@pheem49/mint 1.9.0 → 1.10.0

package/crates/mint-core/src/video_gen.rs

@@ -0,0 +1,235 @@
+use base64::{Engine as _, engine::general_purpose::STANDARD};
+use serde::{Deserialize, Serialize};
+use serde_json::{Value, json};
+use thiserror::Error;
+
+use crate::{MintConfig, PictureEntry};
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct VideoGenRequest {
+    pub prompt: String,
+    pub negative_prompt: Option<String>,
+    pub aspect_ratio: String,
+    pub duration: u32,
+    pub model: Option<String>,
+    pub provider: String,
+}
+
+#[derive(Debug, Clone, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct VideoGenResponse {
+    pub videos: Vec<PictureEntry>,
+    pub model: String,
+    pub provider: String,
+    pub prompt: String,
+}
+
+#[derive(Debug, Error)]
+pub enum VideoGenError {
+    #[error("missing Gemini API key — set GEMINI_API_KEY or configure api_key in config")]
+    MissingApiKey,
+    #[error("API request failed: {0}")]
+    Request(#[from] reqwest::Error),
+    #[error("unexpected response structure: {0}")]
+    UnexpectedResponse(String),
+    #[error("video generation timed out")]
+    Timeout,
+    #[error("failed to save generated video: {0}")]
+    SaveError(String),
+    #[error(
+        "Google Veo model endpoint returned 404 Not Found. Google Veo video generation requires an API Key or Google Cloud project with Veo Model access enabled: {0}"
+    )]
+    ModelNotFound(String),
+}
+
+pub async fn generate_video(
+    config: &MintConfig,
+    request: &VideoGenRequest,
+) -> Result<VideoGenResponse, VideoGenError> {
+    let api_key = if config.api_key.trim().is_empty() {
+        std::env::var("GEMINI_API_KEY").unwrap_or_default()
+    } else {
+        config.api_key.clone()
+    };
+    if api_key.trim().is_empty() {
+        return Err(VideoGenError::MissingApiKey);
+    }
+
+    let mut model_owned = request
+        .model
+        .as_deref()
+        .map(str::trim)
+        .filter(|m| !m.is_empty())
+        .map(str::to_owned)
+        .unwrap_or_else(|| {
+            config
+                .extra
+                .get("veoModel")
+                .and_then(|v| v.as_str())
+                .unwrap_or("veo-2.0-generate-001")
+                .to_string()
+        });
+
+    if model_owned == "veo-2.0-flash-exp" {
+        model_owned = "veo-2.0-generate-001".to_string();
+    }
+
+    let client = crate::HTTP_CLIENT.clone();
+    let mut url = format!(
+        "https://generativelanguage.googleapis.com/v1beta/models/{model_owned}:predictLongRunning"
+    );
+
+    let payload = json!({
+        "instances": [
+            {
+                "prompt": request.prompt
+            }
+        ],
+        "parameters": {
+            "aspectRatio": request.aspect_ratio,
+            "durationSeconds": request.duration,
+            "sampleCount": 1
+        }
+    });
+
+    let mut response_res = client
+        .post(&url)
+        .header("x-goog-api-key", &api_key)
+        .json(&payload)
+        .send()
+        .await?;
+
+    if response_res.status() == reqwest::StatusCode::NOT_FOUND && model_owned != "veo-2.0-flash-001"
+    {
+        model_owned = "veo-2.0-flash-001".to_string();
+        url = format!(
+            "https://generativelanguage.googleapis.com/v1beta/models/{model_owned}:predictLongRunning"
+        );
+        response_res = client
+            .post(&url)
+            .header("x-goog-api-key", &api_key)
+            .json(&payload)
+            .send()
+            .await?;
+    }
+
+    if !response_res.status().is_success() {
+        let status = response_res.status();
+        let err_text = response_res.text().await.unwrap_or_default();
+        if status == reqwest::StatusCode::NOT_FOUND {
+            return Err(VideoGenError::ModelNotFound(format!(
+                "404 for model {model_owned} on generativelanguage.googleapis.com API. ({err_text})"
+            )));
+        }
+        return Err(VideoGenError::UnexpectedResponse(format!(
+            "HTTP {status}: {err_text}"
+        )));
+    }
+
+    let res: Value = response_res.json().await?;
+
+    let operation_name = res
+        .get("name")
+        .and_then(|v| v.as_str())
+        .ok_or_else(|| VideoGenError::UnexpectedResponse("Missing operation name".to_string()))?;
+
+    let poll_url = format!("https://generativelanguage.googleapis.com/v1beta/{operation_name}");
+    let mut done = false;
+    let mut final_response = None;
+
+    // Poll every 5 seconds, up to 60 times (5 minutes max)
+    for _ in 0..60 {
+        tokio::time::sleep(tokio::time::Duration::from_secs(5)).await;
+        let poll_res: Value = client
+            .get(&poll_url)
+            .header("x-goog-api-key", &api_key)
+            .send()
+            .await?
+            .error_for_status()?
+            .json()
+            .await?;
+
+        if let Some(d) = poll_res.get("done").and_then(|v| v.as_bool()) {
+            if d {
+                done = true;
+                final_response = Some(poll_res);
+                break;
+            }
+        }
+    }
+
+    if !done {
+        return Err(VideoGenError::Timeout);
+    }
+
+    let op_res = final_response.ok_or(VideoGenError::Timeout)?;
+
+    // Check if error occurred in operation
+    if let Some(err) = op_res.get("error") {
+        let msg = err["message"].as_str().unwrap_or("unknown operation error");
+        return Err(VideoGenError::UnexpectedResponse(msg.to_string()));
+    }
+
+    let response_obj = op_res
+        .get("response")
+        .ok_or_else(|| VideoGenError::UnexpectedResponse("Missing response object".to_string()))?;
+
+    // Try to find generated samples or outputs
+    let generated_samples = response_obj
+        .get("generateVideoResponse")
+        .and_then(|gvr| gvr.get("generatedSamples"))
+        .or_else(|| {
+            response_obj
+                .get("outputs")
+                .and_then(|o| o.as_array()?.first()?.get("generatedVideos"))
+        })
+        .ok_or_else(|| {
+            VideoGenError::UnexpectedResponse(
+                "No generated samples or videos in response".to_string(),
+            )
+        })?;
+
+    let sample = generated_samples
+        .as_array()
+        .and_then(|arr| arr.first())
+        .ok_or_else(|| {
+            VideoGenError::UnexpectedResponse("Generated samples array is empty".to_string())
+        })?;
+
+    let video_uri = sample
+        .get("video")
+        .and_then(|v| v.get("uri"))
+        .and_then(|u| u.as_str())
+        .ok_or_else(|| VideoGenError::UnexpectedResponse("Missing video URI".to_string()))?;
+
+    // Download the video file bytes
+    let video_bytes = client
+        .get(video_uri)
+        .header("x-goog-api-key", &api_key)
+        .send()
+        .await?
+        .error_for_status()?
+        .bytes()
+        .await?
+        .to_vec();
+
+    // Convert to data URI for saving
+    let data_b64 = STANDARD.encode(&video_bytes);
+    let data_uri = format!("data:video/mp4;base64,{data_b64}");
+
+    // Save to the pictures directory and record index entry
+    let saved = crate::pictures::save_chat_images(
+        vec![data_uri],
+        Some("veo".to_string()),
+        Some(request.prompt.clone()),
+    )
+    .map_err(|e| VideoGenError::SaveError(e.to_string()))?;
+
+    Ok(VideoGenResponse {
+        videos: saved,
+        model: model_owned,
+        provider: "veo".to_string(),
+        prompt: request.prompt.clone(),
+    })
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pheem49/mint",
-  "version": "1.9.0",
+  "version": "1.10.0",
   "description": "A native Tauri desktop AI assistant with a Rust backend and React UI.",
   "scripts": {
     "dev:web": "vite --config vite.config.web.ts",
@@ -56,6 +56,7 @@
     "mint": "src/bin/index.js"
   },
   "dependencies": {
+    "@pouyanafisi/gmail-mcp": "^2.0.0",
     "@tauri-apps/api": "^2.0.0",
     "@vitejs/plugin-react": "^6.0.1",
     "@xyflow/react": "^12.11.1",
@@ -71,6 +72,11 @@
     "@tauri-apps/cli": "^2.0.0"
   },
   "overrides": {
-    "gh-pages": "^5.0.0"
+    "gh-pages": "^5.0.0",
+    "@modelcontextprotocol/sdk": "^1.24.0",
+    "nodemailer": "^9.0.0",
+    "uuid": "^11.1.1",
+    "brace-expansion": "^5.0.8",
+    "@hono/node-server": "^2.0.5"
   }
 }

package/shared-modules/luajit/luajit.json

@@ -12,8 +12,8 @@
         {
             "type": "git",
             "url": "https://github.com/openresty/luajit2.git",
-            "tag": "v2.1-20260620",
-            "commit": "b411bec3ce550ef9968fc83bca094455cf812c1f",
+            "tag": "v2.1-20260717",
+            "commit": "1e5c0db1a86c9d0645bdd05e88e3e7a295831074",
             "x-checker-data": {
                 "type": "git",
                 "tag-pattern": "^v([\\d.-]+)$",

package/skills-lock.json

@@ -0,0 +1,41 @@
+{
+  "version": 1,
+  "skills": {
+    "debug-mantra": {
+      "source": "thananon/9arm-skills",
+      "sourceType": "github",
+      "skillPath": "skills/engineering/debug-mantra/SKILL.md",
+      "computedHash": "7c519ffda61c6a44db8ef12f6a9ed06de920b9b4faa2e7b47b0032b9c78056e3"
+    },
+    "management-talk": {
+      "source": "thananon/9arm-skills",
+      "sourceType": "github",
+      "skillPath": "skills/productivity/management-talk/SKILL.md",
+      "computedHash": "65bb8510492db4e31f465e0ba4cc73570772d1dde75839c49993cd85a8437a54"
+    },
+    "post-mortem": {
+      "source": "thananon/9arm-skills",
+      "sourceType": "github",
+      "skillPath": "skills/engineering/post-mortem/SKILL.md",
+      "computedHash": "124d195540d43cc922ee93d398996ac8c4294528e92e4d3d6ad83deaf5cfe59a"
+    },
+    "qwen-agent": {
+      "source": "thananon/9arm-skills",
+      "sourceType": "github",
+      "skillPath": "skills/engineering/qwen-agent/SKILL.md",
+      "computedHash": "237e1a708655b68ab8066455d16b1e14ccba55ab682764e21843f2e47848f0a3"
+    },
+    "qwenchance": {
+      "source": "thananon/9arm-skills",
+      "sourceType": "github",
+      "skillPath": "skills/productivity/qwenchance/SKILL.md",
+      "computedHash": "116b0e8cc306b2ee891dd9a70160842528420153273e0109933224b49c22adf8"
+    },
+    "scrutinize": {
+      "source": "thananon/9arm-skills",
+      "sourceType": "github",
+      "skillPath": "skills/engineering/scrutinize/SKILL.md",
+      "computedHash": "d82a0b76617a00069f972cd0efd57c591e18b56f7aa0652aec63f6cb37cfd735"
+    }
+  }
+}

package/src/renderer/index-web.html

@@ -6,7 +6,7 @@
   <title>Mint Agent (Web)</title>
   <link rel="icon" type="image/png" href="./assets/icon.png">
   <!-- Load font families used in settings -->
-  <link href="https://fonts.googleapis.com/css2?family=Outfit:wght@300;400;500;600&family=Mali:wght@400;500;600&family=Prompt:wght@400;500&family=Sarabun:wght@400;500&display=swap" rel="stylesheet">
+  <link href="https://fonts.googleapis.com/css2?family=Outfit:wght@300;400;500;600&family=Mali:wght@400;500;600&family=Prompt:wght@400;500;600&family=Sarabun:wght@400;500;600&family=Kanit:wght@400;500;600&family=Mitr:wght@400;500&family=Noto+Sans+Thai:wght@400;500;600&family=Inter:wght@400;500;600&family=Fira+Code:wght@400;500&display=swap" rel="stylesheet">
   <!-- Cubism must be available before the Live2D module is imported. -->
   <script src="./Live2DCubismCore.js"></script>
 </head>

package/src/renderer/index.html

@@ -6,7 +6,7 @@
   <title>Mint Agent</title>
   <link rel="icon" type="image/png" href="./assets/icon.png">
   <!-- Load font families used in settings -->
-  <link href="https://fonts.googleapis.com/css2?family=Outfit:wght@300;400;500;600&family=Mali:wght@400;500;600&family=Prompt:wght@400;500&family=Sarabun:wght@400;500&display=swap" rel="stylesheet">
+  <link href="https://fonts.googleapis.com/css2?family=Outfit:wght@300;400;500;600&family=Mali:wght@400;500;600&family=Prompt:wght@400;500;600&family=Sarabun:wght@400;500;600&family=Kanit:wght@400;500;600&family=Mitr:wght@400;500&family=Noto+Sans+Thai:wght@400;500;600&family=Inter:wght@400;500;600&family=Fira+Code:wght@400;500&display=swap" rel="stylesheet">
   <!-- Cubism must be available before the Live2D module is imported. -->
   <script src="./Live2DCubismCore.js"></script>
 </head>

package/src/renderer/shared/constants/slashCommands.ts

@@ -0,0 +1,27 @@
+export interface SlashCommand {
+  command: string
+  description: string
+  usage?: string
+  category?: 'system' | 'workspace' | 'models' | 'tools'
+}
+
+export const SLASH_COMMANDS: SlashCommand[] = [
+  { command: '/cd', description: 'Change active workspace directory', usage: '/cd <path>', category: 'workspace' },
+  { command: '/code', description: 'Run in code-agent mode', usage: '/code <task>', category: 'tools' },
+  { command: '/fast', description: 'Toggle fast mode (hide thinking traces)', usage: '/fast [on|off]', category: 'system' },
+  { command: '/help', description: 'Show interactive help menu', usage: '/help', category: 'system' },
+  { command: '/image', description: 'Attach image from disk', usage: '/image [path] [prompt]', category: 'tools' },
+  { command: '/image-provider', description: 'List or switch default image gen provider', usage: '/image-provider [name]', category: 'models' },
+  { command: '/learn', description: 'Import persistent skill/instruction file', usage: '/learn [path]', category: 'tools' },
+  { command: '/mcp', description: 'List configured MCP servers & status', usage: '/mcp [list|allow]', category: 'tools' },
+  { command: '/memory', description: 'Manage long-term memory store', usage: '/memory [list|clear|get|set]', category: 'tools' },
+  { command: '/models', description: 'List AI providers or switch active provider/model', usage: '/models [provider/model]', category: 'models' },
+  { command: '/multi-agent', description: 'Toggle Multi-Agent Collaboration system', usage: '/multi-agent [on|off]', category: 'system' },
+  { command: '/paste', description: 'Attach image from clipboard', usage: '/paste [prompt]', category: 'tools' },
+  { command: '/plugins', description: 'List or generate plugins/skills', usage: '/plugins [name]', category: 'tools' },
+  { command: '/skill add', description: 'Add or install global skill file or folder', usage: '/skill add <path>', category: 'tools' },
+  { command: '/stats', description: 'Show session statistics', usage: '/stats', category: 'system' },
+  { command: '/veo', description: 'Generate video using Google Veo', usage: '/veo <prompt>', category: 'tools' },
+  { command: '/video-provider', description: 'List or switch default video gen provider', usage: '/video-provider [name]', category: 'models' },
+]
+

package/src/renderer/shared/platform.ts

@@ -28,6 +28,7 @@ export interface MintPlatformApi {
     message: string,
     imageDataUri?: string | null,
     audioDataUri?: string | null,
+    videoDataUri?: string | null,
     documentAttachment?: any | null,
     workspacePath?: string | null,
     chatId?: string | null,
@@ -38,6 +39,7 @@ export interface MintPlatformApi {
     onChunk: (chunk: string) => void,
     imageDataUri?: string | null,
     audioDataUri?: string | null,
+    videoDataUri?: string | null,
     systemInstruction?: string,
     onProgress?: (progress: AgentProgress) => void,
     documentAttachment?: any | null,
@@ -51,6 +53,7 @@ export interface MintPlatformApi {
   saveSystemInteraction(
     chatId: string,
     userText: string,
+    aiText: string,
     provider: string,
     model: string,
   ): Promise<any>

package/src/renderer/shared/types.ts

@@ -8,6 +8,7 @@ export interface RuntimeStatus {
   backend: string
   configPath: string
   activeProvider: string
+  activeModel?: string
   availableProviders: string[]
   integrations: Record<string, unknown>
   localIp?: string
@@ -76,6 +77,9 @@ export interface ImageGenRequest {
   numImages?: number
   model?: string
   provider?: string
+  imageDataUri?: string
+  maskDataUri?: string
+  mode?: 'generate' | 'edit' | 'inpaint'
 }
 
 export interface ImageGenProviders {

package/src/renderer/shared/utils/approval.tsx

@@ -13,11 +13,30 @@ export interface ApprovalDetails {
   isDangerous: boolean
 }
 
+function parseDiffStats(diffText?: string): string {
+  if (!diffText) return ''
+  let additions = 0
+  let deletions = 0
+  const lines = diffText.split('\n')
+  for (const line of lines) {
+    if (line.startsWith('+++ ') || line.startsWith('--- ')) continue
+    if (line.startsWith('+')) additions++
+    else if (line.startsWith('-')) deletions++
+  }
+  return `(+${additions} -${deletions})`
+}
+
 // eslint-disable-next-line @typescript-eslint/no-explicit-any
 export function renderApprovalDetails(approval: any): ApprovalDetails {
   if (!approval) return { title: 'Action Pending Approval', body: 'No action details available.', isDangerous: false }
-  if (approval.WriteFile) return { title: 'Write File', body: `Path: ${approval.WriteFile.path}`, reason: approval.WriteFile.diff ? `Diff:\n${approval.WriteFile.diff}` : 'Writing new file content.', isDangerous: false }
-  if (approval.ApplyPatch) return { title: 'Apply Patch', body: `Path: ${approval.ApplyPatch.path}`, reason: approval.ApplyPatch.diff ? `Diff:\n${approval.ApplyPatch.diff}` : 'Applying code patch.', isDangerous: false }
+  if (approval.WriteFile) {
+    const stats = parseDiffStats(approval.WriteFile.diff)
+    return { title: 'Write File', body: `Path: ${approval.WriteFile.path}${stats ? ` ${stats}` : ''}`, reason: approval.WriteFile.diff ? `Diff:\n${approval.WriteFile.diff}` : 'Writing new file content.', isDangerous: false }
+  }
+  if (approval.ApplyPatch) {
+    const stats = parseDiffStats(approval.ApplyPatch.diff)
+    return { title: 'Apply Patch', body: `Path: ${approval.ApplyPatch.path}${stats ? ` ${stats}` : ''}`, reason: approval.ApplyPatch.diff ? `Diff:\n${approval.ApplyPatch.diff}` : 'Applying code patch.', isDangerous: false }
+  }
   if (approval.RunShell) return { title: 'Run Shell Command', body: approval.RunShell.command, reason: 'Executing shell commands can modify your system.', isDangerous: true }
   if (approval.NoteWrite) return { title: 'Write Note', body: `Path: ${approval.NoteWrite.path}`, reason: 'Creating or updating workspace notes.', isDangerous: false }
   if (approval.RunPlugin) return { title: `Run Plugin: ${approval.RunPlugin.name}`, body: approval.RunPlugin.instruction, reason: 'Executing a native plugin action.', isDangerous: false }
@@ -30,9 +49,35 @@ export function renderApprovalDetails(approval: any): ApprovalDetails {
   return { title: 'Unknown Action', body: JSON.stringify(approval, null, 2), reason: 'Requires approval to proceed.', isDangerous: false }
 }
 
+function parseHunkHeader(line: string): { oldStart: number; newStart: number } | null {
+  const trimmed = line.trim()
+  if (!trimmed.startsWith('@@')) return null
+  const parts = trimmed.split('@@')
+  if (parts.length < 3) return null
+  const headerBody = parts[1].trim()
+  let oldStart = 1
+  let newStart = 1
+
+  for (const token of headerBody.split(/\s+/)) {
+    if (token.startsWith('-')) {
+      const numStr = token.slice(1).split(',')[0]
+      const parsed = parseInt(numStr, 10)
+      if (!isNaN(parsed)) oldStart = parsed
+    } else if (token.startsWith('+')) {
+      const numStr = token.slice(1).split(',')[0]
+      const parsed = parseInt(numStr, 10)
+      if (!isNaN(parsed)) newStart = parsed
+    }
+  }
+  return { oldStart, newStart }
+}
+
 export function renderDiff(diffText: string): ReactNode {
   if (!diffText) return null
   const lines = diffText.split('\n')
+  let currentOldLine = 1
+  let currentNewLine = 1
+
   return (
     <div style={{
       background: '#0b0f19',
@@ -46,18 +91,52 @@ export function renderDiff(diffText: string): ReactNode {
       lineHeight: '1.4',
     }}>
       {lines.map((line, idx) => {
+        if (line.startsWith('@@')) {
+          const hunk = parseHunkHeader(line)
+          if (hunk) {
+            currentOldLine = hunk.oldStart
+            currentNewLine = hunk.newStart
+          }
+          return (
+            <div key={idx} style={{ color: '#64748b', fontWeight: 'bold', padding: '2px 6px' }}>
+              {'       ' + line}
+            </div>
+          )
+        }
+        if (line.startsWith('---') || line.startsWith('+++')) {
+          return (
+            <div key={idx} style={{ color: '#64748b', fontWeight: 'bold', padding: '2px 6px' }}>
+              {'       ' + line}
+            </div>
+          )
+        }
+
+        let lineNum = ''
         // eslint-disable-next-line @typescript-eslint/no-explicit-any
-        let style: any = { whiteSpace: 'pre-wrap', padding: '2px 6px' }
-        if (line.startsWith('@@') || line.startsWith('---') || line.startsWith('+++')) {
-          style = { ...style, color: '#64748b', fontWeight: 'bold' }
-        } else if (line.startsWith('+')) {
+        let style: any = { whiteSpace: 'pre-wrap', padding: '2px 6px', display: 'flex', alignItems: 'center' }
+        if (line.startsWith('+')) {
+          lineNum = String(currentNewLine)
+          currentNewLine++
           style = { ...style, background: 'rgba(16, 185, 129, 0.12)', borderLeft: '3px solid #10b981', color: '#a7f3d0' }
         } else if (line.startsWith('-')) {
+          lineNum = String(currentOldLine)
+          currentOldLine++
           style = { ...style, background: 'rgba(239, 68, 68, 0.12)', borderLeft: '3px solid #ef4444', color: '#fca5a5' }
         } else {
+          lineNum = String(currentNewLine)
+          currentOldLine++
+          currentNewLine++
           style = { ...style, color: '#e2e8f0' }
         }
-        return <div key={idx} style={style}>{line}</div>
+
+        return (
+          <div key={idx} style={style}>
+            <span style={{ color: '#64748b', marginRight: '12px', userSelect: 'none', display: 'inline-block', width: '36px', textAlign: 'right', flexShrink: 0 }}>
+              {lineNum}
+            </span>
+            <span>{line}</span>
+          </div>
+        )
       })}
     </div>
   )