---
type: "post"
id: "chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f"
slug: "chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f"
title: "Chromium tracker: XBOW WebP Chrome bug has no CVE; CVE-2026-85045 is the V8 stage"
description: "A Chromium issue-tracker account separates the three bugs in XBOW’s Chrome exploit chain and exposes a CVE attribution mismatch: CVE-2026-85045 identifies the V8 bug, not the animated-WebP stage."
retrieval_nugget: "A Chromium issue-tracker account separates the three bugs in XBOW’s Chrome exploit chain and exposes a CVE attribution mismatch: CVE-2026-85045 identifies the V8 bug, not the animated-WebP stage. The distinction matters because exploit chains combine flaws with different security boundaries and impacts; treating every stage as an equivalent vulnerability obscures what each one actually enables."
published_at: "2026-11-06"
updated_at: "2026-11-06"
record_date: "2026-11-06"
date_kind: "published_at"
topics: ["google"]
entities: ["XBOW","Chrome","Chromium","Google","Crashpad","V8"]
related_patterns: ["agent-security-moves-to-runtime-boundaries"]
novelty_label: "new_fact"
what_changed: "The tracker record says the three bugs were separately rewarded $4,500 total, while the complete chain earned $250,000; it also identifies the chain’s issue number, filing date, and fixed status."
prior_refs: ["https://newruntime.com/posts/zero-to-calc-in-six-days-inside-xbow-791f94/"]
source_url: "https://rootnotes.in/article/chrome-full-chain-the-parts-were-priced-at-4500-the-chain-at-250000"
source_urls: ["https://rootnotes.in/article/chrome-full-chain-the-parts-were-priced-at-4500-the-chain-at-250000"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f/","markdown":"https://newruntime.com/posts/chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f.md","json":"https://newruntime.com/posts/chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f.json"}
markdown_url: "https://newruntime.com/posts/chromium-tracker-xbow-webp-chrome-bug-has-no-e2a38f.md"
---

# Chromium tracker: XBOW WebP Chrome bug has no CVE; CVE-2026-85045 is the V8 stage

## Retrieval answer

A Chromium issue-tracker account separates the three bugs in XBOW’s Chrome exploit chain and exposes a CVE attribution mismatch: CVE-2026-85045 identifies the V8 bug, not the animated-WebP stage. The distinction matters because exploit chains combine flaws with different security boundaries and impacts; treating every stage as an equivalent vulnerability obscures what each one actually enables.

## Summary

A Chromium issue-tracker account separates the three bugs in XBOW’s Chrome exploit chain and exposes a CVE attribution mismatch: CVE-2026-85045 identifies the V8 bug, not the animated-WebP stage. The distinction matters because exploit chains combine flaws with different security boundaries and impacts; treating every stage as an equivalent vulnerability obscures what each one actually enables.

## What happened

Root Notes reports that Chromium issue 547965190 describes a full-chain exploit submitted by XBOW against signed stock Chrome Beta 152.0.7977.42 on Windows x64. The issue was filed on 18 August 2026, marked P0 and fixed. The reported chain ended with the existing unsandboxed Crashpad handler creating `cmd /k calc.exe` through its normal `CreateProcessW` path. The article says three separate issues blocked that chain, and that Google’s Vulnerability Rewards Program panel awarded $250,000 for the full chain.

The first issue, 547819997, concerns V8’s Maglev compiler. The article says it created a mutable number object and reused it while reconstructing other objects, allowing distinct JavaScript objects to share one box. Google’s stated impact was arbitrary read/write within the V8 cage from ordinary web content; the V8 heap sandbox contained that primitive, and native renderer code execution required a separate sandbox escape. Root Notes identifies this as CVE-2026-85045, a V8 race condition fixed in Chrome 152.0.7977.82 on 3 September.

The second issue, 548509153, concerns Blink’s animated-WebP decoder. The reported bug used frame geometry measured from one image against bytes that had since been replaced, producing a chosen write outside V8’s pointer cage. Root Notes says the Chromium issue carries the `Security_Impact-None` label, is rated P2 and S2, and has an empty CVE field. A Chromium engineer explained that it was a V8 sandbox escape, not an operating-system sandbox escape. This stage is the specific attribution point: the article says it has no CVE, despite XBOW’s award post attaching CVE-2026-85045 to the animated-WebP bug.

The third issue, 545820931, involves Crashpad. According to Root Notes, a failed memory read was used to infer how much memory could be read, while a renderer thread changed memory protections between attempts. The article identifies this as CVE-2026-78952, an out-of-bounds write fixed in Chrome 152.0.7977.64 on 25 August. It was the stage crossing an operating-system boundary. The article reports individual rewards of $1,000 for each CVE-bearing bug and $2,500 for the WebP issue, compared with $250,000 for the assembled chain.

## What the archive adds

This is a new fact on a known story, not a new exploit announcement. The archive’s [account of XBOW’s Chrome full-chain exploit](https://newruntime.com/posts/xbow-built-a-chrome-full-chain-exploit-affecting-097c4b/) covered the six-day effort and the reported use of GPT-5.6 Sol. [A later technical account](https://newruntime.com/posts/zero-to-calc-in-six-days-inside-xbow-791f94/) described the V8, animated-WebP and Crashpad stages. Root Notes adds a tracker-based distinction: the WebP issue’s CVE field is empty, while CVE-2026-85045 belongs to the V8 issue.

## Trend signal

The distinction between a mitigation boundary and an operating-system boundary supports [agent security moves to runtime boundaries](https://newruntime.com/patterns/agent-security-moves-to-runtime-boundaries/). In this case, the relevant boundary is not an AI agent runtime but Chrome’s layered security architecture: a V8 cage escape is not itself an OS sandbox escape, and a chain’s impact depends on crossing the later boundary. This is a narrow example, not evidence about agent runtime security directly.

## Why now

The attribution matters now because the chain and its reward have already drawn attention, while the tracker evidence gives readers a more precise account of which stage carries which identifier. The issue dates, fix versions and CVE fields let teams distinguish the V8 and Crashpad fixes from the WebP issue rather than repeating a merged description of the chain.

## Why it matters

Security teams need to reason about exploitability as a path, not just a list of bug labels. Here, the V8 bug’s stated primitive remained within the V8 cage; the WebP stage escaped that cage but was not an OS-sandbox escape; Crashpad was the stage that crossed an operating-system boundary. Those distinctions change how defenders interpret severity and what a fix addresses.

The reward figures also show the difference between valuing individual defects and demonstrated end-to-end impact. Root Notes reports $4,500 across the three individual awards against $250,000 for the chain. That is evidence about this case and program decision, not a universal pricing rule. For teams building or securing browser-based products, the practical lesson is to preserve the provenance of each stage: issue number, affected boundary, fix version and CVE status. A chain summary can otherwise transfer an identifier from one component to another and make follow-up harder.

## Limits and uncertainty

This account does not establish that the WebP issue will later receive a CVE; its field is reported as empty. It does not establish whether other Chromium-based browsers contain the same code or were affected. The article describes a Windows x64 Chrome Beta chain and does not prove exploitation against real users. The reported $40,000 in API charges and six-day development period are attributed to XBOW, and the article says the charge figure was reconstructed from archived logs rather than published as a billed line item. The reward and technical details are reported from tracker material; they should not be generalized to other vulnerability programs or exploit chains.

## Reader takeaway

Keep the chain’s stages separate. In Root Notes’ tracker-based account, CVE-2026-85045 is the V8 race condition, CVE-2026-78952 is the Crashpad out-of-bounds write, and the animated-WebP issue has no CVE listed. The distinction matters more than repeating the headline-sized chain reward: each stage crosses a different boundary and has a different documented status.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Google](https://newruntime.com/topics/google.md): Explore the google topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Sources](https://newruntime.com/sources/): See the original sources this archive cites.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)