[GH-ISSUE #3758] rea doctor has no checks for the firmware integrations #1464

New issue

Open

opened 2026-10-16 21:05:41 +00:00 by bsakaguchi · 1 comment

bsakaguchi commented

2026-10-16 21:05:41 +00:00

Owner

Originally created by @oleander on GitHub (Oct 14, 2026).
Original GitHub issue: https://github.com/morluto/rea/issues/3758

Summary

rea doctor has no check for the firmware integrations. On a host where inspect_firmware_regions and extract_firmware run fine, doctor still says nothing about them. On a host where they cannot run, it gives no hint why.

Environment

rea-agents 6.5.0, linux-arm64 container, with binwalk 3.1.0, unblob 26.6.4, util-linux prlimit and procps installed, and REA_BINWALK_COMMAND, REA_UNBLOB_COMMAND and REA_FIRMWARE_PRLIMIT_COMMAND set.

Observed

rea doctor lists checks for node, host, hopper, ghidra, ida-registration, jadx, skill identity and registration. None of its output mentions binwalk, unblob, prlimit or firmware. It reports ok: false only because Hopper, Ghidra, IDA and JADX are absent, although both firmware commands work in the same container.

On macOS, both firmware tools refuse with "This integration requires Linux and an absolute REA_BINWALK_COMMAND path to the caller-supplied tool", and doctor does not flag that either.

Expected

A firmware section in doctor that checks:

the platform (Linux)
each REA_*_COMMAND variable is set to an absolute, executable path
the tool version against the verified one (binwalk 3.1.x, unblob 26.6.x)
prlimit, and ps from procps
unblob's external extractors, for example via unblob --show-external-dependencies

Suggested priority: P3: low (I cannot set labels on this repo).

Originally created by @oleander on GitHub (Oct 14, 2026). Original GitHub issue: https://github.com/morluto/rea/issues/3758 ## Summary `rea doctor` has no check for the firmware integrations. On a host where `inspect_firmware_regions` and `extract_firmware` run fine, doctor still says nothing about them. On a host where they cannot run, it gives no hint why. ## Environment rea-agents 6.5.0, linux-arm64 container, with binwalk 3.1.0, unblob 26.6.4, util-linux prlimit and procps installed, and `REA_BINWALK_COMMAND`, `REA_UNBLOB_COMMAND` and `REA_FIRMWARE_PRLIMIT_COMMAND` set. ## Observed `rea doctor` lists checks for node, host, hopper, ghidra, ida-registration, jadx, skill identity and registration. None of its output mentions binwalk, unblob, prlimit or firmware. It reports `ok: false` only because Hopper, Ghidra, IDA and JADX are absent, although both firmware commands work in the same container. On macOS, both firmware tools refuse with "This integration requires Linux and an absolute REA_BINWALK_COMMAND path to the caller-supplied tool", and doctor does not flag that either. ## Expected A firmware section in doctor that checks: - the platform (Linux) - each `REA_*_COMMAND` variable is set to an absolute, executable path - the tool version against the verified one (binwalk 3.1.x, unblob 26.6.x) - prlimit, and `ps` from procps - unblob's external extractors, for example via `unblob --show-external-dependencies` Suggested priority: P3: low (I cannot set labels on this repo).

bsakaguchi commented

2026-10-16 21:05:42 +00:00

Author

Owner

@XinwangTao commented on GitHub (Oct 15, 2026):

Confirmed — this is a real diagnosability gap: doctor reports tool presence for the RE frameworks but stays silent about firmware integrations, so a host that looks "ok: false due to missing Hopper/Ghidra" can actually have fully working firmware tooling, and a macOS host gives no hint why the firmware commands refuse. Adding a firmware section that checks platform, absolute executable REA_*_COMMAND paths, tool versions against verified ones (binwalk 3.1.x, unblob 26.6.x), and procps dependencies is the right scope. Two suggestions: separate hard failures (wrong platform, unset command) from soft warnings (version drift) so doctor's ok flag isn't skewed, and on non-Linux platforms report the firmware section as explicitly unsupported rather than omitted, so users don't read absence as "fine".

@XinwangTao commented on GitHub (Oct 15, 2026): Confirmed — this is a real diagnosability gap: doctor reports tool presence for the RE frameworks but stays silent about firmware integrations, so a host that looks "ok: false due to missing Hopper/Ghidra" can actually have fully working firmware tooling, and a macOS host gives no hint why the firmware commands refuse. Adding a firmware section that checks platform, absolute executable REA_*_COMMAND paths, tool versions against verified ones (binwalk 3.1.x, unblob 26.6.x), and procps dependencies is the right scope. Two suggestions: separate hard failures (wrong platform, unset command) from soft warnings (version drift) so doctor's ok flag isn't skewed, and on non-Linux platforms report the firmware section as explicitly unsupported rather than omitted, so users don't read absence as "fine".

bsakaguchi referenced this issue

2026-10-16 21:06:05 +00:00

[GH-ISSUE #3868] [Bug] Process capture rendered lines gain an extra blank cell after wide characters #1488

bsakaguchi referenced this issue

2026-10-17 21:11:37 +00:00

[PR #1465] [MERGED] docs(process): clarify normalized terminal line representation #2904

bsakaguchi referenced this issue

2026-10-31 00:50:26 +00:00

[GH-ISSUE #7582] [Bug] One-column process captures silently use a two-column rendered screen #4516