Kjell Tore Guttormsen 32199965f2 fix(llm-security): post-mcp-verify reads live PostToolUse tool_response field ... The PostToolUse hook read input.tool_output, but live Claude Code delivers the tool result as tool_response — so the indirect-injection scan on MCP tool output silently never fired outside the test harness (which sent tool_output). The hook now reads tool_response with a tool_output fallback for older harnesses/fixtures. +3 tests covering string and object tool_response and the precedence. Found via a live-session check during the v7.8.3 sweep; not one of the 52 MEDIUM-tier findings. Suite 2016/0. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TcQyMTQfyrsAapaCMPxTtQ		2026-07-18 11:00:34 +02:00
..
e2e	test(llm-security): cover TRG/SIG/AST in e2e pipeline + SIG miner/hacktool/rot13 families (#58,#59)	2026-07-18 10:46:19 +02:00
fixtures	fix(llm-security): YAML/workflow parser divergence — block scalars + bare if: (#32,#33,#43)	2026-07-18 10:35:56 +02:00
helpers	test(llm-security): add JetBrains fixture tree + build helper	2026-04-18 10:49:49 +02:00
hooks	fix(llm-security): post-mcp-verify reads live PostToolUse tool_response field	2026-07-18 11:00:34 +02:00
lib	fix(llm-security): misc scanner correctness — reflog FP, diff exact-pass, hex dedupe, SARIF version (#20,#22,#50,#54,#56)	2026-07-18 10:36:51 +02:00
scanners	test(llm-security): cover TRG/SIG/AST in e2e pipeline + SIG miner/hacktool/rot13 families (#58,#59)	2026-07-18 10:46:19 +02:00