NitronEdge TechnologiesHelping Enterprises Harness Innovation
Software supply chain · Chainguard

Chainguard adds a feed for breaking security fixes

Chainguard now offers an opt-in feed for security fixes that break APIs, and shared how its Checks CI rode out GitHub outages. What it means for you.

On 5 November 2026, Chainguard added a second package endpoint for members of Athena, its service that patches vulnerable open source libraries. The new endpoint carries security fixes that include breaking changes, which many teams want but cannot safely apply on autopilot. A day later, Chainguard described how it moved its busiest repository from GitHub Actions to Chainguard Checks, its own CI service, now in beta.

What changed

Athena members already pull patched packages from an endpoint called `cgp`. It only delivers fixes that stay backward compatible, so they can flow into builds without anyone checking each one. That is safe, but it leaves a gap. Some vulnerabilities can only be fixed by changing an API or behaviour, and those fixes never reached the safe feed.

The new `cgb` endpoint fills that gap. It mirrors everything in `cgp` and adds the fixes that break compatibility. Teams only receive it if they change their package manager settings on purpose, so nobody gets a breaking change by surprise. Vulnerability data in OSV and VEX formats treats `.cgb` releases as fixed versions, which means scanners will show the issue as resolved once you move to one.

Chainguard says changelogs describing the API changes are coming, with older patches to be documented over time. It also plans a similar opt-in line for its public `.cgr` patches.

The Checks post is a case study of Chainguard's own move. Its main repository grew from roughly 500 merges a day to about 2,000, much of it from bots. Checks runs each build in its own short-lived virtual machine, retries jobs when the infrastructure fails, blocks network access by default and only runs the checks a change actually affects. Chainguard reports image builds about twice as fast, and three GitHub Actions incidents in October that did not slow its rate of change. GitHub still hosts the code, so an outage there still matters.

Why it matters

Most organisations patch open source on two speeds. Minor fixes go in automatically. Anything that might break an application waits for a developer, and that queue is where serious vulnerabilities sit for months. A separate, clearly labelled feed for breaking fixes does not remove that work, but it makes the choice explicit and visible. You can see which fixes exist, decide which services can take them, and keep the safe feed running for everything else.

The OSV and VEX support is easy to overlook. Security teams spend a lot of time arguing with scanners that still flag a vulnerability after it has been patched. Accurate machine-readable status saves that time and makes audit evidence cleaner.

The Checks story speaks to a newer problem. As AI coding agents and update bots open more pull requests, CI becomes a bottleneck and a security control at the same time. Isolated build machines, short-lived identities for each check and no network access by default are the kind of controls auditors now ask about after recent CI supply chain attacks.

What this means for organisations in the Middle East, India, Africa and North America

Software supply chain rules are tightening across our markets. In the UAE and Saudi Arabia, national cybersecurity controls for government and critical sectors expect organisations to manage third-party and open source components. India's CERT-In guidance and the RBI's IT governance directions push banks to track and patch vulnerabilities quickly. In the USA, federal buyers increasingly ask for SBOMs and secure development attestations, and Canadian federal and financial regulators expect similar evidence.

In each case, the hard question is the same: when a fix exists, why has it not been applied? A labelled feed for breaking fixes gives teams a clean way to answer that, service by service. For organisations with small engineering teams, which is common in parts of Africa and among mid-sized firms in the Gulf, the value is in fewer manual patch reviews, not more tools.

Checks is still in beta, so it is something to evaluate rather than standardise on. It is most relevant where bots or AI agents now produce a large share of code changes.

What to do next

  1. Measure how long breaking security fixes sit unapplied in your main applications today.
  2. If you use Athena, pick one service with good test coverage and trial the `cgb` endpoint there first.
  3. Make sure your scanners read OSV or VEX data, so patched packages stop showing as open findings.
  4. Review your CI for shared credentials, open network access and long-lived runners, especially where bots open pull requests.
  5. Keep a record of which services take breaking fixes automatically and who approves the rest.

To talk about securing your software supply chain, see our Chainguard partner page or our cybersecurity services.

Sources

  1. Chainguard endpoints now handle breaking changes (Chainguard blog) (2026-11-05)
  2. GitHub Actions down? No problem. Green with Chainguard Checks. (Chainguard blog) (2026-11-06)
Related: Chainguard
Enquire

Talk to us about Chainguard

Get a free consultation from our AI, data and security specialists. Honest advice, clear options and a reply within 6 hours.

Reply timeOur team replies within 6 hours

Request a callback

Free consultation. Our team replies within 6 hours.

Your details are safe. Our team replies within 6 hours. By sending, you agree to our privacy notice.

WhatsApp