CVE-2026-32621: Apollo Federation vulnerable to prototype pollution via incomplete key sanitization

Should you act?

Critical, preconditions apply

Severe if exploited, but exploitation needs conditions that may not apply to you. This signal is computed from CISA KEV, EPSS and CVSS data; it is not an editorial verdict.

• No confirmed exploitation. It is not in CISA’s KEV catalog and CISA’s SSVC assessment records no known exploitation. That can change quickly after a public write-up; re-check if you are affected.
• EPSS 0.6%, percentile 49: FIRST’s model estimates a 0.6% chance of exploitation activity in the next 30 days, higher than 49% of all scored CVEs (low). EPSS is a probability, not a verdict.
• Needs an account (low (any account)). Exposure depends on who can log in.
• Only applies if your dependency tree includes @apollo/federation-internals (npm) < 2.9.6, fixed in 2.9.6; @apollo/federation-internals (npm) >= 2.10.0, < 2.10.5, fixed in 2.10.5; @apollo/federation-internals (npm) >= 2.11.0, < 2.11.6, fixed in 2.11.6; @apollo/federation-internals (npm) >= 2.12.0, < 2.12.3, fixed in 2.12.3. Check your lockfile, including transitive dependencies.

What it is

Apollo Federation is an architecture for declaratively composing APIs into a unified graph. Prior to 2.9.6, 2.10.5, 2.11.6, 2.12.3, and 2.13.2, a vulnerability exists in query plan execution within the gateway that may allow pollution of Object.prototype in certain scenarios. A malicious client may be able to pollute Object.prototype in gateway directly by crafting operations with field aliases and/or variable names that target prototype-inheritable properties. Alternatively, if a subgraph were to be compromised by a malicious actor, they may be able to pollute Object.prototype in gateway by crafting JSON response payloads that target prototype-inheritable properties. This vulnerability is fixed in 2.9.6, 2.10.5, 2.11.6, 2.12.3, and 2.13.2.

From the GitHub advisory

Impact

A vulnerability exists in query plan execution within the gateway that may allow pollution of Object.prototype in certain scenarios. A malicious client may be able to pollute Object.prototype in gateway directly by crafting operations with field aliases and/or variable names that target prototype-inheritable properties. Alternatively, if a subgraph were to be compromised by a malicious actor, they may be able to pollute Object.prototype in gateway by crafting JSON response payloads that target prototype-inheritable properties.

Because Object.prototype is shared across the Node.js process, successful exploitation can affect subsequent requests to the gateway instance. This may result in unexpected application behavior, privilege escalation, data integrity issues, or other security impact depending on how polluted properties are subsequently consumed by the application or its dependencies. As of the date of this advisory, Apollo is not aware of any reported exploitation of this vulnerability.

Patches Mitigations addressing prototype pollution exposure have been applied in @apollo/federation-internals, @apollo/gateway, and @apollo/query-planner versions 2.9.6, 2.10.5, 2.11.6, 2.12.3, and 2.13.2. Users are encouraged to upgrade to these versions or later at their earliest convenience.

Workarounds A fully effective workaround is not available without a code change. As an interim measure, users who are unable to upgrade immediately may consider placing an input validation layer in front of the gateway to filter operations containing GraphQL names matching known Object.prototype pollution patterns (e.g., proto, constructor, prototype). Users should also ensure that subgraphs in their federated graph originate from trusted sources.

Weakness type: CWE-1321.

Product: @apollo/federation-internals (npm) · Assigned by: GitHub_M · Published: 2026-03-13

Affected and fixed versions

Severity scores

• Attack vector: network (remotely reachable)
• Attack complexity: low
• Privileges required: low (any account)
• User interaction: none
• Scope: changed (can affect other components)

CVSS measures how bad a flaw is if exploited, not how likely exploitation is. Read it next to the exploitation data below.

Exploitation and urgency

• CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
• CISA SSVC assessment: exploitation none, automatable no, technical impact total.
• FIRST EPSS (model of 2026-10-21): 0.62% probability of exploitation activity in the next 30 days; percentile 49.

Timeline

• 2026-03-13 — CVE record published by GitHub_M
• 2026-03-13 — GitHub-reviewed advisory GHSA-pfjj-6f4p-rvmh published (critical)
• 2026-03-16 — CVE record last updated

Sources and references

Original sources first. Version and fix details should always be confirmed against the vendor or project advisory.

Vulnerability records

1. GitHub Advisory GHSA-pfjj-6f4p-rvmh github.com
2. CVE.org record CVE-2026-32621 cve.org
3. NVD entry CVE-2026-32621 nvd.nist.gov

Other references from the CVE record and advisory

1. github.com github.com

Critical npm advisories · Today’s Security Digest