Commit Graph
10 Commits
Author SHA1 Message Date
morluto 339b82e0c8 fix(cutter): preserve bridge directory traversal 2026-10-27 08:58:27 +08:00
47353e494d feat(config): make the bridge request ceiling configurable (#6376)
* fix(reconstruction): dedupe residual unknown ids with uniqueSorted

Current main calls uniqueBy, which is not defined, so typecheck fails.
Residual unknown ids are strings, and the adjacent lists already use
uniqueSorted.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* feat(config): make the bridge request ceiling configurable

REA_BRIDGE_MAX_REQUEST_BYTES replaces the hard-coded 16 MiB Hopper and
Cutter request payload ceiling. The default stays 16777216 bytes. Values
outside 7464 to 268435456 bytes are configuration_invalid.

Part of #6176.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): keep cleanup ownership when a sent command's reply is oversized

An authenticated command may already have changed the Cutter document
when its reply exceeds the frame ceiling. Return the existing unknown
execution, with the ceiling diagnostic in the message, so receipt
cleanup still waits. A frame that overflows before the command is sent
remains a resource constraint.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* style: format the bridge request ceiling sources

oxfmt --check failed on the five files that set and test
REA_BRIDGE_MAX_REQUEST_BYTES. Formatting only; an oversized Cutter
reply still stays an unknown execution.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): exclude the LF from the response frame budget

REA_BRIDGE_MAX_REQUEST_BYTES counts payload bytes. The Cutter response
reader and authenticated encoder were reserving one byte for the
terminating LF, so an exact-limit response was rejected. A reply that
exceeds the ceiling after the command was sent still stays an unknown
execution.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(bridge): measure escaped JSON before allocating the request

A command of U+0001 characters can expand sixfold. JSON.stringify of
that expansion can throw past Node's string ceiling before the payload
check. Count the escaped size within REA_BRIDGE_MAX_REQUEST_BYTES
first and refuse it as a resource constraint.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(config): derive the bridge request maximum from JSON expansion

A 256 MiB ceiling still lets a shorter U+0001 command expand past Node's
536870888-character string limit. The maximum is now
floor((536870888 - 296) / 7) = 76695798, the longest such command whose
authenticated frame fits in one string. Requests are still measured
before JSON.stringify.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(bridge): size the Cutter frame deadline and send the measured Hopper payload

The authenticated Cutter command is encoded after the challenge, so a
one-second read rejected an in-budget frame. That read now allows one
second per mebibyte of the ceiling and never waits longer than 25
seconds. Hello stays at one second. Hopper writes the payload it already
measured instead of serializing the same request again.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(config): bound the bridge request maximum by the serialized string

The ceiling already measures escaped JSON, so dividing Node's string limit by the seven-byte expansion rejected encoded payloads that still fit. The maximum is 536870887, one byte under that string ceiling, leaving room for the framing LF.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(hopper): serialize a bridge request when it reaches the wire

The queue kept each encoded frame, plus the caller's params, until Hopper finished the active request. A raised ceiling made a few waiting calls enough to exhaust the heap. The FIFO now keeps the params, and the frame is encoded only for the request being written.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): hash the bridge proof prefix and payload separately

An exact-limit Cutter frame plus the HMAC prefix exceeds Node's single-string ceiling, so the client threw before sending an admitted command. The prefix and payload are now separate HMAC updates, including the response proof.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): keep recovery fields when a status path exceeds the frame

A long current_file made the status reply miss the ceiling, and the fallback then dropped session_id and the completion receipts. The status now omits the path and version and still returns the session, connection proof, and up to 64 receipts. That required recovery frame is smaller than the 7464 floor.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): report an unknown path when status identity does not fit

A bounded status kept the new document generation and omitted the path, so the client filled in the descriptor's older path. The observed path is sent when it fits. When it does not, current_file is null and is not replaced with that stale path.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): count structured replies by their encoded JSON size

The response guard charged 16 bytes per primitive and two bytes per list slot, so a dense array under the configured ceiling was replaced with output_truncated. Numbers, booleans, null, and container punctuation now use their JSON lengths.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): stream authenticated responses in bounded chunks

A fitting reply was signed from retained raw bytes, escaped bytes, and a
joined frame, so a ceiling near 512 MiB could exhaust Cutter before send.
Measure and sign JSON pieces, then write the same frame in 64 KiB chunks.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(bridge): keep Cutter admission from stalling or hiding sessions

Read an unauthenticated frame off the accept loop. A command reply that
does not fit keeps completion and generation and reports an oversized path
as null. Discovery reports a ceiling mismatch instead of an absent session.
The maximum HMAC proof is opt-in, and both bridges reject non-ASCII digits.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): bound frame scans and client response copies

Search only newly read bytes for the framing LF. Unescape each socket
chunk into one response string, check the proof, and then parse it.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): keep completion, admission, and dense replies bounded

Report completed_connection only through connections that have
terminated. Cap connection threads and keep the listener up if one
fails to start. Batch scalar JSON and fail discovery when any ceiling
mismatches.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): hash requests in parts and scan escapes once

Update the client-proof prefix and request separately. Stop the response
decoder at the first quote or backslash instead of rescanning the tail.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(cutter): confirm live ceiling mismatches and omitted output

A stale descriptor from a previous ceiling is not a discovery mismatch
until its hello proves a live authenticated bridge. A response discarded
for the frame ceiling after dispatch reports output_truncated.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* test(cutter): close the unauthenticated ceiling-mismatch fixture

The listener that fails the hello proof is destroyed after its challenge
so discovery cleanup does not wait on that socket.

Co-authored-by: morluto <morluto@users.noreply.github.com>

---------

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: morluto <morluto@users.noreply.github.com>
2026-10-23 06:11:21 +08:00
morluto e780148305 fix(cutter): preserve producer failure and cancellation facts 2026-10-22 22:07:11 +08:00
morluto 71929ed37c fix(cutter): normalize command checkpoint cancellation
Produce typed cancellation at each command checkpoint while retaining the exact opaque reason as cause. Preserve independent typed physical failures without downstream equality guesses.

Validated 72 focused tests across nine files, static checks, docs and five conformance controls, including real client/service admission races and matching physical-error controls. No native Cutter engine or new public-transport reproduction claimed.

Refs: #4635
2026-10-17 09:24:43 +08:00
morluto 5c0472c571 fix(cutter): attribute status proof failures to command
Parse command status and validate session identity at one bridge boundary. Keep the preflight watermark requirement separate from exact completion receipt proof, and retain status as a compact auxiliary diagnostic.

Reuse authenticated protocol and descriptor fixtures. Validate 66 focused tests across eight files, static checks, docs and five conformance controls. CLI refusal controls use an authenticated loopback fixture; native Cutter was not verified.

Refs: #4607
2026-10-17 09:19:02 +08:00
morluto 2b7b698db8 fix(cutter): retain descriptor owners and join discovery cancellation
Refs #4471, #3946.

Read only admitted regular descriptors without blocking on pipes. Retain failed closes, join discovery workers, and propagate cancellation through CLI and MCP. Preserve command and descriptor cleanup as independent owners.
2026-10-17 00:44:02 +08:00
Ysopwnable a6c6a76d37 fix(cutter): cancel discovery probes and queued work
Propagate CLI/MCP cancellation through shared Cutter discovery, stop queued probes, and drain admitted workers. Deliver terminal bridge results only after socket closure. Preserve confirmed command Evidence and caller reasons when cancellation arrives during finalization.

Validated 66 focused tests, full CI static/docs gates, actual CLI SIGINT, compiled stdio cancellation/recovery, and official Linux Cutter 2.5.0 Qt5 CLI/MCP ELF workflows. Native UI cancellation and descriptor filesystem cancellation remain unverified.

Fixes #3946.
2026-10-16 22:49:06 +08:00
morluto 7e0c7b1244 fix(cutter): resolve bridge paths from the selected caller home 2026-10-15 04:01:36 +08:00
morluto 45d5e80144 fix(cli): preserve Cutter choices and reject invalid runtime arguments 2026-10-15 03:43:29 +08:00
4aa51bfe19 fix(reverse): reconcile cleanup ownership and mutation diagnostics (#2799)
* feat(process): add owned debugger shutdown lifecycle

* feat(reverse): add local inspection commands

* feat(gdb): add persistent debugger integration

* feat(cutter): add Cutter bridge integration

* test(reverse): enforce complete integration coverage

* fix(process): preserve debugger shutdown ownership

* fix(reverse): retain evidence across interrupted commands

* fix(gdb): honor captured provider environment

* test(process): cover debugger shutdown retries

* fix(reverse): retain evidence for interrupted providers

* test(process): cover debugger shutdown retries

* fix(reverse): retain evidence for interrupted providers

* fix(server): snapshot provider environment at creation

* fix(reverse): close review gaps with complete coverage

* fix(reverse): address inspection review findings

* fix(gdb): bound sessions and retain cleanup ownership

* fix(ci): require reverse-tools quality lane

* fix(process): unref inherited debugger output pipes

* fix(reverse): address remaining PR 1960 findings

* fix(gdb): address remaining PR 1961 findings

* fix(reverse): account for unrestricted provider execution

* fix(process): retain debugger ownership until safe shutdown

* fix(process): preserve replacement debugger sessions

* fix(reverse): preserve admission and output provenance

* fix(process): retain exited debugger until output settles

* fix(mcp): bound provider admission and debugger cleanup

* fix(reverse): preserve failure provenance and classify E2BIG

* fix(gdb): preserve startup cleanup and detach safety

* fix(process): retain debugger ownership after detach failure

* fix(reverse): preserve mutation and failure provenance

* fix(gdb): address debugger review findings

* fix(reverse): rebind sessions after external mutation

* fix(process): cap debugger output after exit

* fix(gdb): protect one-shot detach verification

* feat(debugger): add reverse tool coverage gate

* test(debugger): cover integrated failure paths

* fix(gdb): bound and preserve session observations

* fix(gdb): address review regressions

* style: format merged integration files

* fix(reverse): preserve session and output evidence

* refactor(gdb): share repeated test and partial result setup

* fix(gdb): invalidate TTY proof for Python aliases

* fix(reverse): retain failure output byte counts

* test(process): synchronize late debugger output

* fix(gdb): invalidate TTY proof for wrapped commands

* fix(gdb): recognize wit command abbreviation

* test(reverse): wait for cancelled descendant readiness

* test(reverse): bound descendant cancellation cleanup

* style(reverse): format integration tests and adapters

* style: format reverse engineering integration files

* test(reverse): remove duplicate environment harness

* test(gdb): share session startup assertions

* fix(gdb): invalidate output proof for command abbreviations

* test(reverse): share composition edge case scenarios

* fix(ci): recognize verification and mutation tool entries

* test(rizin): cover successful fake inferior detach

* test(cli): align full output failure envelope

* test(cli): isolate reverse command lifecycle service

* test(process): share debugger output fixture setup

* fix(binary): preserve high-bit Windows signature bytes

* style(binary): format Windows signature parser

* test(mcp): preserve call coverage in unchanged comparison

* fix(ci): include reverse tools Vitest configs in lint cache

* fix(mcp): preserve target content on delivery errors

* ci: align duplication ceiling with expanded suite

* test(reverse): allow signal admission mutants

* fix(reverse): bound retained cleanup failures to process owners

* fix(test): replace drifting mutation ranges with owned selectors

* fix(reverse): admit persistent commands and retain cleanup ownership

* fix(cutter): recover timed-out commands from exact receipts

* fix(cutter): reject unsupported bridge request kinds

* fix(ci): repair client format and shared verification fixtures

* fix(native): preserve debugger cleanup ownership and bridge authentication

* test(native): retain diagnostics without perfect coverage requirements

* chore(licenses): refresh locked dependency inventory and notices

* fix(test): exclude transient command locks from mutation sandboxes

* fix(gdb): admit only platforms with safe detach verification

* fix(cutter): retain bounded completion receipts until acknowledgement

* fix(cutter): acknowledge commands rejected before UI dispatch

* test(reverse): fail cancellation checks before fixture teardown

* test(gdb): clean up the retained broken-stdin fixture owner

* docs(test): clarify focused provider verification scope

* fix(rizin): retain close cleanup failures

* fix(gdb): wait for inferior stop and bound MI buffering

* fix(cutter): retain failed receipt acknowledgements

* fix(rizin): return retained session identity on rebind failure

* fix(gdb): preserve startup cancellation and lossy status

* fix(rizin): retain launch provenance and assemble frames once

* fix(cutter): recover receipts across callers and bound discovery

* docs: describe debugger evidence and bridge recovery

* fix(process): keep cleanup incomplete while writers remain

Stopping a debugger no longer reports success when inherited output
streams stay open, so BinarySession cannot reopen a binary a child is
still changing. Concurrent Cutter cleanup retries share one receipt
acknowledgement instead of treating a consumed receipt as failure.

* fix(debuggers): recover cleanup with captured process identities

* fix(analysis): retire sessions after unrestricted commands

* fix(cli): preserve completed evidence after cleanup recovery

* fix(cutter): bound frame collection and coalesce cleanup

* fix(windows): resolve selected provider paths absolutely

* fix(gdb): report decoding loss in failed startup evidence

* fix(rizin): scope decoding loss to collected command frames

* test(mcp): cover snapshot retirement in reverse tools CI

---------

Co-authored-by: Daniel Gonçalves Araujo <47118053+dandgabr@users.noreply.github.com>
Co-authored-by: REA Maintainers <noreply@github.com>
Co-authored-by: OpenAI Codex <codex@users.noreply.github.com>
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
2026-10-14 04:47:34 +08:00