SSRF to the metadata endpoint steals IAM credentials. Those credentials may allow the attacker to access all S3 data, invoke Lambda functions, spin up EC2 instances, or access RDS databases — full cloud account compromise. Capital One impact: 100M records.

SSRF allows the attacker to port-scan the internal network and probe internal services that assume all requests come from trusted sources. Admin panels, monitoring tools, and internal APIs often have no authentication on internal interfaces.

file:// SSRF on a misconfigured server reads configuration files, private keys, database credentials, and environment files from the server’s local filesystem. Anything the application process can read, the attacker can read.

The API server is inside the firewall. SSRF uses it as a pivot point to reach internal services that the firewall would block if the attacker tried to reach them directly. The attacker reaches internal resources through the API server as an intermediary.

Some internal services accessible via SSRF accept commands (Redis EVAL, Elasticsearch /_scripts/, Jenkins API). SSRF that reaches these services with write access can result in RCE on internal systems.