Description

A low-privileged user may poison cached artifact metadata under specific conditions, potentially causing consumers to retrieve untrusted content.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "8.8",
  "pubDate": "2026-08-12T18:18:11.273Z",
  "pubdate": "2026-08-12T18:18:11.273Z",
  "executiveSummary": "An artifact metadata cache poisoning vulnerability exists where a low-privileged user can manipulate cached artifact metadata under specific conditions. This security flaw introduces the risk of downstream consumers retrieving and utilizing untrusted, maliciously altered content instead of legitimate build or dependency artifacts. The primary impact involves potential supply chain compromise, integrity violation of cached components, and unauthorized modification of artifact delivery pipelines. The vulnerability requires a low-privileged user posture and specific operational conditions to successfully execute the poisoning action against the caching mechanism. Risk implications are severe for environments relying on shared caching layers, as compromised metadata can propagate malicious payloads across dependent systems and deployment pipelines without immediate detection by standard integrity checks.",
  "technicalDetails": "The root cause of the vulnerability stems from insufficient access controls and validation mechanisms governing the write operations for cached artifact metadata. Specifically, the caching layer fails to adequately cryptographically verify or restrict write permissions for the entity updating metadata entries, allowing an actor with low-level privileges to overwrite or inject malicious entries into the cache store.\nThe exploitation method involves an attacker leveraging their low-privileged access to interact with the artifact storage or retrieval pipeline. Under specific conditions, the attacker crafts malicious metadata pointing to untrusted content and pushes or poisons the cache entry associated with targeted artifacts. When downstream consumers or build processes request the affected artifact, the caching mechanism serves the poisoned metadata rather than fetching the authentic source. Consequently, the consumer retrieves the untrusted content referenced by the manipulated metadata, leading to downstream execution or integration of unauthorized artifacts.\nThe attack flow proceeds as follows: first, the low-privileged user identifies the caching mechanism and the specific conditions required to submit or update metadata records. Second, the attacker generates a malicious metadata payload designed to redirect artifact resolution paths to untrusted sources or payloads. Third, the attacker injects this poisoned metadata into the shared cache. Finally, when an unsuspecting consumer requests the artifact, the system resolves the request using the cached malicious metadata, resulting in the delivery and potential execution of untrusted content.\nThe vulnerable component is the metadata caching subsystem responsible for indexing, storing, and serving artifact reference data. Exploitation requires low-privileged user access and the presence of specific configuration or operational conditions that permit unvalidated or weakly validated cache updates. Network exposure depends on the architecture of the artifact repository or caching service, which may be accessible internally or externally depending on deployment parameters. Post-exploitation impact includes supply chain contamination, unauthorized code execution in consumer environments, and persistent integrity degradation of the artifact distribution pipeline."
}