What problem does it solve? Large offensive security engagements fail when testers chase isolated bugs without a unifying objective, losing track of surfaces, hypotheses, and chainable primitives across a long campaign. ## Core Features & Use Cases - Campaign Start Sequence: Selects a crown-jewel objective, builds a target dossier, imports HAR/Burp/Caido traffic into a request map, and ranks attack surfaces before choosing bug classes. - Three Operating Modes: Supports crown-jewel route planning, narrow-target deep manual hunting via /deep-hunt, and broad delegated campaigns with specialist subagent briefs. - Chain-Oriented Scoring: Invokes /attack-chain on any primitive to score findings for chainability rather than standalone severity, producing report-ready exploit-chain cards. - Use Case: Starting a bug bounty engagement on a new web target, you run /crown-jewel target.com, import captured traffic with /request-import traffic.har, deep-hunt the billing workflow, and convert a refund logic flaw into a documented exploit chain. ## Quick Start Ask the agent to start an offensive campaign against your in-scope target by choosing the crown-jewel objective and building a target dossier.
