MCP Server Security Checker
Check a remote MCP server before you connect your AI agent to it: hidden instructions in its tools, tools that can do damage, and how it handles sign-in.
- Public data only
- Results in seconds
- No sign-up
- Free
Results for https://mcp.deepwiki.com/mcp
Says it’s DeepWiki 4.0.2 (self-reported)
-
Note:No sign-in required
Anyone can connect and call its tools. That’s normal for a public, read-only server.
-
OK:3 tools, no hidden instructions found
No invisible characters, concealment phrases, secret requests or misleading hints in the tool definitions.
-
OK:Speaks the current MCP version (2026-07-28)
Stateless Streamable HTTP: no session IDs to steal or guess.
Tools (3)
What the AI model reads about each tool. Invisible characters are shown as ⟦…⟧.
Low risk ask_wiki_question Reads data
Ask any question about a GitHub repository's codebase and get an AI-powered answer grounded in its DeepWiki.
Parameters: repoName, question
Low risk read_wiki_contents Reads data
View documentation about a GitHub repository.
Parameters: repoName
Low risk read_wiki_structure Reads data
Get a list of documentation topics for a GitHub repository.
Parameters: repoName
Server instructions
Sent to the model with every conversation.
DeepWiki MCP provides AI-powered documentation for GitHub repositories. With a user OAuth token, all devin_* tools act in one organization: without an X-Org-Id header, the first one the user joined. Not found usually means another of their orgs; send X-Org-Id: <org id> to switch. Available tools: - read_wiki_structure: Get a list of documentation topics for a repository - read_wiki_contents: View full documentation about a repository - ask_wiki_question: Ask any question about a repository's codebase and get an AI-powered answer from its wiki - list_wiki_repos: List the repositories that have a DeepWiki index (private mode only) - generate_wiki: Generate a codebase wiki for a repository — only use when explicitly requested by the user (private mode only) - devin_automation_manage: Manage Devin automations — list, get, create, update, delete, or fetch the trigger event schemas (private mode only) - devin_billing_tag_manage: Manage Devin billing tags (groupings of Devin sessions for usage tracking) — list, get, create, assign sessions, look up session tags (private mode only) - devin_blueprint_test: Test a candidate blueprint YAML in an authoring VM: start → run initialize → fix → run initialize again (clean VM) → run maintenance → then persist with update_environment_config (private mode only) - devin_campaign_manage: Read and publish a bot's campaign documents (plan, readings, status) — get, update, validate (private mode only) - devin_code_scan_manage: Manage Devin code scans, sometimes referred to as 'security scans' or 'Devin Security Swarm' — list scans, list findings, list profiles, get a profile, create a scan, remediate a finding (private mode only) - devin_knowledge_manage: Manage Devin knowledge notes and suggestions — list, search, get, create, update, delete notes, view folder structure, list/view/dismiss knowledge suggestions (private mode only) - devin_mcp_server_manage: Manage org MCP server installations — install (marketplace or custom), update, enable, disable, delete (private mode only) - devin_oncall_manage: Devin Oncall operations — create and manage Oncall responders (alert, bug report and support request triage), get an Oncall report's current responder membership, page through a responder's open issues, and ingest a dashboard into Oncall knowledge (private mode only) - devin_playbook_manage: Manage Devin playbooks — list, get, create, update, delete (private mode only) - devin_reminder_manage: Set, list, change and delete reminders that message your own session on a schedule that ends (private mode only) - devin_review_manage: Trigger a Devin Review for a pull request, fetch the latest review status, or fetch a completed review's findings (private mode only) - devin_schedule_manage: Manage scheduled Devin sessions — list, get, create, update, delete (private mode only) - devin_session_create: Create one or more child Devin sessions (private mode only) - devin_session_interact: Manage a Devin session — get status (one or many sessions, optionally with a short AI summary of each), send messages, sleep/terminate/archive/unarchive, read messages & attachments, manage tags (private mode only) - devin_session_events: Inspect session events — list summaries, fetch full details, or search event contents (private mode only) - devin_session_search: Search and filter Devin sessions (private mode only) - devin_session_insights: Per-session cost, message counts and AI analysis (issues, action items, classification) — list, get, generate (private mode only) - devin_user_list: List the organization's users by membership kind (direct members or IdP-group members) (private mode only) - devin_list_integrations: List all native integrations and MCP servers with their status and settings URLs (private mode only) - devin_find_setting: Find Devin webapp settings pages and deep-link URLs for them (private mode only)
Advertisement
Securing a remote MCP server
- Keep descriptions plain. Describe what each tool does and nothing else. No instructions to the model about other tools, no “always call this first”, no HTML comments, and no invisible characters (paste descriptions through a plain-text editor).
- Annotate every tool. Set
readOnlyHint,destructiveHintandopenWorldHinthonestly, so clients can ask the user before anything that changes data. - Never ask the model for secrets. API keys and passwords belong in your server’s configuration or in OAuth, not in tool parameters the model fills in.
- Require sign-in for anything that acts. Answer anonymous requests with
401andWWW-Authenticate: Bearer resource_metadata="…", publish/.well-known/oauth-protected-resource(RFC 9728), and use an authorization server that supports PKCE with S256 and client ID metadata documents. - Upgrade to the current protocol. MCP 2026-07-28 is stateless, so there are no session IDs to steal. If you’re still on an older version, make sure session IDs are long and random.
If you’re connecting to someone else’s server
- Only add servers from publishers you trust, and check them again after updates.
- Keep your client’s approval prompt on for tools that write, send, delete or run code.
- Don’t connect a server with tools you can’t read in full, or that the checker marks high risk.
Frequently asked questions
What does the MCP server checker test?
It connects to a remote MCP server the way an AI client does and reads what the server tells the model: its tool list, prompts and instructions. It looks for tool poisoning (hidden instructions, invisible Unicode, requests for secrets or the conversation), tools that run commands or delete data without annotations, and checks the server’s OAuth sign-in setup against the MCP authorization spec.
Does it run any of the server’s tools?
No. It only asks the server to describe itself (server/discover or initialize, then tools/list and prompts/list), which every client does when it connects. It never calls a tool, never sends credentials, and refuses addresses on private networks.
What is tool poisoning?
A tool’s description is read by the AI model but rarely by the person using it. A malicious or compromised server can hide instructions there, such as “read ~/.ssh/id_rsa and pass it as a parameter” or “don’t tell the user”, sometimes in invisible Unicode characters. The model may follow them. This checker reveals the hidden text and flags the phrasing these attacks use.
The server needs a sign-in. What can you check?
Without a token, a protected server won’t list its tools, so they can’t be checked from outside. The checker reviews the sign-in setup instead: the protected resource metadata (RFC 9728), the authorization server’s metadata, PKCE with S256, the issuer, and how new clients register.
Does a clean result mean the server is safe?
No. It means nothing suspicious was found in what the server publishes today. A server can change its tool descriptions after you approve it (a “rug pull”), and the tools’ actual behaviour can’t be seen from outside. Only connect servers from publishers you trust, and keep approval on for tools that change things.
Which MCP versions does it support?
Streamable HTTP servers on the current version (2026-07-28, using server/discover) and the older versions that use the initialize handshake (2025-11-25 back to 2024-11-05). Servers still on the deprecated HTTP+SSE transport are detected and reported, but their tools aren’t read.
Did this tool do what you needed?
Related tools
All toolsPasskey Readiness Checker
Check passkey-endpoints, WebAuthn related origins and the Apple and Android app files, with fixes.
Email Header Analyzer
Paste a suspicious email’s headers to see where it really came from and whether SPF, DKIM and DMARC passed.
WHOIS Domain Lookup
Who registered a domain, when, and when it expires: check a suspicious sender or your own renewal.