feat: LLM auto-fallback + codegen pack + OCR pack

LLM auto-fallback (jarvis-app/src/app.rs + jarvis-core/src/config.rs):
- When neither intent classifier nor levenshtein finds a command match,
  route the utterance straight to the LLM instead of just playing the
  "not found" sound. Triggers ("скажи X", "answer Y") still work and
  take precedence — they short-circuit before command lookup.
- Two new config knobs:
    LLM_AUTO_FALLBACK = true        — master toggle
    LLM_AUTO_FALLBACK_MIN_CHARS = 4 — suppress for very short utterances
                                       so background noise doesn't burn
                                       Groq quota
- Requires GROQ_TOKEN; if absent, behaviour is unchanged (play_not_found).

codegen/ command pack:
- Phrases: "напиши код X", "сгенерируй скрипт Y", "write code Z", etc.
- Strips the trigger from the recognized phrase, sends what remains to
  Groq with a strict system prompt ("return ONLY code, no fences, no
  commentary") at temperature 0.2.
- Parses the JSON content, unescapes \n / \" / \\ / \t, strips any
  remaining ```lang ... ``` fences, drops the result into the clipboard
  via jarvis.system.clipboard.set.
- Notifies a 120-char preview + plays ok-sound. Works for any language
  the model handles (Python by default if unspecified).
- GROQ_TOKEN / GROQ_MODEL / GROQ_BASE_URL read from env at call time —
  same envvars the voice-loop fallback already uses.

ocr/ command pack:
- Phrases: "прочитай экран", "что на экране", "read screen", etc.
- Captures the primary screen via System.Windows.Forms + System.Drawing
  to a temp PNG, then shells to tesseract.exe (-l rus+eng for ru/ua,
  -l eng for en).
- Resolves Tesseract by PATH first, then C:\Program Files\Tesseract-OCR
  and the x86 install dir; if none works the user gets a friendly
  "winget install UB-Mannheim.TesseractOCR" hint in a notification.
- Recognized text goes to clipboard + a 200-char preview notification.

All three features are portable (env-var resolution, no hardcoded user
paths). Command-pack total is now 11. cargo test -p jarvis-core --lib
commands::tests passes 3/3.
This commit is contained in:
Bossiara13 2026-05-15 01:39:21 +03:00
parent 429113175a
commit 69a38b0d89
6 changed files with 255 additions and 5 deletions

View file

@ -459,12 +459,21 @@ fn execute_command(text: &str, rt: &tokio::runtime::Runtime) -> bool {
}
} else {
info!("No command found for: {}", text);
voices::play_not_found();
ipc::send(IpcEvent::Error {
message: format!("Command not found: {}", text)
});
if jarvis_core::config::LLM_AUTO_FALLBACK
&& crate::llm_fallback::is_enabled()
&& text.chars().count() >= jarvis_core::config::LLM_AUTO_FALLBACK_MIN_CHARS
{
info!("Auto-routing to LLM (no command match): {}", text);
crate::llm_fallback::handle(text);
} else {
voices::play_not_found();
ipc::send(IpcEvent::Error {
message: format!("Command not found: {}", text)
});
}
}
ipc::send(IpcEvent::Idle);
false // no chain on error or not found
}

View file

@ -197,6 +197,13 @@ pub const CMS_WAIT_DELAY: std::time::Duration = std::time::Duration::from_secs(1
pub const LLM_DEFAULT_ENABLED: bool = true;
pub const LLM_DEFAULT_MAX_HISTORY: usize = 8;
pub const LLM_DEFAULT_MAX_TOKENS: u32 = 256;
// Auto-route unrecognized commands to the LLM (no "скажи" trigger required).
// Triggers still work and take precedence — this only kicks in when neither
// the intent classifier nor the levenshtein fallback found a command match.
pub const LLM_AUTO_FALLBACK: bool = true;
// Don't auto-fallback for very short utterances — too noisy.
pub const LLM_AUTO_FALLBACK_MIN_CHARS: usize = 4;
pub const LLM_SYSTEM_PROMPT_RU: &str = "Ты — J.A.R.V.I.S. (Just A Rather Very Intelligent System), \
ИИ-ассистент Тони Старка из киновселенной Marvel (до событий Age of Ultron ты НЕ Vision и НЕ FRIDAY). \
Ведёшь себя как британский дворецкий: вежливо, иронично, с лёгким сарказмом, обращаешься к пользователю «сэр». \

View file

@ -0,0 +1,99 @@
local lang = jarvis.context.language
local phrase = (jarvis.context.phrase or ""):lower()
local triggers = {
"сгенерируй скрипт", "напиши скрипт", "сделай скрипт",
"сгенерируй код", "набросай код", "напиши код",
"generate code", "write code", "make script", "write script",
"згенеруй код", "напиши код", "зроби скрипт",
}
local request = phrase
for _, t in ipairs(triggers) do
local s, e = string.find(request, t, 1, true)
if s == 1 then
request = request:sub(e + 1)
break
end
end
request = request:gsub("^%s+", ""):gsub("%s+$", "")
if request == "" then
jarvis.system.notify(
lang == "ru" and "Codegen" or "Codegen",
lang == "ru" and "Что писать?" or "What to write?"
)
jarvis.audio.play_error()
return { chain = false }
end
local token = jarvis.system.env("GROQ_TOKEN")
if not token or token == "" then
jarvis.log("error", "GROQ_TOKEN not set")
jarvis.system.notify("Codegen", "GROQ_TOKEN не задан")
jarvis.audio.play_error()
return { chain = false }
end
local base = jarvis.system.env("GROQ_BASE_URL")
if not base or base == "" then base = "https://api.groq.com/openai/v1" end
local model = jarvis.system.env("GROQ_MODEL")
if not model or model == "" then model = "llama-3.3-70b-versatile" end
local system_prompt = lang == "ru"
and "Ты Senior Engineer. На запрос пользователя верни ТОЛЬКО исходный код, без обрамления ``` и без объяснений. Если язык не задан — выбери самый подходящий (по умолчанию Python). Не добавляй комментарии-болтовню."
or "You are a Senior Engineer. Reply with ONLY source code, no ``` fences, no commentary. Pick the most fitting language if unspecified (Python by default). Skip chatty comments."
jarvis.log("info", "codegen request: " .. request)
local payload = {
model = model,
messages = {
{ role = "system", content = system_prompt },
{ role = "user", content = request },
},
max_tokens = 1024,
temperature = 0.2,
}
local headers = { Authorization = "Bearer " .. token }
local res = jarvis.http.post_json(base .. "/chat/completions", payload, headers)
if not res.ok then
jarvis.log("error", "codegen API failed: status=" .. tostring(res.status) .. " body=" .. tostring(res.body):sub(1, 200))
jarvis.system.notify("Codegen", "Ошибка API: " .. tostring(res.status))
jarvis.audio.play_error()
return { chain = false }
end
local body = res.body or ""
local content = body:match('"content"%s*:%s*"(.-[^\\])"')
if not content then
jarvis.log("error", "codegen: could not parse content from " .. body:sub(1, 200))
jarvis.system.notify("Codegen", "Не смог распарсить ответ")
jarvis.audio.play_error()
return { chain = false }
end
content = content:gsub('\\n', '\n')
content = content:gsub('\\"', '"')
content = content:gsub('\\\\', '\\')
content = content:gsub('\\t', '\t')
content = content:gsub("^```[%w]*\n?", "")
content = content:gsub("\n?```%s*$", "")
content = content:gsub("^%s+", ""):gsub("%s+$", "")
jarvis.system.clipboard.set(content)
jarvis.log("info", "codegen result copied to clipboard (" .. #content .. " bytes)")
local preview = content:sub(1, 120)
if #content > 120 then preview = preview .. "..." end
jarvis.system.notify(
lang == "ru" and "Код в буфере" or "Code in clipboard",
preview
)
jarvis.audio.play_ok()
return { chain = false }

View file

@ -0,0 +1,27 @@
[[commands]]
id = "codegen_clipboard"
type = "lua"
script = "codegen.lua"
sandbox = "full"
timeout = 30000
[commands.phrases]
ru = [
"напиши код",
"сгенерируй код",
"набросай код",
"сделай скрипт",
"напиши скрипт",
"сгенерируй скрипт",
]
en = [
"write code",
"generate code",
"make script",
"write script",
]
ua = [
"напиши код",
"згенеруй код",
"зроби скрипт",
]

View file

@ -0,0 +1,26 @@
[[commands]]
id = "read_screen"
type = "lua"
script = "read_screen.lua"
sandbox = "full"
timeout = 30000
[commands.phrases]
ru = [
"прочитай экран",
"что на экране",
"распознай текст с экрана",
"что написано на экране",
"сними текст с экрана",
]
en = [
"read screen",
"what is on screen",
"ocr screen",
"extract text from screen",
]
ua = [
"прочитай екран",
"що на екрані",
"розпізнай текст з екрана",
]

View file

@ -0,0 +1,82 @@
local lang = jarvis.context.language
local temp = jarvis.system.env("TEMP") or jarvis.system.env("TMP") or "C:\\Windows\\Temp"
local ts = jarvis.context.time.year .. jarvis.context.time.month .. jarvis.context.time.day
.. "_" .. jarvis.context.time.hour .. jarvis.context.time.minute .. jarvis.context.time.second
local img = temp .. "\\jarvis_ocr_" .. ts .. ".png"
local capture_ps = string.format(
[[Add-Type -AssemblyName System.Windows.Forms,System.Drawing; $b=[System.Windows.Forms.Screen]::PrimaryScreen.Bounds; $bmp=New-Object Drawing.Bitmap $b.Width,$b.Height; $g=[Drawing.Graphics]::FromImage($bmp); $g.CopyFromScreen($b.Location,[Drawing.Point]::Empty,$b.Size); $bmp.Save('%s'); $g.Dispose(); $bmp.Dispose()]],
img:gsub("\\", "\\\\")
)
local capture_cmd = string.format('powershell -NoProfile -ExecutionPolicy Bypass -Command "%s"', capture_ps:gsub('"', '\\"'))
local cap = jarvis.system.exec(capture_cmd)
if not cap.success or not jarvis.fs.exists(img) then
jarvis.log("error", "screen capture failed: " .. tostring(cap.stderr))
jarvis.system.notify("OCR", lang == "ru" and "Не смог снять экран" or "Could not capture screen")
jarvis.audio.play_error()
return { chain = false }
end
local tesseract_paths = {
"tesseract.exe",
"C:\\Program Files\\Tesseract-OCR\\tesseract.exe",
"C:\\Program Files (x86)\\Tesseract-OCR\\tesseract.exe",
}
local tess = nil
for _, p in ipairs(tesseract_paths) do
local check
if p == "tesseract.exe" then
check = jarvis.system.exec("where tesseract.exe")
else
check = { success = jarvis.fs.exists(p), stdout = p }
end
if check.success and (check.stdout or ""):gsub("[\r\n]+$", "") ~= "" then
tess = p == "tesseract.exe" and "tesseract.exe" or '"' .. p .. '"'
break
end
end
if not tess then
jarvis.log("error", "tesseract not found")
jarvis.system.notify(
"OCR",
lang == "ru"
and "Tesseract не установлен. Установи: winget install UB-Mannheim.TesseractOCR"
or "Tesseract not installed. Install: winget install UB-Mannheim.TesseractOCR"
)
jarvis.audio.play_error()
return { chain = false }
end
local langs = lang == "en" and "eng" or "rus+eng"
local ocr_cmd = string.format('%s "%s" - -l %s', tess, img, langs)
local ocr = jarvis.system.exec(ocr_cmd)
if not ocr.success then
jarvis.log("error", "tesseract failed: " .. tostring(ocr.stderr):sub(1, 200))
jarvis.system.notify("OCR", lang == "ru" and "Tesseract упал" or "Tesseract failed")
jarvis.audio.play_error()
return { chain = false }
end
local text = (ocr.stdout or ""):gsub("^%s+", ""):gsub("%s+$", "")
if text == "" then
jarvis.system.notify("OCR", lang == "ru" and "Текста не найдено" or "No text found")
jarvis.audio.play_not_found()
return { chain = false }
end
jarvis.system.clipboard.set(text)
jarvis.log("info", "OCR result (" .. #text .. " bytes) copied to clipboard")
local preview = text:gsub("\r?\n", " "):sub(1, 200)
if #text > 200 then preview = preview .. "..." end
jarvis.system.notify(
lang == "ru" and "Текст с экрана" or "Screen text",
preview
)
jarvis.audio.play_ok()
return { chain = false }