13.35 · Update and Patch Management

Explore 4 source-supported AI security cases for 13.35 · Update and Patch Management in Databricks DASF 3.0, with evidence, scope, and defensive analysis.

Risk category

A client uses insecure update mechanisms or does not apply security patches promptly. Unverified updates and unsafe distribution channels can introduce malicious changes, while delayed patches leave known weaknesses exploitable. Both the authenticity of updates and the timeliness of their installation affect client security.

Official framework source ↗ · Open interactive explorer ↗

Research · · Practical overlap

McPhantom: Coding Agents Bring Retired MCP Servers Back

Glow Labs observed coding assistants installing retired MCP reference servers during ordinary setup work. The important operational gap is that a locally launched package can escape repository-only dependency inventories. The report supports tighter package and server admission, not a claim that those sessions were malicious intrusions.

Why it relates: Automated client setup fails to account for the maintenance and replacement status of installed MCP packages. The shared issue is lifecycle/update selection. The study concerns installed MCP components, not a demonstrated insecure updater in the client itself.

This is observational research about unsafe dependency selection, not a confirmed malicious installation or breach.

AIA · Read the full case ↗

Incident · · Practical overlap

Cursor Marketplace: A Trojanized Claude Code Extension Plants an SSH Backdoor

AI code editor Cursor confirmed malicious Windows/macOS copies of the Claude Code coding-assistant extension, versions 2.1.312 and 2.1.313, in its marketplace on October 29, 2026. Injected code ran at module load, planted an SSH key, attempted a persistent reverse tunnel and reported host information. Attacker login still required working local SSH and an accepted key; compromise of Anthropic’s official distribution infrastructure is not established.

Why it relates: A client distribution/update path supplies modified extension bytes that execute before normal activation. Verifying the exact installed artifact before load is the shared client-update boundary. DASF places this entry under MCP client risks. This incident establishes a native IDE extension update path, not MCP use or an MCP server response; the relation is the substantial shared client-update mechanism.

The review covers the confirmed malicious Cursor extension distribution and analyzed native payload. Practical overlaps explicitly retain differences from model-mediated, ML-lifecycle and MCP-specific deployments; no missing deployment is inferred.

AIA · Read the full case ↗ · SecureFlow · Read the full case ↗

Validated Research · · Practical overlap

Plugin4Shell: Git Reference Shadowing Defeats Reviewed Plugin Pins

Air Security demonstrated that AI coding clients could install different plugin code from the commit reviewed by a marketplace. An attacker controlling an already trusted plugin repository could shadow a pinned SHA with a default branch of the same name, or shadow FETCH_HEAD in Gemini CLI. Background updates then loaded malicious code without a fresh install decision. The defect is verification of the resolved working-tree commit, not a failure to review the marketplace entry; both repository control and a compatible client/hosting path are material prerequisites.

Why it relates: The update mechanism trusts a requested ref without verifying the installed commit, admitting unreviewed client code. The shared client-update integrity mechanism is established; the exploit does not require or prove an MCP transport session.

Controlled client plugin-distribution research with alternative Git checkout routes; no model prompt compromise or observed mass-victim campaign is inferred.

AIA · Read the full case ↗ · SecureFlow · Read the full case ↗

Validated Research ·

Ollama Windows Auto-Update Persistent RCE

Striga and CERT Polska documented a chained flaw in Ollama for Windows 0.12.10 through 0.22.0: attacker-influenced update responses could provide malicious ETag or Content-Disposition path values, the updater could write a payload outside its staging directory such as the Windows Startup folder, Windows signature verification was a no-op, and the payload could run persistently on later logins. Ollama first shipped the hardened update flow in v0.23.3.

Why it relates: Striga and CERT Polska documented a chained flaw in Ollama for Windows 0.12.10 through 0.22.0: attacker-influenced update responses could provide malicious ETag or Content-Disposition path values, the updater could write a payload outside its staging directory such as the Windows Startup folder, Windows signature verification was a no-op, and the payload could run persistently on later logins. The relevant category is Update and Patch Management: A client uses insecure update mechanisms or does not apply security patches promptly.

Striga and CERT Polska documented a chained flaw in Ollama for Windows 0.12.10 through 0.22.0: attacker-influenced update responses could provide malicious ETag or Content-Disposition path values, the updater could write a payload outside its staging directory such as the Windows Startup folder, Windows signature verification was a no-op, and the payload could run persistently on later logins. Ollama first shipped the hardened update flow in v0.23.3.

SecureFlow · Read the full case ↗