Bright Security launches STAR Fuzzer with agentic fuzzing

Bright Security has released STAR Fuzzer, an agentic fuzzer that picks security-critical functions, writes harnesses and confirms findings with byte-exact payloads. Its first open source run found High or Critical issues in 45% of tested projects, including Crystal, htop and RIOT OS.

• AI agents choose targets and build harnesses; an evolutionary engine runs the search
• High or Critical issues found in 45% of tested open source projects
• Ten published findings across eight projects in the first month
• Every finding ships with a payload that reproduces the crash