of non-well-known packagelib/handlers.js:340 · npx invocation of non-well-known package — may execute untrusted codenext_action: "Run npx -y @jtalk22/slack-mcp --setup"Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagelib/handlers.js:271 · npx invocation of non-well-known package — may execute untrusted code? "Run npx -y @jtalk22/slack-mcp --setup"Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/token-cli.js:153 · npx invocation of non-well-known package — may execute untrusted codeconsole.log("Or enter the token by hand: npx -y @jtalk22/slack-mcp --setup");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandscripts/setup-wizard.js:194 · File permission modification command — agent changes file ownership/permissionsprint(` 1) Token file (chmod 600) + macOS Keychain backup${current !== 'keychain-only' ? ` ${colors.dim}(current)${colors.reset}` : ''}`);Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File permission modification commandscripts/setup-wizard.js:599 · File permission modification command — agent changes file ownership/permissions"Readable by more than you.", `chmod 600 ${TOKEN_FILE}`);Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
MCP server using network transport (SSE/HTTP)server.json:9 · MCP server using network transport (SSE/HTTP) — data sent over network"src": "https://raw.githubusercontent.com/jtalk22/slack-mcp-server/main/docs/assets/icon-512.png",Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
MEDIUM
Process kill commandsrc/server.js:58 · Process kill command — agent can terminate processes// handlers below to kill zombie processes on stdin EOF and signals.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagescripts/setup-wizard.js:450 · npx invocation of non-well-known package — may execute untrusted codeprint("Run setup wizard: npx -y @jtalk22/slack-mcp --setup");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/setup-wizard.js:439 · npx invocation of non-well-known package — may execute untrusted codeprint(` ${colors.cyan}claude mcp add slack -- npx ${args.join(" ")}${colors.reset}`);Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Hidden zero-width characters detectedlib/skills.js:270 · Hidden zero-width characters detected — may conceal contentconst FRONTMATTER = /^?---[ \t]*\r?\n[\s\S]*?\r?\n---[ \t]*(?:\r?\n|$)/;Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST07
HIGH
Collapsed <details> block carrying instructionsREADME.md:67 · Collapsed <details> block carrying instructions<summary><strong>Side by side: the managed integration path vs. the local session path</strong></summary> <br> | | Slack official MCP | **Slack MCP Server — locFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
npx invocation of non-well-known packagesrc/server.js:44 · npx invocation of non-well-known package — may execute untrusted codeconsole.error("Use npx -y @jtalk22/slack-mcp --setup to configure credentials");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/web-server.js:466 · npx invocation of non-well-known package — may execute untrusted codeconsole.error("Run: npx -y @jtalk22/slack-mcp --setup");Fix: Pin versions and sources; vendor anything fetched at install time.
AST06

Cost to the agent

Instructions loaded
≈2,872 tok
3 files · 44 in bundle
Tool list per prompt
≈3,475 tok
20 tools · 174 per tool
Handshake
491 ms
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 157

Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.

Known vulnerabilities

No advisories in OSV for @jtalk22/slack-mcp@5.2.1 as of 3d ago. This is the package itself, not its dependencies.

What would make this Verified

16 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Prompting user to enter sensitive credentials≈ +9 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  2. Unclaimed publisher≈ +8 on the totalClaim jtalk22: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  3. Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
  4. document.cookie access≈ +4 on the totalReplace encoded or assembled strings with plain literals and remove undeclared network calls.
  5. Hidden zero-width characters detected≈ +4 on the totalKeep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
  6. npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.
Publish a new version to npm, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim jtalk22Propose a patch to the maintainer