Changelog¶
All notable changes to lcode are documented here. The format follows Keep a Changelog and the project uses Semantic Versioning.
Unreleased¶
Added¶
- Baselines for the public benchmarks. The Public benchmarks page now gives lcode 0.63.0's results on the dev slice on a 12 GB GPU: qwen3.6-35b at 32K (55%) and 64K (61%), why 256K and qwen3.8-27b don't fit such a machine, and the commands to reproduce them.
0.63.1 - 2026-11-11¶
Fixed¶
lcode -c -pin an image or video session exits with1when nothing came. Whenlcode -c -p "…"(or--resume … -p) carried on an image or video session and the picture or video didn't come (no ComfyUI, a job that failed in ComfyUI, Ctrl+C), lcode said so and exited with0. It now exits with1, aslcode --mode image -pdoes.
0.63.0 - 2026-11-11¶
Changed¶
- A reply that thinks at length without acting is stopped. Sometimes the model weighs one option
after another for minutes without trying any: in
polyglot-rust-cit wrote 45,000 to 53,000 tokens, most of them thinking, and ran 1 or 2 commands in 15 minutes. Now, when a reply has thought for 7,000 tokens without writing anything or calling a tool, lcode stops it and asks for the next step with thinking off. It hands the model the last part of its thinking, so it can act on where its reasoning got to. On the dev slice, about 1 reply in 100 thought that long: they came from 42 of the 110 attempts that failed and 2 of the 102 that passed. Set the budget withlcode config set think_budget,0for no limit (#241).
Fixed¶
lcode -pexits with1when the request fails, also without--output. It showed the error (a model error, Ollama stopping in the middle of a reply, running out of memory) and exited with0, so inlcode -p "…" && next-stepthe next step ran after a failed request. It now exits with the codes--output jsonuses:1for an error,3stopped at--max-steps,4stopped because the model kept repeating itself,130interrupted. What it prints is the same, and the interactive session still carries on after an error.- The line after a load counts the part of the model in system RAM. With Ollama 0.40,
qwen3.6-35bat 32K on a 12 GB GPU showed as10.5 GB, 98% on the GPU and the rest in system RAM, because Ollama leaves out the ~12 GB of experts that stay mapped from the model's file. For the models in the catalog, lcode now counts the whole model at its estimate, the onelcode setupandlcode doctorshow:~23 GB, 42% on the GPU and the rest in system RAM. With Ollama's numbers, the warning that another program is using GPU memory couldn't fire. Now it does, and it says how much of the model fits on the GPU. lcode doctorshows the loaded model. It looked forlcode-qwen3.6-35bwhile Ollama lists it aslcode-qwen3.6-35b:latest, so theLoadedline never appeared for the modelslcode setupcreates. It now says how the model fits, in the same words as the line after a load.lcode benchcounts the part of the model in system RAM, too. Forqwen3.6-35bat 32K on a 12 GB GPU, it said10.4 GB, 98% on the GPUafter loading, and the same in its summary. It now says~23 GB, 42% on the GPU and the rest in system RAM, as the line after a load does.--jsonresults keep Ollama's count inmemory_gbandgpu_percent, so they still compare with older results, and add lcode's estimate inestimated_memory_gibandestimated_gpu_percent.- A passing run of one test file counted as the whole checks. When the model ran the project's
check command with a path or a test name after it (
.venv/bin/python -m pytest -q tests/test_calc.py), lcode took the checks as passed and didn't run them, and/statusandlcode -p --output jsonsaid they passed, even on a project whose full run stops before its first test. Now a run like that, or one with an option that picks tests (-k,--ignore,npm test -- --grep), doesn't count, and lcode runs the checks itself. Options alone, like-xor--tb=short, still count. In 426 attempts on six dev-slice nights this changes 2: one now runs xarray's suite (4.4 minutes, failing on tests that failed before), the other a run that stops before its first test in under a second. - A model server that stops in the middle of a reply no longer ends lcode in a traceback. When the
connection broke while a reply streamed, as it did when
lcode updatereplaced Ollama's container during a reply,lcodeandlcode -pstopped with a Python traceback, which also ended an interactive session. Now lcode says that Ollama stopped in the middle of a reply and what to do, the session carries on, the trace records the failure, andlcode -p --output jsonreportsstatuserror. LM Studio, llama-server, vLLM and MLX servers get the same. A reply stopped on purpose breaks its connection too, and that stays a stop: a cancel from an editor (ACP) came back as an internal error,ChunkedEncodingError, and a subagent stopped with Ctrl+C was reported as failed.
0.62.0 - 2026-11-06¶
Changed¶
- A test run that stops before its first test no longer counts as a failure of the model's change.
pytest stopping on an error while collecting, or unittest ending in a traceback instead of
Ran N tests, says nothing about the change, yet the model was told that the project's checks fail and to fix what its changes broke. On the dev slice at 32K that was 19 of the 34 automatic checks (Django'stests/can't be discovered by unittest; two other projects stop on modules and data that aren't there). The model spent 26 minutes after those notes, rangit stashin 5 attempts to see for itself, and in 2 attempts kept editing: one downloaded a 244 MB database and changed a test file the request had excluded. Now lcode tells you what stopped the run, and the model's answer stands. An error that names a file the request changed still goes to the model, shell commands included, and so do tests that ran and failed.checksinlcode -p --output json, the task sheet and/statusnow say these checks didn't run, also for a project with no tests or no test runner. The notes weren't all loss: in that same attempt, 1 of 47 like it over four nights, the extra work also fixed a test the model had called pre-existing. - Checks that don't finish in 10 minutes are no longer called a failure. The model is told that they don't say whether its change works, instead of to fix what it broke.
- A SPARQL query it never ran now counts like code it never ran (#242): on the dev slice, 2 of 3
sparql-universityattempts on 0.61.0 wrotesolution.sparqland said they were done without running it. lcode sends the model back once, and the note namesrdflib. SQL files stay out, because one can be a migration, and asking to run that could change a real database.
Fixed¶
- The automatic checks run in the project's root. They ran wherever the model's last
cdleft its shell, though their command is written for the root: fromtests/,python -m unittest discover -s testsdoesn't start and.venv/bin/pythonisn't there. 3 of the 34 checks on the dev slice ran fromtests/. A command from your owncheckssetting runs in the folder you started lcode in. - A failing test run piped into
tailcounted as passing. When the model ran the project's checks itself aspytest -q | tail, lcode readtail's exit code of 0, took the checks as passed and didn't run them, andlcode -p --output jsonreported them as passed. A run whose exit code isn't the check's now counts only if its output shows a test summary with nothing failing. Otherwise lcode runs the checks itself. lcode bench --comparetook a task with one attempt as certain, so its noise came out too small, and it said nothing when a run had far fewer attempts: a real run lost 18 of its 72 attempts and compared as usual. A single attempt's pass rate now comes from both runs, and--comparewarns when the two runs didn't have the same number of attempts.
0.61.0 - 2026-11-05¶
Added¶
- Code it never ran (#242). When a request wrote a new code file and no command since its last
change ran it, lcode sends the model back once, to run it or to say plainly that it's untested. A
syntax check, a program that isn't installed (
python3: command not found) or a command that only reads doesn't count. Withuvinstalled, the note says how to run Python code in a throwaway environment. On the dev slice at 32K, 5 of 6torch-tensor-parallelismattempts wrote the module, found nopython3or checked only its syntax, and said they were done: each had a runtime error that one run would have shown. A project with checks of its own gets those instead. - lcode says when the machine is short of RAM for the model (#246). A model that ran well at first
could fall to about a third of its speed, and nothing said why: the part of the model outside GPU
memory and Ollama's prompt cache, which grows to 8 GB in a long session, both have to fit in RAM, and
lcode only compared the model with the machine's total. On Linux, with Ollama on the same machine,
lcode doctorhas aRAMline that compares them with what's available now and, when they don't fit, says what to do: close programs, use a smaller context, or pick a smaller model.lcode setupgives the same warning on its recommendation, and a session says it once when replies fall below half of their earlier speed and RAM is short.
Changed¶
- After a reply that goes nowhere (it went round in circles, filled the context before it said
anything, or was empty), the model's next reply runs with thinking off (with Ollama), so it acts
instead of starting the same deliberation over. lcode now stops the request when two replies in a row go
nowhere, or four in one request, instead of the second one anywhere in it. On the dev slice at 32K,
16 of 150 attempts were stopped this way, and all of them failed: 9 because the second chance went
round in circles again (all 6
polyglot-rust-cattempts among them), and 7 that had recovered and worked on until another reply went nowhere much later (#241).
Fixed¶
- A service you ask lcode to leave running stays up after lcode exits (#240). lcode stopped every
background job when the session ended, and there was no way to keep one: on the dev slice, 5 of 6
pypi-serverattempts failed because the server the task asked for was gone when its tests ran. The model now starts such a service withkeep=true. It writes its output to a log in~/.local/state/lcode/jobs/instead of to lcode, so it keeps running, and lcode says how to stop it. It always asks first, and the question says that the command stays running. - A background command written with
&(server &) looked as if it had ended right away, while the program kept running where lcode couldn't see or stop it, and held its port. The&is now ignored, and the program is a background job like any other. What a background command otherwise leaves behind when it ends at once (server & echo started) is stopped, and the model hears why (#240). - The automatic checks failed on any project whose
tests/folder has no__init__.py, withStart directory is not importable, even when every test passed. The model was then asked to fix that: in two real runs, one added atests/__init__.pynobody asked for, and the other blamed the project. lcode now leaves out-t .unlesstests/is a package.
0.60.0 - 2026-11-04¶
Changed¶
- For contributors: CI type-checks
src/lcodewith basedpyright, and the eight modules that were over 1,000 lines are split into smaller ones, with no change in behavior. The code map indocs/how-it-works.mdlists them (#200).
Fixed¶
- The Valohai MCP server failed to describe an execution with an output that has no name; it now
shows it as
?. Found by the type checks that CI now runs onsrc/lcode(#200). - Cancelling in an editor right after sending a request (lcode acp) could be lost: the request then
ran to the end. The editor's messages are handled in threads of their own, and the request's own thread
reset the cancel when it started late, on a busy machine. The cancel now holds. This was also why
tests/test_acp.pyfailed now and then under load (#200). lcode bench --compareleft out attempts whose tests ran but didn't finish, as if they hadn't run, so a change that makes the tests hang could hide a drop. They count as failed now, as in Harbor's own mean, and--comparereads both runs again from Harbor's results so that they count alike. It also gives each run's model speed, and warns when one ran more than 10% slower, which makes the tasks' time limits bite more often. On 0.59.0's gate, both showed (#199).
0.59.0 - 2026-11-03¶
Added¶
lcode bench --compare BEFORE AFTER(#199) compares two--harborresults: the tasks whose results changed, and whether pass@1 moved by more than the noise, which it estimates from how much each task's attempts disagree (about 9 points on the dev slice). It exits with 1 after a drop beyond the noise. Releases that change prompts, tools or context handling now go through it on the dev slice: see Releases in the contributing guide.
Fixed¶
lcode bench'swide-featurecheck failed an attempt whosebudget listprinted a message such asNo budgets set.without budgets, or whose report listed the budgeted categories first. The task asks for neither behavior, so both pass now. With qwen3.6-35b, this had failed two attempts in three that were otherwise right.
0.58.0 - 2026-11-02¶
Added¶
- Six longer tasks in
lcode bench(#199), closer to real work, after the eight quick ones: a command-line tool built from a spec (cli-tool), a test that fails now and then (flaky-test), a bug reported only by its symptom in a 47-file repository (repo-bug), a feature across modules with older data files still working (wide-feature), a slow job made fast with the same output (fast-dedupe), and a bug that shows only when the code runs (runtime-bug, until now only with--tasks). Each check runs on data or tests the model never saw. With qwen3.6-35b at 32K on a 12 GB laptop GPU, 14 of 18 attempts passed: none atruntime-bug, and 2 of 3 atfast-dedupe, where after lcode made room in the context the model lost track and repeated itself until the loop breaker stopped it. lcode bench --quickruns only the eight quick tasks, once each, aslcode benchdid before 0.57.
Changed¶
- In
lcode bench, the longer tasks stop after 10 minutes and the quick ones after 5.--timeoutsets one limit for all.
0.57.0 - 2026-11-02¶
Changed¶
lcode benchmakes three attempts at each task (#199), each in a new folder and conversation, and reports pass@1, pass^3 and how many tasks passed in each attempt, counted as in the public benchmarks. One run of each task was within the noise between runs.--attempts 1runs each task once, as before, in a third of the time. In the summary, a task's time is per attempt, on average. The JSON addsattempts,summaryand each task'sattempt. Without--session,--roundsis another name for--attempts.
0.56.0 - 2026-11-02¶
Added¶
- Other agents on the same model (#198):
lcode bench --harbor dev --agent claude-code(oropencode,codex) runs that agent on the same tasks, model and context as lcode, for a comparison. Each is set up the wayollama launchsets it up, with approvals off. Their APIs can't set the context window, so lcode makes a copy of the model with it set (lcode-qwen3.6-35b-64k), which uses the model's files.lcode-bench.jsonrecords the agent and its version.
0.55.0 - 2026-11-02¶
Added¶
lcode bench --harborsays how long the whole run took, andlcode-bench.jsonsaves when it started and its duration (started,seconds).
Fixed¶
- Tokens of unfinished work in
lcode bench --harborand in traces. An attempt stopped at its task's time limit reported 0 tokens. So did a reply lcode stopped because it went round in circles, since the server counts a reply only when it ends. The bench now counts what the attempt's trace recorded, and the trace estimates a stopped reply's tokens (estimated_inandestimated_out;lcode trace showmarks them "about"). An attempt with no record shows–, not 0.
0.54.0 - 2026-11-01¶
Added¶
- Public benchmarks (#197):
lcode bench --harbor devruns lcode on a slice of Terminal-Bench 2.0 and SWE-bench Verified with Harbor, each task in a fresh container with your local model, three attempts each. It reports pass@1, pass^3, the spread between runs and the time and tokens per task, and saves them with the lcode version, the model, the context and the command that reproduces them. Theheld-outslice is for confirming a change, never for tuning on it. Forharbor runitself, the agent islcode.harbor:Lcode.
Fixed¶
- Your Python when lcode runs with
uvx.uvxputs lcode's own temporary environment first onPATH, sopythonin the model's commands and in the checks was lcode's Python, without the project's packages, and the checks failed with "No module named pytest". lcode now takes that environment offPATHwhen it starts, if you have a Python of your own onPATH. Other installs don't changePATHand are left as they are.
0.53.0 - 2026-10-31¶
Added¶
- Run traces (#196): each request in code mode is recorded on this machine as it runs: the
model's answers with the time to the first token, the total time and the tokens in and out;
each tool call with its result and time; the model's settings, the lcode version and the git
commit. Subagents get a lane of their own. Traces stay on this machine, are deleted after 30
days and take 200 MB at most;
lcode config set trace falseturns them off. lcode tracelists the traces,lcode trace show 1shows one as a timeline, andlcode trace export 1writes a copy for a bug report with secrets, home-folder paths, user names, and email and IP addresses masked.lcode trace replay 1runs a recorded request again with the recorded answers instead of a model, in a copy of the folder at the recorded commit, and shows where the tool results now differ. Shell commands keep their recorded results unless--commandsruns them again. It needs no GPU, andlcode.replay.rundoes the same in a test.
Fixed¶
- Screenshots of web pages when the browser starts slowly. The first start after a restart reads the browser from disk, which took up to 12 seconds on GitHub's machines and sometimes over the 30 seconds lcode gave it, so the shot failed with "the browser didn't answer". The browser now gets 90 seconds to start. An error says whether it didn't start in time, quit, or didn't answer a command, and quotes the browser's last lines.
- What to do when Ollama in Docker doesn't answer. lcode and
lcode doctorsuggested installing Ollama even when it runs in a Docker container. They now name the container and give the command to start it:docker compose upwhen Docker Compose made it,docker startotherwise. A running container that doesn't answer getsdocker logs. When there's no container or Docker itself isn't running, the message says so. - The Ollama commands on macOS. When Ollama wasn't running, or had a new version out, lcode showed the Linux install command on a Mac. It now shows the Homebrew one.
0.52.0 - 2026-10-27¶
Added¶
/resumein two sections (#141): This folder lists the conversations saved in this folder and anywhere in its git repository, also from a subfolder, a worktree or another clone (the same first commit). Elsewhere lists the 20 newest from other folders, with each one's folder. They're numbered across both. Before, only sessions saved in exactly this folder were listed, so starting lcode one folder up made them look lost.- The context each conversation uses, in the list:
18.2K / 32K (57%)against the current window, in yellow when resuming it means a summary first. Sessions now save what they used, and the status bar shows it right after resuming. Older sessions get an estimate, marked~. - Notes follow a repository that moved, was renamed or was cloned again. A repository's
notes folder names its first commit (
repo.json), and a checkout whose own folder has no notes finds them by it. Notes from older versions get the file the first time lcode runs there. lcode doctorshows how many sessions are saved, how many for this folder, and where.
Changed¶
- No memory check when a session ends, by default. Quitting,
/clearand summarizing no longer wait for a model call asking what's worth remembering. Turn it back on withlcode config set memory_check true./remember, the model's memory tool and saved notes work as before, and the conversation is always saved. lcode -ccontinues the latest conversation in this folder, else the latest in its repository, and says when that takes it to another folder./resumeandlcode --resumeleave out sessions from temporary folders that no longer exist, such as scripted test runs;/resume alllists them, and everything else.
Fixed¶
- A session file that isn't a conversation (broken JSON, a list, odd messages) no longer stops
/resumefrom listing the others.
0.51.0 - 2026-10-27¶
Added¶
- Clips up to 60 seconds (#185), also from a start to an end picture:
@start.png @end.png the flower opens --seconds 25. A clip longer than the 8 s a video model makes at once is made in parts, each starting on the last frame of the one before (the last one ending on the end picture), and joined into one clip with ffmpeg, without the frame each join would repeat./extend 4 --seconds 20works the same way; Esc keeps the parts made so far as a clip. - Clips through several pictures:
@bud.png @half-open.png @open.png the tulip opens --seconds 25goes to each picture in turn, in an equal share of the clip.
Fixed¶
lcode updateending inModuleNotFoundError: No module named 'rich._unicode_data…'. The update itself had worked; only its last message failed, and the Ollama update after it didn't run. Updating lcode can replace the files the running lcode is made of, so lcode now prepares what it will say before that and updates Ollama first. A git checkout is reinstalled on the Python lcode already runs on. Before, uv picked one from the folder you were in, such as a nearby.venv, and built a new environment when it was another version. An update runs the old version's code, so the update to this version can still show the error once. lcode is updated anyway: runlcode update ollamafor the rest.
0.50.0 - 2026-10-26¶
Added¶
- Ask lcode how (#184):
/help how do I make a clip from two pictures?answers from lcode's own manual, which now ships inside the package (the docs' pages, for the installed version, offline). The answer says what to type, filled in with your details, and the first step is typed into the next prompt for you. In image and video mode, a question is answered instead of being made into a picture (--rawmakes it anyway); in code mode, the coding model gets alcode_helptool that searches the manual. The prompt assistant's small model answers in image and video mode (qwen3.5:4b); every command and option in an answer must exist in lcode, and the lines to type go through lcode's own option parser (--seconds 90is caught), with one retry before what's still wrong is marked.
0.49.1 - 2026-10-26¶
Fixed¶
- Image and video jobs no longer run out of GPU memory right after switching modes. Switching to video mode in the first half minute after lcode starts unloaded the coding model while it was still loading; the load then finished and kept the GPU (10.8 GB of 12 here), and the next job sent to ComfyUI ran out of memory. A load that ends outside code mode now unloads itself, a job waits for a load still going, and unloading waits until Ollama has really freed the memory.
- Running out of GPU memory says what holds it (
Other programs hold GPU memory: obs (1.6 GB)…) instead of ComfyUI's whole error, and lcode starts its ComfyUI afresh: after running out of memory ComfyUI crashed on the next job. A crash now says it was one, with the log's path, instead of "ComfyUI closed the websocket". - A nearly full disk stops a job before it starts (under 1 GB free), and ComfyUI's "No space left on device" is said plainly: a film used to stop at shot 6 on it.
0.49.0 - 2026-10-24¶
Added¶
- Cloud generation in image and video mode (#157): xAI's Grok Imagine as cloud models,
grok-image(pictures, edits of one picture) andgrok-video(clips from text or a start picture), chosen with/modeland marked cloud in/modelsandlcode media. Before each request lcode shows what goes where and about what it costs, and sends nothing until you say yes; the session's total is shown after. The key comes fromXAI_API_KEYonly and goes to xAI's API only. What a cloud model can't do here (masks, several pictures, clips between two frames…) is made by a local model. See In the cloud.
Changed¶
- The README no longer says code mode makes images through MCP servers (it hasn't since 0.40.0).
Fixed¶
- Films: things and places no longer become cast members. A kite in the cast got a "character" picture (a person behind a kite) and walked on legs in every shot. The storyboard now keeps the cast to living characters and describes things in the shots: in 12 storyboards for ideas with props, things in the cast went from 6 to 1 (a plant).
0.48.0 - 2026-10-24¶
Added¶
- Films (#155):
/film a fox finds a lost kitten --shots 3makes a short film with the same characters in every shot. The prompt assistant writes the storyboard (saved, editable); each character gets a reference picture; each shot starts from a keyframe made with the cast as references and is animated; the shots are joined with cuts or fades, with an optional sound track./film shot 2remakes one shot,/film stitchjoins again. A real 3-shot film took 4 minutes on a 12 GB laptop GPU. See Films.
0.47.0 - 2026-10-24¶
Added¶
- The prompt assistant (#156): a small local model (
qwen3.5:4bwhen installed) writes fuller prompts in English for image and video mode, reads a start picture, takes follow-ups ("darker") as changes, and checks pictures against the request (check_results). See The prompt assistant. - Measured: video prompts get much better (a near-black "city at night rain" became a neon
street scene), picture prompts don't (FLUX.2 klein already reads short and non-English prompts
well), so by default it writes video prompts only (
prompt_assistant = video);alladds pictures and follow-ups.--rawsends one request as typed. - It's loaded for each job and unloaded right after, and your own words stay in the gallery.
Fixed¶
- The modes page has its "ComfyUI" heading back (it had slipped under video mode).
0.46.0 - 2026-10-24¶
Added¶
- Video mode: references, restyling, smoother and bigger clips, sound (#154, part 2). See Video mode.
@cat.png @hat.png --ref …makes a clip that follows reference pictures, and/restyle 4 …makes a clip again in a new look following its outlines (Wan 2.1 VACE 1.3B)./smooth 4doubles (or quadruples) the frames with FILM,/upscale 4now works on clips, and/sound 4 …adds sound made by Stable Audio Open to a clip.
Fixed¶
- Extending a clip keeps its pace: a 16 fps camera clip extended with the 24 fps model played too fast.
0.45.0 - 2026-10-24¶
Added¶
- Video mode: first and last frames, camera moves, longer clips (#154, part 1). See Video mode.
@start.png @end.png …makes a clip from one picture to the other (Wan 2.2 Fun InP 5B).--camera pan-leftand twelve other moves (pan, tilt, zoom, dolly, rotate), from a description or a start picture (Wan 2.1 Fun Camera 1.3B, 16 fps)./extend 4 --seconds 3 …in video mode continues a clip from its last frame into one longer clip.- lcode picks the model a job needs, installs it when you agree, and says so.
0.44.0 - 2026-10-24¶
Added¶
- Video mode makes videos (#153). Describe a clip, or start it from a picture (
@photo.png), and Wan 2.2 TI2V 5B makes it on your GPU. See Video mode. - Options:
--seconds(1 to 8, 5 by default, 24 fps) plus the image options. Quality presets: 864×480 in 12 or 20 steps, or 1280×704 in 30. A 5-second clip takes 100 to 150 s on a 12 GB laptop GPU. Clips are 16:9 (9:16 from an upright start image) unless you give a shape. - Results: an MP4 with its settings in a JSON file next to it, in the session's gallery; the
first frame shows in the terminal,
/playopens it, and/vary//againmake new clips. /animate 3in image mode makes a clip from image #3.lcode medialists video models, and/modelsin video mode picks one (thevideo_modelsetting).- With
notifyon, a long render ends with a desktop notification.
0.43.0 - 2026-10-24¶
Added¶
- Image mode edits pictures (#152). See Editing.
- Edit by instruction (
/edit 3 make the sky purple, or@photo.png make the sky purple) and references (@photo.png @#3 put the mug from image 1 on the table in image 2), with FLUX.2 klein 4B./use 3makes every next request follow an image until/use off. - Inpainting (
--mask mask.png: only the white part changes) and outpainting (/extend 3 --ar 16:9, or by sides: only the new border is painted, the original stays as it was). /upscale 3(Real-ESRGAN, 2× or 4×) and/nobg 3(BiRefNet, a transparent PNG), installed on first use./varyon an edit makes new edits of the same images.- ComfyUI's reasons for refusing a workflow now include its explanation, not just the input's name.
0.42.0 - 2026-10-24¶
Added¶
- Image mode's options, results and follow-ups (#151). See Image mode.
- Options in a request (
a red fox --n 4 --ar 16:9 --seed 42 --quality best --style photo) or for the session (/options n=4 ar=16:9): how many images, the shape, the seed (counting up over several images), a quality preset (0.5, 1 or 2 megapixels), ten style presets, and a negative prompt for models that use one. - Numbered results, shown in the terminal: the picture itself in kitty, Ghostty, iTerm2 and
WezTerm, and a half-block colour preview elsewhere (the
previewsetting). - Follow-ups:
/again(new seeds),/vary 3(four variations of image #3),/galleryand/open 3. The gallery and options are saved with the session. lcode --mode image -p "… --n 2"reports every image's file and seed.- One image is now named
<time>-<prompt>.png(no-1).
0.41.0 - 2026-10-23¶
Added¶
- Image models to choose from, installed by lcode (#150). lcode's media catalog lists the image models it can run, with their size, license, measured speed and whether they fit your GPU. Z-Image Turbo (photorealistic) joins FLUX.2 klein 4B (fast): 17 s and 3.5 s per 1024×1024 image on a 12 GB laptop GPU. See Image models.
- Choose one:
/modelsin image mode (numbered) or/model z-image-turboswitches; theimage_modelsetting is the default, and sessions keep theirs. - Installed on demand: a model that isn't installed downloads with your next image, after you
say yes, or ahead with
lcode media install <model>. lcode downloads only the missing files (models share text encoders and VAEs) into ComfyUI's model folders, checks the disk first, verifies each file's SHA-256, and resumes interrupted downloads.HF_TOKENworks for gated models. lcode medialists image and video models with what fits and what's installed, andlcode doctorsays how to install the image model when it's missing.
0.40.0 - 2026-10-23¶
Added¶
- Image mode makes images (#149). In image mode, what you type becomes a picture, made by FLUX.2 klein 4B on your GPU in about 8 seconds, with no language model involved. lcode drives ComfyUI itself. See Modes.
- ComfyUI is found or started: lcode uses a running ComfyUI as it is. Otherwise it starts
yours in the background when you enter image mode (
comfyui_path, thecomfyCLI's folder or a usual one). It stops that ComfyUI when you quit, go back to code mode or close the terminal. With none, it offers to install ComfyUI with comfy-cli. - Progress, cancel and type-ahead: the status line shows the step, the time left and your place in ComfyUI's queue. Esc or Ctrl+C cancels the job in ComfyUI too. What you type meanwhile waits in the next prompt.
- Files with their settings: images go to
.lcode/media/(themedia_dirsetting), each with a JSON file of its prompt, model, seed and size, also stored in the PNG itself. Images that finish after a crash are saved the next time. lcode --mode image -p "…"makes an image from a script, and--output jsongives the paths.lcode doctorreports ComfyUI and the image model.- No Ollama needed for images: lcode starts in image mode without Ollama or the coding
model;
/mode codesays what's missing.
Fixed¶
- lcode cleans up when its terminal closes. A closed terminal or a
killused to end lcode without stopping what it had started. Now it saves the session and stops its background jobs and servers. On Linux, a ComfyUI lcode started also stops if lcode is killed outright.
0.39.0 - 2026-10-23¶
Added¶
- Modes: code, image and video (#148).
/mode image,/mode videoand/mode codeswitch what lcode makes.lcode --modestarts in one, and themodesetting is the default. The banner, status bar and prompt show the mode, and sessions keep theirs. - The GPU goes to one model at a time: switching to image or video mode unloads the coding model, and switching back asks ComfyUI to free its memory.
- Image and video mode don't generate yet: they say what's coming (#149–#157) instead of sending your words to the coding model. See Modes.
Changed¶
- Code mode makes code only. The MCP servers that generated images and videos (
comfyui,xai) are out of the MCP catalog, and lcode doesn't load them in code mode even if they're still inmcp.json(it says so once). Image and video generation are moving to their own modes. /modeswitches code, image and video. The permission mode is set with/permissionsnow;/mode yoloand the like still work, with a note. Shift+Tab cycles the permission mode as before.
0.38.0 - 2026-10-21¶
Added¶
- The screenshot tool can click first. For a Python Qt program,
actionsrun before the picture, the way a user would:double_click Camera 1(a widget by its text) ordouble_click 300,200,click,right_click,key Escape,key ctrl+s,type …,wait 2. The result says what each step hit. Bugs that only show after a click were out of reach before. - Qt's warnings are listed at the top of a screenshot's report, taken from everything the program
printed. In a real run the cause (
setLayout: … which already has a layout) was there, among 30 lines of the program's logs, and the model looked past it.
Fixed¶
From a real session that "fixed" a double-click feature four times without once looking at the window, while the user saw a blank screen each time:
- Windows it never looked at are sent back once before the answer counts. When a request
changed a Python file that draws a window (PyQt, PySide, tkinter, Kivy, wx, GTK) and the model
never took a screenshot or looked at an image, it's told to look (with clicks first, if that's
where the change shows) or to say plainly that it's unchecked. The real cause was in what Qt
printed:
setLayout: … which already has a layout. - A rerun that only took a different time counts as a repeat: the model ran the same tests four
times in a row (
75 passed in 0.14s, then0.12s…) without the repeat check noticing. - A reply lcode drops for going round in circles is saved, and the message says where, so you can read what the model was thinking instead of taking the stop on trust.
- The checks use the project's own Python. They ran
python -m …from PATH, which doesn't have the project's dependencies, so a project with a.venv"failed" its checks on imports. Now.venv/bin/python(orvenv,env) runs them. - Tests written for pytest get pytest when the project has no pytest configuration: the tests
import it, or a requirements file lists it. They used to get
unittest, which can't run them.
0.37.0 - 2026-10-20¶
Added¶
- Ollama's cloud models (tags ending in
cloud, such asminimax-m3.1-flash:cloud) work as the main model or as the stronger model when stuck, and lcode says what they are: what it sends goes to ollama.com and is billed per token. The stronger-model panel said "free, on this machine" for them; it now says ollama.com and shows the cost. A request Ollama refuses because it isn't signed in says to runollama signin(in Docker, with your container's name), and lcode skips the local loading and GPU checks for them. The docs explain the setup and list the cloud models that call tools, with prices and published coding scores: see Cloud models through Ollama.
0.36.0 - 2026-10-18¶
Changed¶
/modelsand/modelshow a numbered list of models, and you switch by typing a number: no second/model <name>./model 2switches straight away, and a name still works. The list ends with the other Ollama models you've installed that can call tools, and a model that isn't installed says how to get it./contextsays why it stops at the model's largest window (256K for qwen3.6-35b): Ollama caps the window at what the model was trained on, whatever the memory, so 512K or 1M would only look larger.
0.35.1 - 2026-10-15¶
Fixed¶
Found in a real session that spent an hour on an Android build whose downloads this machine couldn't reach, then ended without a word:
- A site this machine can't reach now stops the guessing. lcode tracks, per site, the calls that
failed to reach it (HTTP errors, names that don't resolve, refused or timed-out connections) and
the ones that got an answer. After three different failed calls and none that worked, the model is
told to check one address that must exist and, if that fails too, to tell you what fails and what
you could do. Two more and lcode stops the request (status
blocked). Before, every new address looked like a new attempt, so nothing stepped in. - Thinking in circles is stopped as it streams. A model that repeats the same few paragraphs used to think until the context was full (here 70,000 characters, about ten minutes behind the spinner). Now the reply is dropped once its sentences keep repeating, and the model is asked for its next step or what's blocking it.
- A reply with no answer no longer ends the request as if it were done. An empty reply, or one
that filled the context before it said anything, gets a second chance; the second time lcode stops
(status
loop) and offers the stronger model first, when one is set. An answer the context cut off midway now says so. OpenAI-compatible servers report a cut-off reply too. - A
cdout of the project no longer moves lcode there. A model that rancd /tmptook the session, the project checks and the memory notes with it, and the session was saved under/tmp, solcode -cin the project couldn't find it. Now acdwithin the project is kept, and one out of it isn't: the next command starts in the project again. Sessions are saved under the folder lcode works in. - Scripts it never ran are sent back once before the answer counts: a model that wrote
build_android.shand called the job done never ran it, and its first line failed. It now runs it and fixes what fails, or says plainly that it's untested and why.
0.35.0 - 2026-10-14¶
Added¶
- A stronger model when stuck: with
escalate_modelset (a cloud model with your API key, such asanthropic/…,openai/…orxai/…, or a bigger local one,ollama/qwen3.8-27b), lcode offers to ask it for advice when the loop breaker would stop a request or the checks fail twice in a row, and/escalateasks at any time. Nothing is sent without asking, in any permission mode: you see why, what goes (your requests, the task sheet, the failure, the last steps and the files in question, never the whole session), where, and the cost (escalate_price), and can read it all first. Keys come only from environment variables and go only to their provider.lcode -p --escalateallows it for one run, andlcode bench --retry --escalate MODELmeasures it. Measured on a 12 GB laptop with qwen3.6-35b and the bigger local qwen3.8-27b as the advisor, onruntime-bugwith a second try: 1 of 3 passed with its advice, as without it (6 of 9 hidden-test problems fixed against 4), at about 9 minutes per question. A local advisor helps a little; a cloud model is the one to try. See A stronger model when stuck.
Fixed¶
-
Long command output can't fill the disk any more. lcode keeps the full text of long outputs so the model can read parts of them, but a command that printed gigabytes was saved whole, and the saved outputs had no total limit, only an age (7 days). Now one is kept to its first 20 MB (the model still gets the end and the error lines), and all of them together to 200 MB, oldest first.
-
A screenshot of a web app no longer starts its server a second time when one already answers at the URL (one the model started itself, say), which used to fail on the busy port and put the error in the result. A page that answers also counts when the server command itself exits.
0.34.0 - 2026-10-14¶
Added¶
- Screenshots without a screen: a
screenshottool runs a desktop program or opens a web page the way you would see it, saves the shot in.lcode/screenshots/, and returns what it shows with the sizes behind it. For a Python Qt program, that's each window against the screen, the size its layout needs, and the widgets that stick out or overlap. For a web page or a local server (headless Chrome), it's the page's size, what scrolls sideways, and the console errors and failed requests. Other desktop programs run on a virtual display (Xvfb). See Screenshots.
Fixed¶
- In a project with a virtual environment (
.venv,venvorenv), the model is told to run its Python (.venv/bin/python) rather than plainpython, which lacks the project's packages. Found when the model checked a PySide6 app with the system Python, failed, and stopped checking.
0.33.1 - 2026-10-14¶
Changed¶
- A new logo, on GitHub, PyPI and the docs site (header, favicon, home page and link previews),
with a light version for dark backgrounds.
scripts/make_logos.pymakes every size from the original.
Fixed¶
- The README's images and links work on PyPI: the logo, the demo and the links to the license and contributing guide were relative, so PyPI showed them broken.
- The source package is half the size (0.6 MB): it no longer carries the docs' images.
0.33.0 - 2026-10-14¶
Added¶
- Steer while it works. Type while a request runs and press Enter: the line goes to the model at its next step, as a note from you, without interrupting anything. The spinner shows what you're typing and the notes waiting. Esc pauses after the current step, so you can look, then go on with or without a note. Notes count as your requests for summaries and the review, and anything that never reached the model goes back into the prompt. A command that runs for a while now shows a spinner with its elapsed time. See Steer while it works.
Fixed¶
- "Where you left off" (
lcode -c) and session titles showed lcode's own messages to the model, such as a failed check, as if you had written them. They show your last request now.
0.32.0 - 2026-10-14¶
Added¶
- A review before the answer. After a request that changed files, the model checks its change
against your request, from the diff and its own answer, and real problems (part of the request
not done, an untested claim, a leftover debug print) go back to it once. One line says what the
review found. It's off by default: measured with qwen3.6-35b on a 12 GB laptop, it made requests
73% slower for one real catch. Turn it on with
lcode config set review auto.lcode -p --output jsonreports it inreview, andlcode bench --reviewcompares. See A review before the answer.
0.31.0 - 2026-10-14¶
Added¶
- A task sheet that survives long sessions. The plan (the todo list), the model's notes (decisions,
findings, open questions, kept in
todo_write's newnotes, or inside a todo item, where small models tend to put them), the files changed and the last check are pinned next to the summary and next to your request when old output is removed. It's saved with the session, solcode -cbrings it back (the todo list used to be lost), and/statusshows it. See Context management.
0.30.0 - 2026-10-13¶
Added¶
- Edits that the model copied imperfectly land where they belong instead of failing: line numbers
copied from
read_file, stray escapes, a block at the wrong indentation or spaces for tabs (the new text is re-indented), dropped blank lines, and very close matches. A close match changes only the lines the model meant to change, so an earlier edit its copy doesn't show survives. Ambiguous or weak matches are refused, and the model is shown the closest lines with their numbers. See Imperfect edits. apply_patch: the model can apply a unified diff to several files at once. Hunks are found by their text, so wrong@@line numbers don't matter, and a file changes only if every hunk fits.- Common tool slips get a useful answer instead of an error: a
greppattern that isn't a valid regex (def add() is searched as plain text, a path that doesn't exist comes back with the closest existing ones (Did you mean: src/pkg/math_utils.py?),read_fileon a folder lists it, andlist_diron a file says to read it. lcode bench --jsonbreaks failed tool calls down by tool (errors_by_tool).
0.29.0 - 2026-10-13¶
Added¶
- lcode runs the project's checks before calling work done. After a request that changed files, it
runs the tests (pytest, unittest, npm/pnpm/yarn/bun test, cargo test, go test or make test, found
from the project, or the command in
.lcode/settings.tomlor thecheckssetting). When they fail, the model gets the output and keeps working, twice at most, and then says what still fails. Checks the model already ran after its last change aren't repeated.lcode -p --output jsonreports them inchecks, andlcode bench --checks offcompares. See Checks before done.
Fixed¶
- A tool call that Ollama couldn't read because its XML was out of order (
expected element type <function> but have <parameter>) ended the request with an error. Like other malformed calls, it's now sent back to the model to make again.
0.28.1 - 2026-10-13¶
Fixed¶
Found by testing lcode's tools on real files and commands (#118):
- Some commands that change things ran without asking, because they were taken for read-only:
sort -o,uniq in out,tree -o,git branch -Dand other branch changes,git remote add/remove,git log/diff --output,rg --pre(which runs a program),find -fls,date -s, andnvidia-smisettings. Arguments are now checked.2>&1and>/dev/nullno longer make a read-only command ask. - "Always allow" for a command also covered chains that started with it: allowing
npm testletnpm test && rm -rf buildrun without asking. Now a chain needs its own "always". - A JSON tool call that the model only showed as an example in its answer was run. Text is now taken as a call only when the whole message is one.
- A change you made to a file while lcode asked to edit it was overwritten. lcode now notices and doesn't apply the edit, and the model redoes it on the current content.
replace_all: "false", and other values sent as text, were taken as true. They now get the type the tool declares.- The shell timeout wasn't enforced while a command kept printing, so a watcher started without
backgroundhung the request. Long output is no longer held in memory either: 400 MB of output used to cost 2.5 GB, and now it goes to a file, keeping its start, end and error lines. - One tool result could overflow the context window:
read_filereturned up to ~40K tokens, however small the window, and compaction only looked at the previous request. A result now takes at most a quarter of the window (read_filesays where to read on), and lcode makes room based on the size of the next request. - Sessions were saved only when a request ended, so a crash lost the whole request. They're now saved after every step, and atomically, so a crash while writing can't corrupt them.
- Editing a file with Windows line endings turned the whole file to Unix line endings, and editing a
file that isn't UTF-8 (Windows-1252 or Latin-1) replaced every accented letter in it with
�, not only in the edited line. Files now keep their encoding, line endings and byte order mark. A character the encoding can't hold is refused instead of written wrong.
0.28.0 - 2026-10-13¶
Added¶
- lcode says what's happening under the hood, in a line, only when it happens:
- loading the model and then reading the conversation are named in the status line instead of "Thinking";
- after a load, how long it took and how much of the model is on the GPU, with a warning when less is than usual;
- at startup and on
/model, a warning when a dense model won't fit on the GPU and will be slow, with what to use instead; - the vision model making the main model step aside, and the reload after it;
- an image tool taking the GPU;
- clearer explanations when Ollama runs out of memory.
The status bar shows when the model is still loading at startup. See What's happening under the hood.
Fixed¶
- After
/model, images were still described by the vision model chosen for the previous model. - The tests no longer write into your lcode state folder; one of them had left about a hundred
entries in
extensions-approved.json.
0.27.0 - 2026-10-10¶
Changed¶
- Long sessions are summarized earlier: once the conversation reaches 64K tokens (
compact_at), even when the window is larger. Before, a 256K window was summarized only at about 220K, far past where small models work well. lcode still removes old tool output first, and summarizes only if that's not enough.lcode config set compact_at 0keeps only the 85% rule, andlcode bench --compact-atcompares ceilings. Measured with qwen3.6-35b: making room earlier doesn't cost tasks (21 of 24 either way) and saved 17% of the time. See Context management. - Summaries keep your latest requests (up to eight) word for word, so the goal survives the
retelling.
/contextshows where the session makes room.
Fixed¶
- After lcode removed an earlier file read to make room, the model sometimes answered questions
about that file from memory, and got them wrong, instead of reading it again. The files whose
only copy went are now named next to the request it's working on, and the note in their place
says plainly that the content is gone. A conversation that was just summarized isn't summarized
again until it has grown by a quarter of
compact_at, which must be at least 16K (or 0).
0.26.0 - 2026-10-10¶
Added¶
- lcode checks for secrets before the model runs
git commitorgit push: passwords, keys and tokens in code or configuration, credentials in URLs (rtsp://user:pass@…), private keys, tokens with a known form, and.envfiles, in the lines a commit adds or the commits a push sends. Inaskmode you decide; inauto-editandyolothe command doesn't run, and the model is told to move the secret out of the code./commitand/prshow the findings before you confirm, and the GitHub Action doesn't commit at all.lcode: allow-secretin a comment allows a line;secret_check = falseturns the check off. See Secrets in commits.
0.25.0 - 2026-10-10¶
Changed¶
- The model is told what counts as checking its work: run what it changed the way it will be used,
reproduce each reported problem before fixing it, check what it can't see (GUIs, web pages,
layouts) without a screen, for example with Qt's
offscreenplatform, a headless browser or a rendered image, and never call something tested or done without having run it. See Checking its own work.
Added¶
lcode bench --tasks runtime-bug: a bug that shows only when the code runs, a camera wall whose window grows with every video frame. Reading the code suggests band-aids that fail its hidden test. It isn't in the default set.
0.24.0 - 2026-10-10¶
Added¶
- Ctrl+V in lcode's prompt pastes a screenshot: lcode reads the clipboard (with
wl-pasteon Wayland,xclipon X11,pngpasteorosascripton macOS), saves the image and attaches it as@path. Other clipboard content is pasted as text.lcode doctorshows whether it works here. - Image paths in a message are attached without
@: absolute and~paths, quoted paths with spaces, andfile://paths dragged into the terminal.@"path with spaces"works for any file.
Fixed¶
- An image path that doesn't exist is no longer ignored: lcode says so, and tells the model it can't see that image, so it asks instead of guessing what the picture shows. See Images.
0.23.0 - 2026-10-10¶
Added¶
- lcode notices when the model goes round in circles, making the same tool call with the same result
again and again, as small models sometimes do. The third time in a request (the second, for a call
that failed) the result tells the model so; two more, and lcode ends the request (status
loop, exit code4withlcode -p). Results that change, and checks run after an edit, don't count.repeat_limitchanges the threshold;0turns it off. See When the model repeats itself.
Fixed¶
- A tool call Ollama can't parse because its XML is broken (
XML syntax error … unexpected EOF, from models such as Qwen3.6 that write their calls in XML) is now sent back to the model to make again, like a call with broken JSON, instead of ending the request with an error.
0.22.0 - 2026-10-06¶
Added¶
lcode mcp add xai: images, image edits, videos and speech with xAI's Grok Imagine and text to speech APIs, in the cloud and billed to your xAI account. It asks for your xAI API key, and runslcode-mcp-xai, a new MCP server that ships with lcode, with the toolsgenerate_image,edit_image,generate_video(waits for the video and downloads it),video_statusandtext_to_speech. Files are saved ingenerated/in the project or where you ask. See Images, videos and speech with xAI.
0.21.0 - 2026-10-06¶
Added¶
lcode updateupdates lcode the way it was installed (uv tool, pipx, Homebrew, pip, orgit pullfor a checkout), and Ollama when it runs in Docker. The newest Ollama release comes from Docker Hub, not a locallatesttag, and is pulled by its version tag (such asollama/ollama:0.35.1) sodocker psshows the version. A container from Docker Compose gets the new tag in its compose file and is recreated withdocker compose up; any other container is recreated with the same settings, and the old one comes back if the new one doesn't start.lcode update --checkonly shows which versions are out. See Updating.
0.20.0 - 2026-10-06¶
Added¶
- Background commands: the model starts a dev server, a watcher or another process that keeps
running with
bashandbackground: true, reads its new output withbash_outputand stops it withbash_stop./jobslists them, and they all stop when the session ends. - Desktop notifications (
notify-sendon Linux,osascripton macOS, else the terminal bell) when a request that ran longer than 30 seconds is done, or lcode waits for your answer during one. Settings:notifyandnotify_after.
Fixed¶
- A command that starts something with
&no longer hangs until its timeout while that process holds the output open. What it left running is stopped when it ends, and the model is told to use a background command instead.
0.19.0 - 2026-10-06¶
Added¶
lcode acp: lcode inside editors that speak the Agent Client Protocol (Zed, JetBrains IDEs, Neovim, Emacs and others). The editor shows the answers and thinking as they stream, each tool call with the files it touches and a diff for edits, and asks permission questions and plan approvals in its own dialog. lcode's permission modes are the session's modes, the todo list is the editor's plan, and saved sessions reopen with their conversation. The model reads the editor's buffers (unsaved changes included) and edits through them, and MCP servers the editor passes join lcode's own. See Editors for setup.
Fixed¶
- Cancelling a request whose answer was cut off mid-stream no longer ends it as if it were done.
0.18.0 - 2026-10-06¶
Added¶
lcode -p "…" --output jsonprints one result object (status, answer, tool calls, files changed, token usage);--output stream-jsonprints an event per model step and tool result, then the result.--max-stepsand--allowed-toolslimit a run, and the exit code says how it ended. Without a person to ask, anything that needs permission is refused.- A Python API:
from lcode import Session, withrun(),stream()and anapprovecallback for permission requests. Seeexamples/api/fix_tests.py. - A GitHub Action for self-hosted runners (
uses: nasser1941/lcode@v0.18.0). It answers@lcodein issues and pull requests from the repository's owners, members and collaborators, proposes changes as a pull request (or pushes them to the pull request's branch), and can review pull requests. See Automation.
0.17.0 - 2026-10-06¶
Added¶
/commitwrites a commit message in the repository's style, from the diff, the conversation and recent commits, and commits once you approve it. It commits what you staged, or every change except files the model thinks are unrelated and files that may hold secrets./review [base]reviews the uncommitted changes, or the branch against a base, and reports findings withpath:line. The model can read but not change anything during a review./pr [base]pushes the branch and opens a GitHub pull request withghafter you approve the title and description; commits made onmainitself go to a new branch.- A command of your own called
commit,revieworprreplaces the built-in one. lcode --worktree [NAME](-w) runs the session in a git worktree on its own branch, so several sessions can work on one repository at once. It's removed at the end if nothing changed. See Git workflow.
Fixed¶
- With another model server,
/ctxno longer says the model reloads: it changes how much of the server's context window lcode uses.
0.16.0 - 2026-10-05¶
Added¶
- Other model servers: LM Studio, llama.cpp's llama-server, vLLM, MLX's server, or any other
server with an OpenAI-compatible API, instead of Ollama.
lcode config set backend lmstudio(orllama.cpp,vllm,mlx,openaiwithbase_url); a key can come fromLCODE_API_KEY. Tool calls, streaming, reasoning (on and off), images and structured answers work as with Ollama. lcode reads the context window from the server, matches the model setting to the models it serves, and moves thinking that a server sends as part of the answer out of it.lcode doctor,lcode models,lcode benchandlcode indexwork with every backend. See Other model servers, with setup for each.
0.15.0 - 2026-10-05¶
Added¶
- Hooks: shell commands that run on lcode's events (
before_tool,after_tool,after_request,session_start,notification), filtered by tool and file, with the event as JSON on stdin. Abefore_toolhook that exits with code 2 blocks the call and tells the model why; anafter_toolhook's output reaches the model when it fails (a linter) or withfeedback = true. Notification hooks run when lcode waits for you and when a long request is done. - Permission rules:
[permissions]allow and deny lists for shell commands, file edits, web domains and MCP tools, checked before lcode asks. Deny always wins, even inyolomode and for read-only commands, and is enforced for subagents too. - Both go in
config.tomlor a repository's.lcode/settings.toml, which is used only once approved (and approved again after any change).lcode doctorlists them. See Hooks and permission rules, with recipes for ruff, prettier, blocking force-pushes, running tests and desktop notifications.
Fixed¶
lcode config setno longer drops tables you added toconfig.tomlby hand.
0.14.0 - 2026-10-05¶
Added¶
- A repository map: the important files with their classes and functions, signatures and line
numbers, ranked by how much of the code uses them and cut to a token budget. Small repositories
(up to 300 source files) get it in the model's instructions; larger ones get a
repo_maptool, which can zoom into a folder. Python is read with its own parser; TypeScript, JavaScript, Go, Rust, Java and Kotlin by their declarations. Measured with qwen3.6-35b, it halved the steps the model took to find code and cut the tokens it read by 39%. Therepo_mapsetting turns it off. - Semantic code search: with an embedding model in Ollama (qwen3-embedding, nomic-embed-text, …),
lcode indexindexes the repository, and the model gets asearch_codetool that finds code by what it does. Changed files are embedded again before each search. During a session the embedding model runs on the CPU, so it never pushes lcode's model off the GPU. Theembed_modelsetting picks the model or turns it off. lcode bench --tasks find-concept(a task in a larger repository),--repo-mapand--code-search. See Repository map and code search.
0.13.0 - 2026-10-05¶
Added¶
- Code intelligence from language servers. With one installed (basedpyright or pyright for Python,
typescript-language-server or TypeScript 7's
tsc --lsp, gopls, rust-analyzer, clangd), the model gets anlsptool for a symbol's definition, references and type, and a file's or the workspace's symbols; and after every edit to a file in that language, the errors the edit introduced are added to the result, so the model fixes them right away. Servers start on demand, one per language; a project's ownnode_modules/.bincomes first. Without any server installed nothing changes.lcode doctorshows what it found; thelspsetting and~/.config/lcode/lsp.jsonconfigure it. See Code intelligence.
0.12.0 - 2026-10-05¶
Added¶
/contextshows what uses the context window, by category: the system prompt and its parts (AGENTS.md, memory notes, skills, MCP servers), tool definitions, your messages, attached files, the model's replies and each tool's results.lcode bench --sessionruns all tasks in one conversation, to test long sessions and context management;--rounds Nrepeats the tasks and--no-pruneturns the removal of old tool output off, to compare.
Changed¶
- Smarter context management, which matters most at 32K. When the window is 85% full, lcode first
replaces old tool output (file contents, command output, search results from before your last
two requests, and earlier copies of files that were read again) with one-line notes saying what
was there, then the output of your previous request if needed; it summarizes the conversation
only if that doesn't free enough room. Measured on 24-task sessions at 32K: fewer summaries, with
about as many tasks solved. The
prunesetting turns it off. - Long command and MCP output keeps its beginning, its end and the lines that look like errors; the full output is saved to a file the model can read.
0.11.1 - 2026-10-05¶
Fixed¶
lcode mcp add encordfailed with "needs the Encord SDK" when lcode itself was installed withuv tool install:uvxreused that installation, which doesn't have the SDK. The Encord and Valohai presets now runuvx --isolated. If you added either server with 0.11.0, add it again (lcode mcp add encord) or put--isolatedfirst in itsargsinmcp.json.
0.11.0 - 2026-10-05¶
Added¶
- Ready-made MCP servers for data and ML platforms:
lcode mcp add metabaseconnects to the MCP server built into Metabase 60 and later, with browser sign-in, so the model sees your data with your Metabase permissions.lcode mcp add encordandlcode mcp add valohai: those platforms have no MCP server of their own, so lcode now shipslcode-mcp-encord(projects, datasets, ontologies, labeling progress, tasks and labels, built on the Encord SDK) andlcode-mcp-valohai(executions with their logs, metrics and outputs, pipelines). Both are read-only unless started with--allow-writes, which adds assigning tasks and setting priorities (Encord) or starting and stopping executions (Valohai), and both work with other MCP clients too. See MCP servers.- Catalog questions that ask for an address now accept it without
https://.
0.10.0 - 2026-10-05¶
Added¶
- Custom commands: prompt templates in
.lcode/commands/<name>.mdor~/.config/lcode/commands/, run as/name args, with$ARGUMENTSand$1…$9, and an optional frontmatter for the description, an argument hint and the tools allowed for the request. - Skills in the Agent Skills format, from
.lcode/skills/,.agents/skills/and.claude/skills/(in the repository and in your home folder). Only their names and descriptions are in the system prompt; the model loads a skill's instructions with the newskilltool when a task matches, and its scripts and reference files when the instructions call for them./namestarts a skill yourself./helplists commands and skills and where each comes from. Theskillssetting limits them to lcode's own folders (lcode) or turns them off. An example command and skill are inexamples/. See Commands and skills.
Changed¶
- A repository's own commands, skills and agents are used only after you approve them; lcode asks
the first time and again when they change, like a project's
.mcp.json. Custom agents from a repository (new in 0.8.0) used to load without asking.
0.9.0 - 2026-10-05¶
Added¶
- Plan mode: a
planpermission mode (Shift+Tab fromask,/mode plan,lcode --plan, or/plan <request>) in which the model can read, search, run read-only commands, use the web and read-only subagents, but can't change anything. It presents a plan with the newpresent_plantool, and you approve it (inaskorauto-editmode), edit it, save it to.lcode/plans/, or send it back with feedback. An approved plan fills the todo list and is kept through compaction;/planshows it. See Plan mode.
Fixed¶
lcode mcp addshows the default value in its questions again, such asAWS region [us-east-1]; the terminal formatting swallowed it.- The end-of-session "Worth remembering?" list shows each note's type (
[feedback],[project], …), which the same problem hid.
0.8.0 - 2026-10-05¶
Added¶
- Subagents: the model can hand a task to a subagent with its own fresh context through the new
agenttool, and only the subagent's report comes back, so exploring a large codebase or reading long output doesn't fill the main context. Built-in types:exploreandplan(read-only) andworker(all tools). Custom agents are markdown files in.lcode/agents/or~/.config/lcode/agents/;@nameasks for one. Subagents use the session's model and context window (no reload), permissions (prompts are labelled with the agent), sandbox and checkpoints (/undocovers their changes). Withmax_parallel_agentsabove 1 and Ollama'sOLLAMA_NUM_PARALLEL, several run at the same time; workers that do get their own git worktree and their changes come back as a diff to approve./agentsshows the types and what each subagent did;lcode doctorestimates how many could run at once. See Subagents.
Changed¶
- A shell command that starts with
cd <folder> &&counts as read-only when the rest of it does, socd src && grep -n foo *.pyruns without asking. Acdwith$(…)or backticks still asks.
0.7.0 - 2026-10-05¶
Added¶
- Memory across sessions: lcode keeps short notes that later sessions load, only key knowledge and
not whole conversations. The model saves them with its new
memorytool (and updates or deletes them), you add them with/remember, and when a session ends (on quit,/clearor before compaction) one quick question asks the model what from it is worth remembering. Notes are per repository, shared by its git worktrees, plus a few for every repository; they're plain markdown files, managed with/memory. The index in the system prompt is capped at about 2,000 tokens. Notes that look like they contain secrets are never saved. Thememorysetting isask(confirm each note, the default),autooroff;--no-memoryturns it off for a session. See Memory.
0.6.1 - 2026-10-01¶
Fixed¶
- Tools marked
free_gpu(such as ComfyUI'srun_workflow) now always run to completion (wait: true): a job left running in the background competed with lcode's reloading model for the GPU.
0.6.0 - 2026-10-01¶
Added¶
- lcode can look at images: attach a screenshot, mockup or diagram with
@path, and the model can open images itself with the newview_imagetool. A model that can see describes the image in detail (all text transcribed, with your question in mind): your model itself if it can see, or the original model behind lcode's text-only variant, whichlcode setupalready downloaded (vision_modelsetting;lcode doctorshows which). Screenshots returned by MCP tools, such as Playwright's, are described too. lcode mcp add comfyui: generate and edit images with models you run locally in ComfyUI (FLUX, SDXL, Stable Diffusion 1.5, Qwen-Image), through ComfyUI's official MCP server, with only its local tools enabled.- MCP servers can mark tools that need the GPU to themselves (
free_gpuinmcp.json): lcode unloads its own model before they run and reloads it afterwards. On for ComfyUI's generation tools, so image models work on a 12 GB GPU next to a large coding model.
0.5.0 - 2026-10-01¶
Added¶
- An optional sandbox for shell commands: with
lcode --sandbox(orsandbox = "docker"/"podman"), the model's commands run in a container that only sees the project folder, as your user, with no network access unless you allow it (/sandbox network on,sandbox_network), no Linux capabilities and a process limit. The file tools are limited to the project too. lcode builds a default image with Python, Node.js, git and build tools on first use (sandbox_imagefor your own). Inauto-editmode, sandboxed commands run without asking. If the sandbox can't start, lcode stops instead of running commands without it.
0.4.1 - 2026-10-01¶
Added¶
- Homebrew:
brew install nasser1941/tap/lcodeon macOS (Apple Silicon and Intel) and Linux. The tap tests the formula on both and updates it automatically for each release, about a day after it's published. - Python 3.14 is supported and tested.
0.4.0 - 2026-10-01¶
Added¶
- MCP (Model Context Protocol) servers: the model can use tools from Jira, GitHub, AWS, databases and more, with your approval for every call.
lcode mcp add <name>sets up a ready-made server and tests it:atlassian,aws,aws-knowledge,google-drive,grafana,gcp,github,playwright,context7,sentry,postgres,kubernetes,linearandnotion(lcode mcp cataloglists them).- Any other server works too, local (stdio) or remote (Streamable HTTP), configured in
~/.config/lcode/mcp.jsonin the standardmcpServersformat, or per project in.mcp.json(used only after you approve it). - Browser sign-in (OAuth 2.1 with PKCE and dynamic client registration) with automatic token
refresh;
lcode mcp login/logout. - Speaks both the current MCP protocol (2026-07-28) and the earlier
initialize-based versions. - When the tool definitions would take more than 15% of the context window, the model finds tools
on demand instead (
mcp_toolssetting)./mcpshows servers, tools and their context cost;lcode --no-mcpstarts without them.
Fixed¶
lcode bench: stopping a run with Ctrl+C no longer leaves the running task's temporary folder behind, and the summary marks the model as stopped.lcode benchno longer lists the model it's about to benchmark as "already loaded".
0.3.1 - 2026-09-30¶
Added¶
lcode benchscores models on this machine with eight small coding tasks: fixing a bug, finding code, writing a script from a spec, renaming across files, a one-line edit in a long file, recovering from failing commands, fixing a function to match its spec and adding a feature across files. Each task runs in a temporary folder and has an automatic check, some with hidden tests. It reports tasks passed, time, generation and prompt speed, tool-call errors and memory use.lcode bench model-a model-bcompares models,--jsonsaves the results (a documented, versioned format) and--markdownprints a table for a model test report.
0.3.0 - 2026-09-30¶
Added¶
/undotakes back the file changes of the last request: lcode saves a checkpoint before the model first changes files in a request, and restores edited and deleted files and removes new ones, including changes made by shell commands./undoagain goes further back,/rewind Ngoes back to before request N (optionally removing those requests from the conversation too), and/checkpointslists them. Checkpoints live in a separate git repository in lcode's state folder, so your own repository is never touched, and work in folders that aren't git repositories. Turn them off withlcode config set checkpoints false.
0.2.2 - 2026-09-30¶
Added¶
/contextnow lets you change the context window mid-session: it lists the sizes the model supports with the memory each needs and whether it fits on the GPU, and you pick one (or type/context 128k). If the conversation is too long for a smaller size, lcode summarizes it first./ctxis a shortcut for the same command.
0.2.1 - 2026-09-29¶
Fixed¶
- File edits and writes failed with "IndexError: list index out of range" in the default
askpermission mode, a regression in 0.2.0.auto-editandyolomodes were not affected. - Slow models no longer look stuck: the status line keeps counting the elapsed time while the model thinks, answers or silently writes a long file into a tool call, and reminds you that Ctrl+C stops it.
0.2.0 - 2026-09-29¶
Added¶
- Web access for the model:
web_searchfinds current information (latest releases, docs, error messages) through Ollama web search, Brave Search, Tavily or a self-hosted SearXNG, andweb_fetchreads pages as clean text. On by default;lcode config set web ask|offorlcode --no-weblimits it. Search API keys are read from environment variables only. Closes #7.
0.1.2 - 2026-09-29¶
Added¶
- All eight catalog models are now tested end to end (
qwen3.8-27b,qwen3.6-27b,laguna-xs-2.1,nemotron-3.5-lightningandgpt-oss-20bjoined the three tested before), with a results table in the models guide.laguna-xs-2.1's cache size is now measured rather than estimated. - When a model fails to load because its context doesn't fit in GPU memory, lcode retries with half the
context and remembers the size that worked for that model (
~/.local/state/lcode/limits.json, shown bylcode doctor), so later sessions don't fail first. Found withnemotron-3.5-lightning, whose 1M context doesn't fit on a 12 GB GPU. - Named sessions and a session picker:
/rename <name>names the current session,/resumelists saved sessions and resumes the one you pick (by number, name, id or title),/resume allshows every folder, andlcode --resume [SESSION]does the same from the shell. Resuming shows a short recap of where you left off.
Changed¶
lcode setuprecommendsgpt-oss-20bat 64K (instead ofqwen3.5-4b) for 8 GB GPUs with 16 GB of RAM, and prefersqwen3.5-9bovergpt-oss-20bwhere both fit.- lcode is on PyPI as
lcode-cli; the installer now installs the latest release from PyPI instead of themainbranch.
Fixed¶
qwen3.6-35bcould crash Ollama with "CUDA error: an illegal memory access" on 12 GB GPUs: its prompt batch of 1024 left too little VRAM for the model's speculative-decoding context. The default is now Ollama's 512 (1024 remains available withlcode config set num_batch 1024), and when the GPU runs out of memory before answering, lcode retries automatically with a batch of 512.- A tool call with invalid JSON arguments no longer aborts the request: lcode tells the model what was wrong and lets it try again. Unknown or missing tool arguments are reported with the list of valid arguments.
lcode -ccontinues the most recently used session in the folder, not the most recently created one. Empty sessions are no longer saved./renamesaves the session right away, so a session named before its first request shows up in/resume.- Typing
exitorquit(without a slash) quits instead of being sent to the model as a request. - Compacted sessions keep their original title instead of showing the summary header.
0.1.1 - 2026-09-28¶
Added¶
qwen3.5-9bandqwen3.5-4bare now tested end to end, with measured speed and memory on a 12 GB GPU in the docs.lcode setupnow recommends only tested models on common machines.- Social preview image for the repository and link previews for the docs site.
- Release workflow that publishes to PyPI with Trusted Publishing.
Changed¶
- Memory estimates use ~1 GB of runtime overhead instead of 1.5 GB, matching measurements, so small
models get larger context windows (e.g.
qwen3.5-9bat 128K on a 12 GB GPU).
Fixed¶
lcode --model Xno longer reuses the context window saved for the default model; it picks the largest window that fits X./modeldoes the same inside a session.lcode modelsfits in 80-column terminals.
0.1.0 - 2026-09-28¶
First public release.
Added¶
- Interactive terminal agent with streaming markdown output, reasoning display,
@fileattachments, slash commands, a status bar and resumable sessions (lcode -c). - One-shot mode for scripts:
lcode -p "...". - Tools for the model:
read_file,write_file,edit_file(with diff previews and a whitespace-tolerant fallback),list_dir,glob,grep(ripgrep with a pure-Python fallback),bash(live output, timeouts, persistent working directory) andtodo_write. - Permission modes
ask,auto-editandyolo(Shift+Tab cycles), with a read-only command allowlist and "always allow" per command. - Model catalog with eight open-weight models, hardware detection for Linux + NVIDIA and Apple Silicon Macs, and memory-fit estimates that pick the largest context window that fits.
lcode setup(recommend, download and configure a model),lcode models,lcode doctorandlcode config.- Choice of context window per run (
--context 128k), per session (/ctx) or in the config file. - Text-only model variants that drop unused vision projectors to free GPU memory.
- Automatic conversation compaction when the context window is 85% full, plus
/compact. AGENTS.mdproject instructions and/initto generate them.- One-line installer for Linux and macOS.