Explore 9 source-supported AI security cases for 13.17 · Confused Deputy in Databricks DASF 3.0, with evidence, scope, and defensive analysis.
Risk category
An MCP server acts for the wrong user or uses permissions inappropriate for the request. Its legitimate authority is effectively borrowed to perform an operation the actual requester should not be allowed to perform. In a multi-user environment, this can bypass authorization, expose another user's data, or escalate privileges.
LiteLLM is a gateway that routes applications' requests to language models. Its semantic cache can reuse an earlier answer when two questions have similar meanings. NVISO showed that in the tested configuration, /v1/responses reused cached entries across different API keys. This exposed simulated support data and allowed a fabricated answer or tool-call response to reach a different user. The researchers also demonstrated downstream execution in a mock banking tool and the pi coding agent. CVE-2026-89032 was fixed in v1.101.0-rc.1.
Why it relates: The mock MCP bank acted on a transfer call originating in a different caller’s stored result. The observed overlap is the shared runtime cache or inherited gateway exposure in the reported configuration. The test did not change model weights, compromise package distribution or establish takeover of a live MCP account.
All downstream consequences are from separate tests against the vulnerable cache behavior.
A normalization error in the locally run AWS API MCP Server could cause custom deny and approval lists to miss AWS operations containing consecutive capital letters. AWS IAM remained a separate authorization boundary. The fix is to upgrade the server and verify the effective local policy with the actual operation names.
Why it relates: The server uses its AWS identity for an operation the configured custom policy should deny or require confirmation for.
The mapping concerns the custom local policy boundary, with its exact affected versions and unaffected modes.
Two Filesystem MCP disclosures show a local process replacing a directory with a symlink after validation. One route reads an outside file by reopening its pathname; another moves an outside file into an allowed directory before reading it. The attacker needs local directory-mutation capability and tool calls, while the server must possess the required filesystem access. Version 2.3.0 is the announced repair.
Why it relates: The filesystem server uses legitimate OS authority for a final target that differs from the authorized pathname.
The read and move races are distinct paths; a path check is not an atomic object binding.
pgEdge disclosed independent SQL-injection, read-only-enforcement and bundled-client approval defects affecting versions through 1.0.0 and repaired in 1.1.0. A custom pl-do tool could turn argument data into SQL; query_database could defeat an application-level read-only setting; two client defects skipped write confirmation. Database permissions still bound the resulting actions. The source establishes controlled validation, not a malicious customer breach.
Why it relates: The MCP server can use its write-capable database authority for requests its application-level read-only policy should deny.
Each advisory contributes its own route and prerequisites; client confirmation and database authorization are different controls.
Chainstack provides blockchain infrastructure services. Chainstack traced its October 2026 intrusion from a test Grafana instance to an unauthenticated, cluster-admin MCP server, then to production through compromised machine credentials. The investigation found database access, secret theft and several persistence routes. It did not recover individual MCP calls or establish customer-node modification. ARTEX is the operator’s assessment of the attacker’s tooling, not actor attribution.
Why it relates: The MCP service’s legitimate cluster-admin identity could serve unauthenticated callers, borrowing authority that those callers did not possess.
The relation follows the assessed MCP integration and the documented infrastructure effects; separate observed routes remain separate.
The MCP Python SDK connects AI applications to tools and supports experimental long-running tasks. When a multi-client server enabled the affected default task handlers, another client could list tasks, read results and cancel work without a task-owner check. CVE-2026-52870 was fixed for default generated task IDs in 1.27.2; explicitly supplied IDs and direct TaskStore writes need separate ownership review.
Why it relates: The MCP server uses its legitimate task-store authority for the wrong requesting client.
Experimental MCP task handlers cross client-session ownership; the default-ID patch does not establish custom-store isolation.
Johann Rehberger tested Copilot, the AI query assistant in SQL Server Management Studio (SSMS), and made a database owner's stored instructions induce execution under another user's more privileged SQL connection. A regex-based read-only check failed to contain dynamic SQL, and the final demonstration added the attacker's login to sysadmin. Separate direct-prompt experiments sent table data over SMB; these are distinct demonstrations, not evidence of a single observed intrusion.
Why it relates: The SQL assistant acts as a deputy with a victim's broader authority on a lower-authority metadata writer's instructions. The shared boundary is untrusted context influencing a privileged tool. The tested SQL tools are not established as MCP servers.
Controlled SSMS metadata-to-SQL escalation plus separate direct SQL/SMB tests; no joined intrusion or OS escape is asserted.
Manifold Security showed that an HTML comment hidden in an Azure DevOps PR description was returned verbatim by the official MCP server's PR tool. Claude Code and Copilot CLI agents reviewing the PR followed it to run a pipeline and read a wiki in a project the contributor could not access, then posted the wiki content as a comment on the contributor's PR. MSRC acknowledged and triaged the report; the cited post did not confirm a fix.
Why it relates: Manifold Security showed that an HTML comment hidden in an Azure DevOps PR description was returned verbatim by the official MCP server's PR tool. The relevant category is Confused Deputy: An MCP server acts for the wrong user or uses permissions inappropriate for the request.
Manifold Security showed that an HTML comment hidden in an Azure DevOps PR description was returned verbatim by the official MCP server's PR tool. Claude Code and Copilot CLI agents reviewing the PR followed it to run a pipeline and read a wiki in a project the contributor could not access, then posted the wiki content as a comment on the contributor's PR. MSRC acknowledged and triaged the report; the cited post did not confirm a fix.
Manifold Security reported eight GitSpawn findings across seven AI coding agents. A transferred repository with its local Git configuration intact can make Git execute a configured command during status, diff, or background context collection before the agent's trust prompt; the research did not report malicious real-world compromise, and a normal clone does not preserve this local configuration.
Why it relates: The related official Git MCP advisory confirms that read-only tools can lend the server process’s command authority to repository-local configuration. The MCP server is a separately verified integration. The original graph retains the coding-agent workspace path.
The original coding-agent graph remains intact; the Git MCP advisory adds an independently bounded affected integration and repair.