Each change to an AI's tool list runs as its own job, and jobs can finish in any order, so two quick changes could leave Settings showing the first while the second was saved. The row now waits for its save, as the request-rate row does. Also escapes the `|` characters that split the STATUS table row.
* Take target file signatures from the shared craft-sniff table
* Pin craft-sniff to the craft-libs #1841 merge and record the cutover
The shared table is on main at 070005125f0af1a5ae4a59002270e5849505bfc0.
* Keep named-but-unsupported formats as Unknown
A CPIO file starting 070701 was identified by craft-sniff, then fell
through to the unsigned path and was called text. Identified kinds this
app does not inspect stay Unknown; the unsigned fallback is only for
files with no signature.
A Ghidra folder setting that named a file was run as the headless launcher whatever it was. A start script for Ghidra's window (as Linux packages put on PATH) was listed as found, failed as a launcher, and Deep mode quietly used the built-in decompiler. A file is now run only when it is the headless launcher or a script that runs it; a link or start script is followed to its Ghidra folder; any other file isn't listed as Ghidra.
Follow-up to #1941. Where a program is about to start or be checked (Play, making a pin, the build's run and screenshot checks), Cargo itself is asked for its target directory, which also covers CARGO_BUILD_TARGET_DIR, the Cargo home's config and a parent workspace's target folder. The Project screen keeps the file lookup, so it starts no process; without Cargo the result is the file lookup. Outside the project only a program the manifests name is taken.
An API backend's `tools` setting (auto, lean or full) could only be changed by editing config.toml. Settings → AI now has a "Builder's tools" choice per API backend, with a caption saying what Auto works out to for that AI, and `ai tools <name> [auto|lean|full]` shows or sets it from the CLI. Auto is saved as no setting; agent apps are refused, as they choose their own tools.
Docs only: the guided install was tried on a clean Debian 13 and a Debian 13 desktop, followed by a Deep-mode decompile. macOS stays open, and a new roadmap line records that a Ghidra folder setting naming a file is listed as found while Deep mode quietly uses the built-in engine.
A build on an API model refuses every tool that needs consent outside the Build tools without asking anyone, and the model was told "The user said no", so it made up a reason (#1938). The model is now told that nobody was asked and that a plugin's `consent` setting decides; the journal records the case as `refused`. Whether such a build should ask the person is left as a decision line in the roadmap.
The file lookup from #1941 stays for frequent reads (the Project screen). Where a
program is about to start or be checked (Play, pins, the build's run and
screenshot checks), launch_now also runs cargo metadata from the project and
looks in its target_directory first, which covers what the files don't:
$CARGO_HOME/config.toml, CARGO_BUILD_TARGET_DIR and a parent workspace's
target/. Outside the project, only a program the manifests name is taken.
The guided install (Ghidra 12.1.4 and its Java, through engines_install) on a
clean Debian 13 and a Debian 13 desktop, then a Deep-mode decompile from Ghidra.
Ghidra laid out as Arch-based packages install it is found (#85). New open line:
a Ghidra folder setting naming the /usr/bin/ghidra start script is listed as
found, but Deep mode quietly uses the built-in lifter.
A shared or ancestor target directory outside app now supplies only a program the manifests name. Without a name, only target directories inside app are searched, so another project's executable isn't started. built_program_in(app, env) makes the tests independent of the process's CARGO_TARGET_DIR.
The ancestor walk starts from a normalized absolute app, so a relative app reaches a config above the working directory; normalized paths also keep .. out of selected programs. Cargo's legacy config wins over config.toml when it exists, and a config without the key or with invalid TOML doesn't fall through.
A pin copies a program built outside app into pin-program/ and records that relative path; other executables in a shared folder are not copied, and the frozen pin keeps working when the external build is removed.
A project whose Cargo settings move the build output (build.target-dir in .cargo/config.toml, or CARGO_TARGET_DIR) built, but Play and the screenshot/self-test checks only looked in app/target, so they could not start the program. built_program now looks under CARGO_TARGET_DIR, then the closest project or ancestor .cargo setting, then the default target/ as a fallback, and starts the first target directory that has the named program.
The new AI regression uses a project with target-dir = "out" and fails before the fix; engine tests cover the environment path (relative and absolute), a project setting, an ancestor setting, the environment coming first, and the default fallback.
The process and UI source-policy gates stopped at a file's first test attribute, so production code below an interleaved test module escaped them. They now read Rust paths through the whole file, follow import aliases, ignore comments and string examples, and exempt only items that can't exist outside tests. The lock gate uses the same scope, and module files named by a path attribute are checked wherever they live.
A failed API-backend turn wrote the error's message to its log, and that message repeats what the service, or a proxy in front of it, answered, which can include the key or words from the request. The log now keeps our own code for the error, the HTTP status, the context window and request size, and the service's error code only when it is one of a known list. The error shown in the app and the CLI still has the service's message.
write_failure_log wrote the error's message, which repeats what the
service or a proxy in front of it answered (up to 1,000 characters of an
HTTP error, all of a stream's error message), so a key or words from the
request repeated there were kept in logs/. The log now keeps Craft
Anything's code for the error, the HTTP status, the context window and
request size, and the service's error code only when it is one of
failure::KNOWN_PROVIDER_ERRORS, written from that list.
The ticked line covered only what ran: the window on a clean Ubuntu 24.04 and
the snake rebuild on a desktop. The website rebuild examples, a rebuild on a
clean machine, a real screen and Wayland get their own open line.
On a clean Ubuntu 24.04 under X11 the app stopped before its window:
"Library libxkbcommon-x11.so could not be loaded". With libxkbcommon-x11-0
installed it opened on its welcome screen. The README's snake-game rebuild
reached Ready with Claude Code. Also notes, as an open line, that the first
build check didn't find the program when Cargo's target-dir pointed elsewhere.
* Let Claude Code use the model its own settings name by default
Claude Code runs are no longer passed a hardcoded --model: with no choice
saved, the agent uses whatever its own settings name (a custom model, an
alias, opusplan…), and `ai model --default` clears a saved override back
to that. The Settings → AI picker lists "Default (its own settings)" first
for Claude Code (saved as `claude-code` or `claude`), choosing it clears
the saved model, and the typed "Another model" field is gone since the
list carries the choice. effort_levels detects Claude Code by adapter id
so the `claude` agent id gets every effort level too.
* Translate "Default (its own settings)" in the remaining eight catalogs
The i18n gate requires every catalog to carry the same key set; the
first commit only added the new picker label to six of them.
Line numbers in JavaScript facts counted only LF, so anything after a bare CR, U+2028 or U+2029 was reported on the previous line. The line index now counts every ECMAScript line terminator, and CRLF once.
A turn refused by a used-up Claude plan limit ends with a closing result error, which is the one the run keeps, and it carried the generic advice, so "which limit, and when it resets" was lost. The closing error now carries the same plan-limit advice as the error before it. Follow-up to #1931.
When a Claude plan's 5-hour or weekly limit was used up, Claude Code's rate_limit error was read as a per-minute rate limit, so the build waited and tried again and failed the same way each time. Claude Code's stream reports each turn's plan limits; when that report says the turn was rejected, the failure is now a usage limit, which isn't retried, and the fix names which limit is used up and roughly when it resets. A rate limit on a turn the plan allowed is still waited for and retried. Fixes#1919.
Engine downloads (Ghidra and its JDK, Godot, pinned installs) used only the proxy environment variables, so behind a VPN that sets only the Windows system proxy they went direct and timed out. When no proxy variable is set, they now use the person's manual proxy from Windows' settings, read from the registry once per download; this computer and Windows' own exceptions go direct, and a NO_PROXY the person set replaces those exceptions. Settings with credentials, SOCKS only or no port are ignored. AI clients are unchanged. Not run behind a real system proxy. Fixes#1827, #1787.
A walkthrough with keep_bodies.response_body kept bodies for the first
200 requests and said nothing about the rest, so a request without a
response_body read as one that had no body (#1811; web_page_show's
outlines were fixed in #1845). The network result now counts the kept
bodies, and a response_bodies_partial limitation says how many were
left out: past the cap, after the run's deadline, or unreadable.
Co-Authored-By: htjulia <htjulia1@gmail.com>
A local MCP plugin that exited before answering only said "The X plugin exited." It now gives its exit code and the last lines it printed, and a write to a server that has already gone gives the same error. When a local plugin can't be started, exits, or doesn't answer in time, a failure log is written to logs/ in the config folder and named in the error. The plugin's environment is left out of the log, and a remote plugin or a stop writes none. The failure-log writer moves to the engine crate so plugins and agent runs share it.
Aider's setup check read this computer's environment and home folder even for a run on another computer over SSH, so a working remote setup could be refused and local keys could pass a remote run that would then wait at a sign-in page. The check now runs on the host: one bounded, stoppable ssh command looks at that account's environment and Aider's settings files and returns only an exit status. Only the two definite answers refuse the run (no model or key; an OpenRouter model without its key); a check that can't run lets the run start and logs why. Not run against a real SSH host.