CVE-2026-47707: Strawberry GraphQL Fragment Spread Alias Bypass DoS

Technical summary

The MaxAliasesLimiter extension in Strawberry GraphQL implements alias counting at the Abstract Syntax Tree (AST) level to prevent query amplification attacks. However, it does not track the multiplicative expansion of aliases that occurs when FragmentSpreadNode references are resolved and executed. When a fragment containing multiple aliases is spread (reused) multiple times across a query, the actual number of alias resolutions at runtime exceeds the statically counted value by the number of spreads. An attacker can exploit this discrepancy by structuring a query with nested or repeated fragment spreads to cause exponential or linear alias multiplication, bypassing the configured limit and forcing excessive resolver execution. CWE-400 (Uncontrolled Resource Consumption) classifies this issue.

Business impact

For organizations operating Strawberry GraphQL endpoints, successful exploitation results in denial of service without requiring authentication or special privileges. Attackers can make legitimate services unavailable by sending specially crafted queries that force excessive computational work. This impacts API availability and user experience, particularly for public or semi-public GraphQL APIs. The low authentication barrier (none required) and network accessibility make this a practical threat in production environments. Organizations relying on alias limits for query cost control or rate-limiting strategies will find those protections circumvented.

Affected systems

All deployments of Strawberry GraphQL versions 0.172.0 through 0.315.6 are vulnerable. Organizations using earlier versions (prior to 0.172.0) or version 0.315.7 and later are not affected. The vulnerability is present in the core library; exposure depends on whether the MaxAliasesLimiter extension is actively configured and enabled in the GraphQL schema.

Exploitability

Exploitability is high. No authentication, user interaction, or special network positioning is required; an attacker needs only network access to the GraphQL endpoint. Crafting a malicious query requires understanding of GraphQL fragment spreads and alias mechanics, but this knowledge is well within the capability of competent attackers. Public documentation on GraphQL query optimization makes the attack pattern discoverable. The CVSS vector (AV:N/AC:L/PR:N/UI:N) reflects this accessibility. Real-world exploitation likelihood is elevated given the specificity of the flaw and the availability of GraphQL debugging tools.

Remediation

Upgrade Strawberry GraphQL to version 0.315.7 or later immediately. This version includes a corrected MaxAliasesLimiter that properly accounts for fragment spread expansion. Test the upgraded library in a staging environment to ensure compatibility with your GraphQL schema and resolver implementations before production deployment. If immediate upgrading is not feasible, consider disabling the MaxAliasesLimiter temporarily and implementing alternative query cost or complexity controls at the application or network layer until patching is completed.

Patch guidance

Verify your current Strawberry GraphQL version by checking your dependency lock file or running pip show strawberry-graphql in your Python environment. Update to version 0.315.7 or any later version by running pip install --upgrade strawberry-graphql and then restarting your GraphQL service. Strawberry's versioning follows semantic versioning; no breaking changes are expected between 0.315.6 and 0.315.7. Consult the official Strawberry changelog at https://github.com/strawberry-graphql/strawberry/releases for any migration notes specific to your deployment.

Detection guidance

Monitor GraphQL query logs for unusual patterns of alias usage, particularly queries containing multiple nested or repeated fragment spreads that expand to significantly more aliases than statically visible in the query text. If query monitoring is enabled, look for sudden spikes in alias count metrics relative to query complexity scores. Network-level detection is challenging but possible: inspect GraphQL request payloads for fragment spread patterns combined with high alias counts. Alerting on queries that exceed your configured alias limit (if logged separately) can identify attack attempts. Consider implementing query depth and complexity analysis that evaluates the runtime expansion of fragments to catch similar bypasses.

Why prioritize this

Though the CVSS score is moderate (5.3/MEDIUM), prioritization should be elevated due to the unauthenticated, network-accessible nature of the attack and its direct impact on availability. Organizations operating public GraphQL APIs or those for which availability is critical should treat this as a higher priority than the base score suggests. The fix is straightforward (a patch version upgrade), which supports rapid remediation. If your deployment does not use MaxAliasesLimiter or uses alternative query cost controls, urgency is lower but patching remains advisable.

Risk score, explained

The CVSS:3.1 score of 5.3 reflects low impact (availability only, no confidentiality or integrity breach), but the attack vector is network-based with no authentication or user interaction required. The score appropriately captures a denial-of-service scenario that is accessible but not catastrophic. However, practical risk to a specific organization depends on query cost strategy, the presence of downstream rate-limiting, and the criticality of the GraphQL endpoint. A public API with heavy reliance on alias limits warrants higher internal risk scoring than an internal service with defense-in-depth controls.

Frequently asked questions