Initial public release: dockerized reverse-engineering workbench

ChipBench packages Ghidra, radare2, binwalk, chip-programming tools,
simulators, and firmware-unpacking utilities into one reproducible container
for analyzing raw chip dumps entirely from the command line or an AI CLI.
Headless Jython scripts drive import, forced-disassembly sweeps, live
queries, and bulk decompilation exports without any GUI.

Derived from a private engagement environment, generalized for public
release under AGPLv3. No engagement-specific artifacts are included.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2

docker/conf/profile.d/chipbench.sh

@@ -0,0 +1,13 @@
+# chipbench shell setup
+# Prepend the isolated Python venv so `python3` resolves to the interpreter
+# that has capstone / unicorn / mcp installed.
+case ":${PATH}:" in
+    *":/opt/venv/bin:"*) ;;
+    *) export PATH="/opt/venv/bin:${PATH}" ;;
+esac
+
+# Helpful banner once per login shell.
+if [ -n "${BASH_VERSION:-}" ] && [ -z "${CHIPREV_BANNER:-}" ]; then
+    export CHIPREV_BANNER=1
+    printf '\n  chipbench RE shell — try: re-identify, re-analyze, re-binwalk (host) or\n  r2 / ghidraRun / binwalk / python3 (here). Artifacts: /data/artifacts (ro)\n  Project: /data/work  Output: /data/output\n\n'
+fi