Developing AI features in custom actions - Port Swigger
Summary: PortSwigger published guidance on adding AI-powered features to Burp Suite custom actions through the Montoya API, using the Prompt interface to send system and user messages to the AI via ai().prompt().execute, which returns a PromptResponse read with content(). The article describes temperature as a value between 0 and 2, defaulting to 0.5, with lower settings of 0.0 to 0.8 producing more predictable output for technical or factual tasks and higher settings of 0.8 to 2.0 producing more creative output.
Key Takeaway: The article states that lower temperatures generally produce better results for security-related applications, so a default of 0.5 sits in that range - a setting a security team can verify rather than assume when AI is wired into vulnerability analysis and payload generation inside Burp Scanner.
Read the original at PortSwigger Research ↗
- Topics
- Vulnerabilities & Patching
- Related
- Attackers Exploit UnrealIRCd Flaw That Enables Server-Level Command Injection Cyber Risks · Nov 14, 2026U.S. CISA adds Check Point and UnrealIRCd flaws to its Known Exploited Vulnerabilities catalog Cyber Risks · Nov 14, 2026PostgreSQL Patches 52 Flaws, Including 22-Year-Old Bug Enabling Authenticated RCE Cyber Risks · Nov 13, 2026