aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2026-10-23history: keep the model a conversation is asked withDanilo M.1-4/+19
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-23cloud: provider keys from key_cmd, routing by model prefix, --cloud-modelsDanilo M.1-1/+96
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-23config: [[cloud]] providers and the tool cloud flagDanilo M.1-0/+48
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-23docs: plan cloud models, daemon sideDanilo M.1-0/+806
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-23docs: spec cloud models, chosen per chatDanilo M.1-0/+152
Providers are OpenAI-compatible [[cloud]] tables with a key_cmd; a chat stores its model; cloud chats get only tools marked cloud = true, and audio is always transcribed locally. Replaces the everything-local rule with no silent cloud fallback. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-23titles: let the model name a conversation after its first replyDanilo M.2-2/+44
A new conversation is still saved under the first six words of the question, then the model is asked for a title of at most six words in the conversation's language, thinking off, and the row is renamed. The request runs after done under the reply's turn, through ask(), so the next ask, stop, new or open hangs it up through interrupt(): a title never holds the next reply back, and a dropped one leaves the six words. A failure is logged only, since the reply is already saved. The self-check's fake llama-server answers title requests apart from the scripted replies, so a title can never take a reply meant for an ask in the test that follows. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-20clone tool: fail instead of prompting, lowercase host, update only on 1Danilo M.2-1/+14
A missing or private https repo made git ask for a username through SSH_ASKPASS, which on this desktop is pinentry-qt: a login dialog popped up and the tool waited up to its 300 s timeout. SSH_ASKPASS is now unset and GIT_TERMINAL_PROMPT=0, so the clone fails in about 0.2 s with "terminal prompts disabled". The host is lowercased before the map lookup, so GitHub.com finds GITHUB. update pulls only when it is 1, so a model passing 0 gets the question, not a pull. test-clone.sh checks all three; the new checks fail on the previous script. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19clone tool: name the target owner-repoDanilo M.4-21/+54
A generic repo such as evgen2571/wallpapers cloned into wallpapers; the target now takes the owner's name, collapsing when owner and repo match and falling back to the bare name when the URL has none.
2026-10-19docs: clone tool, tested through the modelDanilo M.1-0/+8
Known host clone, repeat asking to update plus update=1, unknown host asking to create the folder, to= cloning and the map append, and a remembered host cloning with no question, all driven through a second daemon on scratch dirs.
2026-10-19clone tool: clone a public repo into the host's folderDanilo M.5-0/+325
A public URL is cloned into ~/Programming/GIT/<host-folder>/<repo>, the folder chosen by host from ~/.config/desktop-assistant/clone-hosts (github.com, gitlab.com, forge.slackware.nl seeded). An unknown host or an already-cloned repo stops and asks with a concrete action; approving a new host folder appends it to the map, so later clones of that host ask nothing. Update pulls with --ff-only, never clones over the folder. test-clone.sh stubs git and uses a scratch tree and map.
2026-10-19docs: the new project tool, live checkDanilo M.2-1/+29
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19project tool: end the prompt with a period, plan-friendly opencode promptDanilo M.2-4/+10
The model sends the description without a final period, so the prompt read "...tool Fill in". opencode now starts in its plan agent, which cannot edit, so its prompt asks for a proposal instead. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19project tool: start a new project and open an agent on itDanilo M.7-1/+1254
project.sh makes a folder in ~/Programming/GIT with AGENTS.md, CLAUDE.md, README.md and LICENSE (GPLv2 only, GPLv2+, GPLv3 or MIT, texts shipped in licenses/), commits it signed, and opens Claude Code or opencode in a kitty tab, or a new window when no kitty listens. The tool description asks the model to confirm before calling. Check: test-project.sh. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19docs: plan the new project tool, agent choice in the specDanilo M.2-15/+274
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19docs: spec the new project toolDanilo M.1-0/+119
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19docs: voice input works end to end with the canvas micDanilo M.1-3/+4
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19voice input: ask takes a WAV, the model transcribes itDanilo M.4-13/+130
An ask with audio sends a transcribing state, has the loaded model transcribe the clip (thinking off, no tools, the prompt naming Italian or English), then answers the transcript as a typed question, so the user event, history and search hold the words. A stop while transcribing ends the turn like any reply. Refused: a relative path, a file that is not a WAV, over 60 s (the ubatch bound, untried for audio), a model without audio input, and an empty transcript. The length comes from the file size, since a recording stopped by a signal leaves the header's length unset. Live: 10/10 clips heard right, the right tool each time. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19docs: spec voice input, the model transcribesDanilo M.2-0/+118
Gemma4 heard 10/10 short clips right as a transcript, but answered straight from the audio in Estonian or Spanish 5 times in 6. So the daemon transcribes first and asks the words as text. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19docs: the desktop modes tool, live checkDanilo M.2-2/+24
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19mode tool: read and set the desktop modesDanilo M.4-0/+198
dnd, presentation, nolock and quiet through statusctl. The output is read back after a set, so a set that did not land is not reported as done, and a running game's hold on presentation is named. test-mode.sh checks it against a scratch runtime dir. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19docs: spec the desktop modes toolDanilo M.1-0/+72
The first tool that changes something: dnd, presentation, nolock and quiet through statusctl, applied with no confirmation since a mode is one bit and undone by asking again. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-17the ctx event carries the modelDanilo M.1-1/+3
The model name reached clients only in hello. The router can swap the model after that, and context_size() notices on the next request, but the ctx event that follows said only used and max, so the quickshell canvas kept showing the hello's name while another model answered. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-17--defaults prints the built-in config as JSONDanilo M.2-2/+12
For the quickshell settings window's Assistant page, which shows each default as the placeholder of its field and falls back to it when a field is emptied. Reading them from here keeps one copy of the defaults. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-17docs: autostart logs to the runtime dir, and whyDanilo M.1-0/+10
Moved from a comment in hypr-theme's autostart.lua, which no longer carries comments now that the quickshell settings window edits it. The log holds the questions asked, so it belongs somewhere private that is gone at reboot. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16write: always end with an outcome; empty replies and the meterDanilo M.1-12/+28
From review. A write ended by anything other than its own stop (an ask, a new conversation, another client's stop) sent only state idle, so the canvas waited for written or an error forever; interrupt() now sends error stopped with turn. A refused write carries turn too. An empty reply is refused instead of saved, a document made of several code blocks is no longer stripped as one fence, the write sizes its budget on the whole context since no history goes with it, and its usage no longer moves the chat's context meter. The file is written as UTF-8 whatever the locale. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Write a reply to a file: the write commandDanilo M.3-5/+129
For the canvas's To .md: a one-off request with attachments, no history, no thinking and no tools, whose reply is written to the first free name in a folder instead of the conversation. It streams progress as a character count, unwraps a reply the model fenced whole, and shares the turn counter, so it stops a reply in flight and stop ends it with no file left behind. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Spec the write commandDanilo M.1-0/+45
A one-off request with attachments whose reply goes to a file, for the canvas's To .md: no history, no thinking, a free name, nothing written when stopped. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Hang up on llama-server at once when a reply is stoppedDanilo M.1-11/+68
A stop was noticed at the reply's next chunk, and llama-server stops a task only when its client hangs up. Through a long prompt pass, measured at 12232 tokens for a file with page images, no chunk comes, so the GPU kept working after Stop until the first one. interrupt() now shuts the reply's socket from the stop handler; the request goes through http.client, which exposes it. A fake server that stays silent records the hang-up within a second of the stop. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Read a nameless file's content to tell whether it is textDanilo M.1-2/+34
A file whose name gives no type, such as a shebang script or a Makefile, was refused as not text. When mimetypes knows nothing, the first 4 KB decide: no NUL and valid UTF-8 (a character cut by the window allowed) is text. A name that does say something, even application/octet-stream, is still taken at its word. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Docs: images past the ubatch kill the model; the preset fixDanilo M.1-0/+11
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Docs: the two-image retry after the render fixDanilo M.1-0/+3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Docs: page render boundDanilo M.3-3/+3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Page renders never upscale; a non-stream answer is an errorDanilo M.1-11/+42
-scale-to 2048 also enlarges: a 600x300 pt scan went out as 2048x1024 beside a second image, and llama-server's model process died on it. The router then answered with a plain 'proxy error: Failed to read connection', which the parser skipped as a non-data line, so the turn finished as an empty reply. Pages now render at 150 dpi, lowered only for a page whose longest side would pass 1600 px (size from pdfinfo), and an answer with no data lines raises with its first line. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Docs: attachmentsDanilo M.3-3/+26
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Attachments: bound page renders, resolve once, no images for blind modelsDanilo M.1-37/+82
From review. A PDF page rendered at 150 dpi had no size bound: a 3000 pt page came out 6250x6250 at 171 MB, past the 20 MB cap an image file has; renders are now scaled to 2048 and refused over the cap. Images resolved on every tool round re-read files and re-rendered pages between rounds, and could flip to gone mid-answer; they resolve once per answer. A conversation holding images, reopened with a text-only model loaded, sent image parts the server rejected; the references become a note instead. Scanned pages past the fourth were dropped silently; they are listed as not shown and the chip is truncated. A pdftotext timeout or a missing poppler is a refusal, newlines in a name no longer close the file block, an image is hashed from the bytes it sends, and /models is read once per answer. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Take files with a questionDanilo M.1-31/+307
ask carries paths; the daemon reads them as the user. Text files and a PDF's text go inline within half the free context, shared equally, and are stored that way, so a reopened conversation still has them. Images, and PDF pages with no text, go as image parts to a model the router reports as seeing images, and are stored as a path and hash, resolved when sent: a file changed or gone since is reported as no longer available. turns gains a files column for the chips, added to an older database on open. A refused file is one log line, not a traceback. Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
2026-10-16Plan attachmentsDanilo M.1-0/+615
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Spec attachmentsDanilo M.1-0/+99
ask takes file paths; text and PDFs are extracted and inlined within half the free context, images and scanned pages go as image parts to a model that reports vision. History keeps extracted text inline and images as a path and hash, re-read on reopen while unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Mark the errors that end a replyDanilo M.2-5/+9
Every client sees every event, so a command error from one client, such as opening a conversation another just deleted, reached a canvas watching a reply and was taken as that reply's end: the canvas stopped rendering tokens the daemon kept streaming. An error from the reply thread now carries turn: true; command errors carry none. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Final review: startup probe, error messages, refusal exit, docsDanilo M.7-17/+44
A startup probe that failed with anything but OSError (llama-server not in router mode, an empty model list) crashed the daemon at boot, against its own comment; it is caught like the per-request failures now. llama-server's reasons reach the client: an HTTP error carries its body and an error chunk mid-stream its message, where the client used to see 'choices'. A daemon that refuses to start, or fails before serving, exits 1 instead of 0. assistant.sh no longer mistakes a --test run for the daemon. AGENTS.md said every section from Hardware to Kokoro was voice-only, which included findings the daemon still applies; it now names the voice ones. It also claimed a stopped reply emits nothing after done, while a late ctx can. The spec lists every error message, the example config no longer shows a non-default temperature as the default, and the updates tool stops telling the model not to read lists aloud. The spec is marked implemented. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Docs: desktop-assistant is a text daemon nowDanilo M.9-132/+128
README and AGENTS.md describe the socket protocol and the history instead of the voice pipeline; the voice-era findings stay, marked as such, since a return to voice starts from them and from the voice-v1 tag. The overlay spec is marked superseded, the daemon spec records what changed while planning, and the example tools lose their spoken announcements. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-16Make thinking a config key, on by defaultDanilo M.1-1/+2
Measured live: with reasoning on, a two-line question took 21 s of thinking before its first token, against about 5 s for questions that needed little. thinking = false in config.toml skips it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16assistant.sh: manage the daemon onlyDanilo M.1-34/+14
The overlay is gone from the quickshell repo, and the canvas that replaces it lives in the desktop shell, which has its own lifecycle. status now reports the process, the socket and the log. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Protocol: check command arguments, keep readers alive, one daemonDanilo M.1-33/+107
From review of the protocol commit. An exception inside a command, such as an id SQLite cannot bind, killed that client's reader and left a deaf connection still receiving events; commands now run under a try that answers error. Ids must be integers: delete with no id compared None to an unsaved conversation's None and wiped it, and a string id deleted the row through SQLite's affinity while the Python comparison missed, leaving the current conversation pointing at nothing so every later answer failed to save. ask, search and their arguments are type-checked instead of coerced. serve() refuses to start over a socket that answers, since unlinking it orphaned the live daemon's clients, and main() removes the socket only if it served it. The accept loop logs and continues on errors instead of dying silently, and a client line over a megabyte is dropped. The two known ceilings, a stalled client delaying everyone once for 2 s and a stopped reply holding the llama-server slot until its next chunk, are marked. New checks cover each of these and a client connecting mid-reply. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Take questions on the socket and stream replies backDanilo M.1-49/+363
Each client gets a reader thread; commands are ask, stop, new, list, open, search and delete. A reply streams think, token and tool events through say(), which checks the turn counter under the same lock that stop takes, so a stopped reply cannot send a token after its done. A finished turn is saved; a stopped one is not. hello carries the current conversation and any reply in flight, so a reconnecting canvas redraws both. Errors from llama-server become an error event and the daemon stays up; every exit path of main() logs a line, which the voice version's silent death lacked. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-16History: enforce modes on existing paths, index turns, strip NULDanilo M.1-4/+12
From review of the storage commit. makedirs and connect leave an existing directory or file at whatever mode it had, so a history dir made earlier with the default umask stayed 0755; both are now chmodded. A NUL in a search ended the FTS string and raised instead of returning nothing. The index on turns(conv, seq) serves the seq lookup per save, load_conv and the cascade, which otherwise scan every turn. The self-check now also asserts that an OR query matches nothing, which a non-raising but unquoted query would not. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Keep conversations in SQLite with full-text searchDanilo M.1-0/+126
One file under XDG_DATA_HOME, 0700 directory and 0600 file. turns.messages holds what the model is sent for the turn, user/reply/tools what the canvas draws. FTS5 with remove_diacritics finds città from citta. The query is quoted as one phrase so FTS syntax typed by the user is matched as text, and the self-check's integrity-check proves a conversation delete cascades into the index. Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
2026-10-16Drop the voice pipelineDanilo M.1-463/+33
The assistant becomes a text daemon behind the quickshell canvas. STT, TTS, the mic button loop, the presentation gate, the audio levels and the overlay config table go; the socket, the tool runner, the never-swap model rule and the config merge stay. The reply loop goes with the voice and returns with the text protocol. The voice version is tagged voice-v1, so later voice work starts from the tag instead of from dormant code. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Plan the text daemonvoice-v1Danilo M.1-0/+1574
Six tasks: tag voice-v1 and strip voice, SQLite history, the two-way protocol and reply loop, assistant.sh, a live check against the router, docs. Every code block was assembled stage by stage into a scratch copy and its self-check run; the planning-time deviations from the spec are listed at the top and folded back into it by the docs task. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-16Spec the text assistant daemonDanilo M.1-0/+180
Voice is dropped for now: the daemon becomes the single brain behind a quickshell conversation canvas, with typed turns over the existing socket, streamed think/token/tool events, and SQLite FTS5 history. The last voice commit gets tagged voice-v1 before deletion, so later voice work starts from the tag instead of dormant code. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>