Cursor, Codex, Gemini CLI, Antigravity hit by sandbox escapes
By Dillip Chowdary • Jul 21, 2026 • Source: BleepingComputer
Security reporting from BleepingComputer detailed sandbox escapes affecting Cursor, Codex, Gemini CLI, and Antigravity. Researchers demonstrated that AI agents could be manipulated into writing files that trusted host tools later run, prompting multiple CVEs, official patches, and Google downgrading two Antigravity findings.
The attack vector hinges on specific product mechanics between agent outputs and host system operations. Under this design, the AI agent writes files to disk which are subsequently executed by trusted host tools, effectively breaking out of the sandbox without directly overriding host execution controls.
What happened
Read BleepingComputer's account next to the product docs, not instead of them. Names and figures in the lede are the ones we can stand behind; everything else below is how teams usually absorb a story like this. If a number, ship date, or quote is not in the source excerpt, it is not in this briefing. That is deliberate — day-one coverage is where invented specifics do the most damage.
Security reporting from BleepingComputer detailed sandbox escapes affecting Cursor, Codex, Gemini CLI, and Antigravity. Researchers demonstrated that AI agents could be manipulated into writing files that trusted host tools later run, prompting multiple CVEs, official patches, and Google downgrading two Antigravity findings.
How it works
Under the hood this is a systems change, not a press-release adjective. Ask what surface area moved — API, policy, hardware, model behavior, or go-to-market — and which of those you actually ship against. A useful working question: if you had to draw the before/after on a whiteboard, which box would you erase? That is the mechanism. Everything else is packaging.
The attack vector hinges on specific product mechanics between agent outputs and host system operations. Under this design, the AI agent writes files to disk which are subsequently executed by trusted host tools, effectively breaking out of the sandbox without directly overriding host execution controls.
Why it matters
Advertisement
Tech Pulse Daily
Developer Action Items
- ☐ Inventory whether Gemini / Google / Cursor runs in prod, CI, staging, or on laptops before you debate severity.
- ☐ Confirm the vendor's fixed build for Gemini / Google / Cursor from the official advisory, then schedule the patch window.
- ☐ If you cannot patch today, isolate the service, rotate tokens that sat on the affected surface, and raise the logging floor.
- ☐ Record the decision and residual risk so the next on-call does not re-litigate whether you are exposed.
Get tomorrow's pulse first
Join engineers who read Tech Pulse before stand-up. Free, weekday mornings.
If you build on or compete with the parties named in Cursor, Codex, Gemini CLI, Antigravity hit by sandbox escapes, the practical hit is on roadmap sequencing and risk reviews this quarter, not on a vague 'future of the industry'. Put one owner on the story, give them a day to read the primary material, and decide whether this is a this-sprint item, a this-quarter item, or noise.
For engineers and system builders, these findings expose a structural weakness in AI agent isolation models. Security boundaries fail when host environments grant implicit trust to host tools executing agent-generated files, turning routine file generation into an indirect local code execution path.
Who is affected
Incumbents, customers, and adjacent open-source projects do not feel this equally. Map the change to your own stack: what you operate, what you buy, and what you will have to explain to a security, legal, or finance review. Partners and resellers often feel it before the end user does — check those contracts before you assume nothing moved.
In the broader market context, the vulnerability impacts several leading AI coding tools concurrently. While affected vendors issued security patches and registered CVEs, Google chose to downgrade two findings associated with Antigravity, reflecting differing vendor assessments of threat severity.
What to watch next
Treat the next two weeks as a verification window. Watch the vendor's own changelog, any regulator or standards follow-up, and whether a competitor ships a matching capability. Do not change production on day-one coverage alone. If nothing new is published in that window, the story was smaller than the headline.
Engineers deploying autonomous developer agents must restrict background execution permissions for host tools interacting with agent-generated files. Security teams should monitor vendor patch advisories, updated CVE classifications, and ongoing host-tool permission model updates across their development environments.
A 3–5 minute news post is a briefing, not a runbook. Keep BleepingComputer and the vendor's primary page in another tab, quote only what they printed, and write down the single decision this story forces (upgrade, wait, or ignore) before you Slack it to the rest of the team. If you need more than that decision, you want the primary docs or a later engineering deep-dive — not another recap of Cursor, Codex, Gemini CLI, Antigravity hit by sandbox escapes.
Advertisement