Skip to main content
The OWASP MCP Top 10 names the ten security risks that matter most for MCP. mcpscore checks the part of each risk that is visible on a running server, from outside, without calling a tool. Most of those checks are the Security & Auth category of the score. Two come from elsewhere: one Primitives rule, and the package audit, which scores a published package instead of a running server. This page maps each risk to the rules that cover it and says where the coverage stops.
Every rule line above is a Security & Auth rule. The auth_* rules skip because DeepWiki serves anonymous requests, so there is no authorization flow to grade. A server behind a login grades them; see authenticated servers.

The mapping

A risk is covered in part when mcpscore checks some of what the risk describes. No risk is covered in full: each one also has a part that only code, configuration or the client can show. One Security & Auth rule maps to no OWASP risk. security_malformed_request_handling checks that a malformed request gets the JSON-RPC parse error, which is error hygiene rather than one of the ten risks. The specification sets the same footing for these checks. Its §Security and Trust & Safety says tool descriptions are untrusted unless they come from a trusted server, and that implementers should follow security best practices. The Origin and TLS rules enforce requirements the spec states directly, in Transports §Security Warning and Authorization. Every rule’s citation is in the rules reference.

Where an audit from outside stops

An audit sees what a server publishes and how it answers requests. It never calls a tool, never reads source code and never sees the client. Each gap below follows from one of those limits.
  • MCP01: mcpscore finds a credential the server publishes. It does not see how the server stores tokens, how long they live, or what it writes to its logs.
  • MCP02: mcpscore sees that scopes are advertised. What a granted token can actually do is decided inside the authorization server and the tools.
  • MCP03 and MCP06: the catalog-text rules read for hidden characters and a fixed list of phrasings. A description that misleads in plain words passes. Text returned by a tool is out of reach, because the audit never calls one. A catalog that changes after the audit is out of reach too; running mcpscore in CI catches the change on the next run.
  • MCP04: a package audit reads registry metadata only. It does not download, run or scan the package or its dependencies.
  • MCP05: injection happens when a tool runs with untrusted input. The audit never sends tools/call. Smoke mode calls your own read-only tools with inputs derived from their schemas, not attack payloads. The exceptions are opt-in: the filesystem boundary check sends file tools paths that try to escape their root, and the network boundary check sends fetch tools URLs that try to reach a local listener their policy denies.
  • MCP07: mcpscore checks the authorization discovery chain and that a made-up bearer token is refused. Whether each tool enforces authorization, and whether the server rejects a real token issued for another service, needs credentials and tool calls.
  • MCP08: logging and telemetry live inside the server.
  • MCP09: an inventory of the MCP servers running in an organization is a governance task. mcpscore audits the server you point it at.
  • MCP10: context is shared or kept in the client and in server-side session state, neither of which a request from outside can observe.
For the risks marked Not covered, pair mcpscore with code review, a dependency scanner, and the logging and inventory controls your organization already runs. The testing tools comparison covers which kind of tool answers which question.

Local servers

When you audit a server on localhost, 127.0.0.1 or ::1, two more rules run. They skip for any other address, so they never run in the web audit.
  • security_host_header_validation (MEDIUM): DNS rebinding points a website’s hostname at 127.0.0.1, and the browser then sends that hostname as Host. The rule sends a request the server accepts, then the same request with Host: mcpscore.invalid, and passes when the second one is refused with any 4xx. The official Python and TypeScript SDKs do this by default for a server bound to localhost. Origin validation stays the main defence; this is the second layer.
  • security_local_bind_loopback (MEDIUM): the transport spec says a server running locally should bind only to localhost, not to all network interfaces. The rule connects to the same port on this machine’s network addresses and sends the request the server accepted on localhost. It fails when the server answers there: it listens on 0.0.0.0 or ::, and other machines on the network can reach it. Another program on that port does not count, and a connection that times out leaves the rule unjudged. It runs for http:// targets only. The report records how many addresses were tried, never the addresses.

How the catalog-text rules decide

The five catalog rules read every string a server publishes: instructions, server info, names, titles, descriptions, schema strings, URIs, MIME types, _meta and icon URLs. A finding names where the string is and what class matched. It never repeats the string, so a report cannot leak a credential or replay an injected instruction.
  • catalog_hidden_unicode flags Unicode tag characters outside a valid emoji tag sequence, bidirectional overrides and isolates, control characters other than tab and line breaks, and runs of two or more zero-width characters, directional marks, soft hyphens or variation selectors. One such character alone passes, because emoji and right-to-left text use them.
  • catalog_no_embedded_secrets matches credentials by their provider’s issued format: AWS access key IDs, GitHub, GitLab and Slack tokens, Stripe live keys, Anthropic, OpenAI and Google API keys, private-key headers, JWTs and long bearer tokens. Documentation samples and placeholder values pass.
  • catalog_prompt_injection_phrasing matches a fixed list of directives to the model: overriding earlier instructions, keeping an action from the user, reassigning the model to a persona, and chat-template control tokens. A phrase in matched quotes, listed with a slash, or introduced as an attempt (“attempts to override the system prompt”) names an attack rather than performing it, and passes. No model judges the text.
  • catalog_exfiltration_links reads each string as Markdown, the way a client renders it, including reference-style links and raw HTML. In HTML it reads the attributes a browser fetches on render (<img src>, srcset, <video poster>, <object data>, a stylesheet, icon or preload <link href>, SVG <image href>, a refresh <meta>, a <script src> that runs (a data block such as JSON-LD does not), url() and the strings @import, image-set() and custom properties take, in a style attribute on any element or in a <style> block, url() in an SVG presentation attribute such as fill or cursor, and the values an SVG <set> or <animate> gives an href or such an attribute) or follows on a click (<a href>, <form action>); the same attribute on an element that ignores it is inert. A dns-prefetch or preconnect <link> sends only its hostname, so only the hostname is checked. It flags javascript: and vbscript: links, data: links other than a raster image, and a link or image URL with a slot for the model to fill, such as {conversation}, {{secret}}, ${TOKEN} or <data>. An image loads without a click, so a slot anywhere in its URL counts; a link counts only with the slot in its query, because a path such as /users/{id} is usually an API route. A form action’s query counts whatever the form’s method. Code spans and image alt text are text, not links, and pass, and so does the content of a <template> (unless it declares a shadow root), <script> and <textarea>, and of a <noscript> except in an iframe srcdoc sandboxed without allow-scripts, where it renders. A relative URL is resolved against the document’s <base href>. Markup that cannot be checked fails as uncheckable_markup: a string with markup over 64 KiB (real catalog strings stay far below), Markdown whose reference links, or a long <base href>, would repeat a destination past four times that size, iframe srcdoc documents nested more than three deep, and constructs that browsers and parsers read differently: a <base href> whose host is not a plain ASCII domain, IPv4 or IPv6 address, a CDATA section in HTML, a comment closed early by <!--> or --!>, a <frameset>, a raw-text tag, <base> or SVG/MathML inside a <select>, an end tag that would close a <select>, raw-text content such as <script> or <xmp> inside SVG or MathML, SVG or MathML inside a <template>, </form>, </body> or </html> with elements still open inside, an end tag that closes unclosed elements across the edge of SVG or MathML content, out of a shadow-root <template>, past an open block such as <div> for a custom element, or out of a <video>, <audio> or <picture>, a <source> inside a media element but not its direct child, or a raw-text end tag such as </script/> that a browser accepts and the parser reads as text.
  • catalog_cross_tool_directives reads each sentence that directs the model: one with must, should, make sure or ensure; always, never or do not opening a clause (“Never use …”, ”…, then always call …”, not “the tool always returns”); first or instead before a verb such as call; one that opens with a verb such as call or use, also after please, then or next; or one that opens with when, before, after or if and whose main clause does not start with a subject (“When asked, call …”, not “When available, the tool returns …”). Cues inside quotes or inline code do not count. In such a sentence it flags a reference to any or other tools or servers (“with any tool”, “using another server”, “every tool must …”) as any_tool_directive, and a tool name the server does not list as unlisted_tool_directive. A tool name is code-shaped: in backticks and made of tool-name characters (letters, digits, _, -, . and /, so not iam:CreateRole), joined with _ or ., or camelCase, or a / path with such a segment (mail/send_email, not “read/write”), after call or invoke or before the word tool. Names of the server’s own tools, prompts and resources pass, also behind a client namespace such as mcp__notes__search or notes.search, and so do plain-word mentions such as “pairs well with the GitHub server” and phrases in double quotes or named as attempts (“attempts to call …”). Unlisted names are judged only when every catalog the server declares was listed whole, from the same lifecycle as the text (without a session, the stateless tools page must say resultType: "complete"). MCP method names such as resources/read and ping are protocol, not tools, and pass.
  • All five skip a partial audit as insufficient-data, because a server behind a login shows no catalog without credentials.
  • skills_hidden_unicode, skills_no_embedded_secrets and skills_prompt_injection_phrasing run the same checks on a server’s skills: every frontmatter string and every line of each SKILL.md read. A finding names the skill and the line. Chat-template tokens inside a code block or an inline code span count as an example, not a directive; every other class counts inside code too.
A finding you intend, such as a game whose instructions give the model a role, can be turned off for your own CI in a mcpscore.toml. The public score on mcpscore.dev keeps every rule.

Security findings in CI

--sarif writes every failed rule as a code scanning alert. Security & Auth rules carry GitHub’s security-severity, so they sort into the Security tab’s critical, high, medium and low bands next to your other security alerts.
In that file, the security_origin_validation finding has level error and security-severity 7.0, which code scanning files as high. The CLI reference has the full mapping. The GitHub Action writes the same file through its sarif-path input.

What’s next