Updating README

Fixing hooks in CC and Cursor, Enforce AI scan phase 5-6, add support for Rules in CC and Cursor, tests updates

Add missing tainted_args field to EntryPoint dataclass to fix generate_entrypoints_json crash

Remove fragile decompiled regex from struct_scanner, delete exploitability-assessment skill, consolidate hunt-plan from 8 to 4 modes, fix hook code quality issues, add feature requests, and clean stale doc references

Remove legacy programmatic scoring: delete exploitability formula, dead taint scoring, naive taint-from-type inference, and phantom docs. Centralize severity mapping to ordinal sort keys. LLM is sole scoring authority

Add /compare-scans command and report_comparison helper. Add AI scanner orchestration rules. Update agents, commands, and docs

Replace heuristic vulnerability scanners with AI-driven agents; prune obsolete skills, commands, and helpers

Add unified IPC workspace discovery, --workspace flag for COM/RPC/WinRT resolvers, and fix DB query thread safety

Exclude auto-generated C# RPC stubs from language stats via .gitattributes

Add agent-tool guardrails and json_extract helper; forbid >/dev/null, 2>&1, and inline python

Rename /verify to /verify-decompiler, add /verify-finding command, and harden logic-vulnerability scanner patterns

Updating testcase to match sample extracted repo

Updating function index support

Updating docs

Fixing unit and integration tests