Technical impact
In Guard Core, the whitespace‑collapse step in the content preprocessor removes carriage returns and newlines before pattern matching. This action disables the http‑splitting and newline‑anchored command injection detectors that rely on CRLF characters, allowing malicious input containing CRLF sequences to bypass Guard Core’s built‑in protection. The weakness is a logic flaw that leads to missing detection of CRLF injection attempts (CWE‑693).
Affected Systems
The affected product is Guard Core (rennf93:guard-core), a security middleware for FastAPI‑Guard, FlaskAPI Guard, and DjAPI Guard adapters. Versions from 1.0.0 up through 3.13.0 include the vulnerable preprocessor logic. All deployments using the default enhanced detection mode during this version range are affected.
Risk and Exploitability
The CVSS score is 3.7, indicating a moderate severity. EPSS is not available and the vulnerability is not listed in CISA KEV, so there is no current evidence of exploitation in the wild. An attacker would need to supply input containing CRLF characters to a Guard Core protected endpoint; the preprocessor would strip those characters, allowing the downstream application to receive the raw CRLF sequence. If the downstream stack echoes or executes that input in a CRLF‑tolerant context such as HTTP headers or command lines, the attacker could achieve HTTP response splitting or command injection. This scenario requires that the application persist or re‑emit the value, which is a common, but not guaranteed, condition. The attack vector is therefore application‑specific and requires an additional vulnerable sink downstream.