Conversation
- Create `mastra-reference.md` to document the Mastra backend structure, including agents, tools, networks, workflows, configuration, A2A coordination, MCP server, evaluations, services, data processing, and policy definitions. - Create `reference.md` to provide an extended reference for AI elements, agent configuration schema, and API routes for the dashboard, linking to the newly created `mastra-reference.md`. - Implement `MainSidebar` component in `main-sidebar.tsx` to enhance the user interface with navigation for agents, tools, workflows, and observability features. - Fetch and display agents, tools, workflows, traces, threads, and vector indices from the backend. - Include collapsible sections for better organization of sidebar items. - Add functionality to select agents and create new threads directly from the sidebar.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Reviewer's GuideAdds comprehensive AI SDK UI and Mastra backend documentation plus a new MainSidebar UI for the chat page, and enhances server/API-side integrations for multi-tenant streaming, threads, memory, and observability data in the sidebar. Sequence diagram for updated AI streaming completion and chat-extra routessequenceDiagram
actor User
participant ChatUI as ChatUI_frontend
participant ApiCompletion as NextApi_completion
participant ApiChatExtra as NextApi_chat_extra
participant Mastra as Mastra_agent
participant AISDK as AISDK_bridge
rect rgb(245,245,245)
User->>ChatUI: Submit prompt or messages
ChatUI->>ApiCompletion: POST /api/completion {prompt,agentId,threadId,resourceId,memory,data}
end
rect rgb(235,245,255)
ApiCompletion->>ApiCompletion: Parse body
ApiCompletion->>ApiCompletion: Select agentId (agentId > data.agentId > weatherAgent)
ApiCompletion->>Mastra: mastra.getAgent(selectedAgentId)
Mastra-->>ApiCompletion: agent
ApiCompletion->>ApiCompletion: Build RequestContext
ApiCompletion->>ApiCompletion: set MASTRA_RESOURCE_ID_KEY, MASTRA_THREAD_ID_KEY
ApiCompletion->>ApiCompletion: Merge remaining data into RequestContext
ApiCompletion->>Mastra: agent.stream(messages, {threadId,resourceId,memory,requestContext})
Mastra-->>ApiCompletion: stream (Mastra stream)
ApiCompletion->>AISDK: toAISdkStream(stream,{from:agent,sendReasoning:true,sendSources:true})
ApiCompletion->>AISDK: createUIMessageStream({originalMessages,execute})
AISDK-->>ApiCompletion: uiStream
ApiCompletion-->>ChatUI: createUIMessageStreamResponse({stream:uiStream})
ChatUI-->>User: Incremental UIMessage updates (text,reasoning,tools,sources)
end
rect rgb(245,245,255)
ChatUI->>ApiChatExtra: POST /api/chat-extra {messages,data,id,threadId,resourceId,memory}
ApiChatExtra->>ApiChatExtra: Select agentId (data.agentId > id > weatherAgent)
ApiChatExtra->>Mastra: mastra.getAgent(agentId)
Mastra-->>ApiChatExtra: agent
ApiChatExtra->>ApiChatExtra: Build RequestContext with resourceId,threadId and extra data
ApiChatExtra->>Mastra: agent.stream(messages,{threadId,resourceId,memory,requestContext})
Mastra-->>ApiChatExtra: stream
ApiChatExtra->>AISDK: toAISdkStream(stream,{from:agent,sendReasoning:true,sendSources:true})
ApiChatExtra->>AISDK: createUIMessageStream(...)
AISDK-->>ApiChatExtra: uiStream
ApiChatExtra-->>ChatUI: createUIMessageStreamResponse({stream:uiStream})
ChatUI-->>User: Streamed assistant response
end
Class diagram for MainSidebar component and sidebar actionsclassDiagram
class MainSidebar {
+SidebarAgent[] agents
+SidebarTool[] tools
+SidebarWorkflow[] workflows
+any[] networks
+SidebarTrace[] traces
+SidebarThread[] threads
+SidebarVectorIndex[] vectors
+boolean loading
+string error
+MainSidebar()
+handleAgentClick(agentId string) void
+handleNavClick(href string) void
}
class ChatContextHook {
+string selectedAgent
+function selectAgent(agentId string) void
+function setThreadId(threadId string) void
}
class SidebarActionsModule {
+function fetchAgents() SidebarAgent[]
+function fetchTools() SidebarTool[]
+function fetchWorkflows() SidebarWorkflow[]
+function fetchTraces(limit number) SidebarTrace[]
+function fetchThreads(resourceId string, agentId string) SidebarThread[]
+function fetchVectors() SidebarVectorIndex[]
+function fetchProcessors() SidebarProcessor[]
+function fetchScorers() SidebarScorer[]
+function createThread(agentId string, title string, resourceId string) any
+function deleteThread(agentId string, threadId string) any
+function updateWorkingMemory(agentId string, threadId string, workingMemory string, resourceId string) any
+function saveMessage(agentId string, threadId string, messages any[]) any
}
class MastraClientWrapper {
+function listAgents() any
+function listTools() any
+function listWorkflows() any
+function listMemoryThreads(resourceId string) any
+function listTraces(limit number) any
+function listVectors() any
+function listProcessors() any
+function listScorers() any
+function createMemoryThread(agentId string, title string, resourceId string) any
+function getMemoryThread(agentId string, threadId string) any
+function updateWorkingMemory(agentId string, threadId string, workingMemory string, resourceId string) any
+function saveMessageToMemory(agentId string, messages any[]) any
}
MainSidebar --> ChatContextHook : uses
MainSidebar --> SidebarActionsModule : calls
SidebarActionsModule --> MastraClientWrapper : uses
File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
|
🤖 Hi @ssdeanx, I've received your request, and I'm working on it now! You can track my progress in the logs for more details. |
|
Caution Review failedThe pull request is closed. Summary by CodeRabbit
WalkthroughAdds extensive AI SDK UI documentation, implements multi-tenant agent selection and UI-stream bridging in chat/completion routes, introduces a data-backed MainSidebar component and related sidebar actions, and adjusts chat page layout and storage metadata. No public API type declarations were removed. Changes
Sequence Diagram(s)sequenceDiagram
participant Client as Client (UI)
participant API as API Route (completion/chat-extra)
participant Agent as Mastra Agent
participant Bridge as Stream Bridge (toAISdkStream / createUIMessageStream)
Client->>API: POST { agentId?, threadId?, resourceId?, messages }
API->>API: resolve agentId (param | body | default)
API->>Agent: agent.stream({ messages, threadId, resourceId, requestContext, memory })
Agent-->>API: streaming chunks (ReadableStream or AsyncIterable)
API->>Bridge: wrap/convert agent stream -> UIMessage stream
Bridge-->>Client: createUIMessageStreamResponse(stream) (UIMessage protocol)
Client-->>Client: render partial UI messages as they arrive
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Possibly related PRs
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.
Contributor
Summary of ChangesHello @ssdeanx, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request introduces a new, interactive main sidebar component that provides comprehensive navigation and access to various backend entities such as agents, tools, and workflows. It also includes extensive documentation for the Mastra backend and AI Elements, improving clarity and maintainability. Furthermore, the API routes for chat and completion have been refactored to support dynamic agent selection and robust stream handling, while new server actions have been added to facilitate richer interaction with backend data from the UI. Highlights
Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
|
🤖 I'm sorry @ssdeanx, but I was unable to process your request. Please see the logs for more details. |
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
There was a problem hiding this comment.
Hey - I've found 1 issue, and left some high level feedback:
- The new
MainSidebarcomponent duplicates entire sidebar groups (e.g.ResourcesandObservabilitysections appear twice with identical content); consider consolidating these to a single instance per section to simplify the UI and maintenance. MainSidebarmaintainsnetworksandvectorsstate but never loads networks (nofetchNetworkscall) and never renders vectors, so either wire these up to real data/sections or remove the unused state to avoid confusion.- In
fetchThreads, the second parameteragentIdis unused, and several new sidebar actions (createThread,deleteThread,updateWorkingMemory,saveMessage) hard-code defaults like'default-resource'; either use the passed parameters consistently or remove them/defaults to avoid incorrect multi-tenancy behavior.
Prompt for AI Agents
Please address the comments from this code review:
## Overall Comments
- The new `MainSidebar` component duplicates entire sidebar groups (e.g. `Resources` and `Observability` sections appear twice with identical content); consider consolidating these to a single instance per section to simplify the UI and maintenance.
- `MainSidebar` maintains `networks` and `vectors` state but never loads networks (no `fetchNetworks` call) and never renders vectors, so either wire these up to real data/sections or remove the unused state to avoid confusion.
- In `fetchThreads`, the second parameter `agentId` is unused, and several new sidebar actions (`createThread`, `deleteThread`, `updateWorkingMemory`, `saveMessage`) hard-code defaults like `'default-resource'`; either use the passed parameters consistently or remove them/defaults to avoid incorrect multi-tenancy behavior.
## Individual Comments
### Comment 1
<location> `app/chat/components/main-sidebar.tsx:172-181` </location>
<code_context>
+ <SidebarGroup>
</code_context>
<issue_to_address>
**suggestion:** The `Resources` and `Observability` sections are duplicated, which complicates the UI and state.
There are two `Resources` groups (Workflows/Networks/Tools) and two `Observability` groups (Recent Traces). This duplication increases maintenance cost and risks the two copies diverging in behavior. Consider keeping a single instance of each or extracting them into reusable components instead of copy-pasting.
Suggested implementation:
```typescript
const ResourcesSidebarGroup = () => (
<SidebarGroup>
<SidebarGroupLabel>Resources</SidebarGroupLabel>
<SidebarGroupContent>
<SidebarMenu>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/workflows">
<span>Workflows</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/networks">
<span>Networks</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/tools">
<span>Tools</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
</SidebarMenu>
</SidebarGroupContent>
</SidebarGroup>
);
const ObservabilitySidebarGroup = () => (
<SidebarGroup>
<SidebarGroupLabel>Observability</SidebarGroupLabel>
<SidebarGroupContent>
<SidebarMenu>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/traces/recent">
<span>Recent Traces</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
</SidebarMenu>
</SidebarGroupContent>
</SidebarGroup>
);
return (
<Sidebar className="border-r">
```
```typescript
<ResourcesSidebarGroup />
```
```typescript
<ObservabilitySidebarGroup />
```
```typescript
<ResourcesSidebarGroup />
```
```typescript
<ObservabilitySidebarGroup />
```
Because I only see a small portion of the file, please:
1. Adjust the four SEARCH blocks that replace inline `Resources` / `Observability` groups so they match the exact JSX currently in your file. The important part is that you:
- Move the **first** `Resources` group JSX into `ResourcesSidebarGroup`.
- Move the **first** `Observability` group JSX into `ObservabilitySidebarGroup`.
2. Remove the duplicated second `Resources` and `Observability` groups and replace them with `<ResourcesSidebarGroup />` and `<ObservabilitySidebarGroup />` respectively, as shown.
3. Ensure `Link` is already imported (likely from `next/link`) at the top of the file; if your existing implementation uses a different navigation component or different hrefs, reuse those instead of the example hrefs I provided.
</issue_to_address>Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.
Comment on lines
+172
to
+181
| <SidebarGroup> | ||
| <SidebarGroupLabel>Agents</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible | ||
| defaultOpen={true} | ||
| className="group/collapsible" | ||
| > | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> |
There was a problem hiding this comment.
suggestion: The Resources and Observability sections are duplicated, which complicates the UI and state.
There are two Resources groups (Workflows/Networks/Tools) and two Observability groups (Recent Traces). This duplication increases maintenance cost and risks the two copies diverging in behavior. Consider keeping a single instance of each or extracting them into reusable components instead of copy-pasting.
Suggested implementation:
const ResourcesSidebarGroup = () => (
<SidebarGroup>
<SidebarGroupLabel>Resources</SidebarGroupLabel>
<SidebarGroupContent>
<SidebarMenu>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/workflows">
<span>Workflows</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/networks">
<span>Networks</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/tools">
<span>Tools</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
</SidebarMenu>
</SidebarGroupContent>
</SidebarGroup>
);
const ObservabilitySidebarGroup = () => (
<SidebarGroup>
<SidebarGroupLabel>Observability</SidebarGroupLabel>
<SidebarGroupContent>
<SidebarMenu>
<SidebarMenuItem>
<SidebarMenuButton asChild>
<Link href="/traces/recent">
<span>Recent Traces</span>
</Link>
</SidebarMenuButton>
</SidebarMenuItem>
</SidebarMenu>
</SidebarGroupContent>
</SidebarGroup>
);
return (
<Sidebar className="border-r"> <ResourcesSidebarGroup /> <ObservabilitySidebarGroup /> <ResourcesSidebarGroup /> <ObservabilitySidebarGroup />Because I only see a small portion of the file, please:
- Adjust the four SEARCH blocks that replace inline
Resources/Observabilitygroups so they match the exact JSX currently in your file. The important part is that you:- Move the first
Resourcesgroup JSX intoResourcesSidebarGroup. - Move the first
Observabilitygroup JSX intoObservabilitySidebarGroup.
- Move the first
- Remove the duplicated second
ResourcesandObservabilitygroups and replace them with<ResourcesSidebarGroup />and<ObservabilitySidebarGroup />respectively, as shown. - Ensure
Linkis already imported (likely fromnext/link) at the top of the file; if your existing implementation uses a different navigation component or different hrefs, reuse those instead of the example hrefs I provided.
Contributor
There was a problem hiding this comment.
Pull request overview
This pull request adds comprehensive reference documentation for the AI SDK UI and Mastra backend integration, along with a new MainSidebar component to enhance navigation in the chat interface. The PR provides detailed documentation covering AI elements, agent configuration, API routes, and the Mastra backend structure.
Changes:
- Added extensive reference documentation (reference.md, mastra-reference.md, ai-elements-reference.md, SKILL.md) providing detailed guidance for AI SDK UI integration, Mastra backend structure, and ai-elements components
- Implemented MainSidebar component for displaying agents, tools, workflows, and observability features with collapsible navigation sections
- Updated API route implementations (completion, chat-extra) with enhanced multi-tenancy support via RequestContext and improved stream handling
- Updated API documentation (AGENTS.md) to reflect current route structure and patterns
Reviewed changes
Copilot reviewed 12 out of 13 changed files in this pull request and generated 10 comments.
Show a summary per file
| File | Description |
|---|---|
app/chat/components/main-sidebar.tsx |
New navigation sidebar component with agent selection, resource display, and new thread functionality |
app/chat/actions/sidebar-actions.ts |
Added server actions for fetching sidebar data (processors, scorers) and managing threads/memory |
app/chat/page.tsx |
Integrated MainSidebar with SidebarProvider for dual-sidebar layout |
app/api/completion/route.ts |
Enhanced with multi-tenancy support, proper message formatting, and robust stream handling |
app/api/chat-extra/route.ts |
Updated with RequestContext integration and improved agent selection logic |
app/api/AGENTS.md |
Simplified documentation focusing on route overview and key implementations |
.agents/skills/ai-sdk-ui-master/*.md |
Comprehensive reference documentation for AI SDK UI, Mastra backend, and ai-elements |
.gitignore |
Added entries for various build artifacts and temporary files |
src/mastra/public/storage/**/*.json |
Runtime Crawlee storage artifacts (should not be committed) |
| const [agents, setAgents] = useState<SidebarAgent[]>([]) | ||
| const [tools, setTools] = useState<SidebarTool[]>([]) | ||
| const [workflows, setWorkflows] = useState<SidebarWorkflow[]>([]) | ||
| const [networks, setNetworks] = useState<any[]>([]) |
There was a problem hiding this comment.
The networks state variable is defined but always remains empty because there's no corresponding fetch function call in the loadData useEffect. Either implement a fetchNetworks function and include it in the Promise.all, or remove the networks state and related UI sections if this functionality is not yet ready.
Suggested change
| const [networks, setNetworks] = useState<any[]>([]) |
Comment on lines
1
to
84
| { | ||
| "usableSessionsCount": 4, | ||
| "retiredSessionsCount": 0, | ||
| "sessions": [ | ||
| { | ||
| "id": "session_1dsmUSuHpC", | ||
| "id": "session_m2f8Zbu3Hq", | ||
| "cookieJar": { | ||
| "version": "tough-cookie@6.0.0", | ||
| "storeType": "MemoryCookieStore", | ||
| "rejectPublicSuffixes": true, | ||
| "enableLooseMode": false, | ||
| "allowSpecialUseDomain": true, | ||
| "prefixSecurity": "silent", | ||
| "cookies": [] | ||
| }, | ||
| "userData": {}, | ||
| "maxErrorScore": 3, | ||
| "errorScoreDecrement": 0.5, | ||
| "expiresAt": "2026-02-20T17:58:03.821Z", | ||
| "createdAt": "2026-02-20T17:08:03.821Z", | ||
| "expiresAt": "2026-02-21T16:45:45.229Z", | ||
| "createdAt": "2026-02-21T15:55:45.229Z", | ||
| "usageCount": 1, | ||
| "maxUsageCount": 50, | ||
| "errorScore": 1 | ||
| }, | ||
| { | ||
| "id": "session_7eaqkMpwYQ", | ||
| "id": "session_7theSs4SJL", | ||
| "cookieJar": { | ||
| "version": "tough-cookie@6.0.0", | ||
| "storeType": "MemoryCookieStore", | ||
| "rejectPublicSuffixes": true, | ||
| "enableLooseMode": false, | ||
| "allowSpecialUseDomain": true, | ||
| "prefixSecurity": "silent", | ||
| "cookies": [] | ||
| }, | ||
| "userData": {}, | ||
| "maxErrorScore": 3, | ||
| "errorScoreDecrement": 0.5, | ||
| "expiresAt": "2026-02-20T17:58:07.283Z", | ||
| "createdAt": "2026-02-20T17:08:07.283Z", | ||
| "expiresAt": "2026-02-21T16:45:48.498Z", | ||
| "createdAt": "2026-02-21T15:55:48.498Z", | ||
| "usageCount": 1, | ||
| "maxUsageCount": 50, | ||
| "errorScore": 1 | ||
| }, | ||
| { | ||
| "id": "session_ajGg989YMc", | ||
| "id": "session_FqMubR3jYS", | ||
| "cookieJar": { | ||
| "version": "tough-cookie@6.0.0", | ||
| "storeType": "MemoryCookieStore", | ||
| "rejectPublicSuffixes": true, | ||
| "enableLooseMode": false, | ||
| "allowSpecialUseDomain": true, | ||
| "prefixSecurity": "silent", | ||
| "cookies": [] | ||
| }, | ||
| "userData": {}, | ||
| "maxErrorScore": 3, | ||
| "errorScoreDecrement": 0.5, | ||
| "expiresAt": "2026-02-20T17:58:08.042Z", | ||
| "createdAt": "2026-02-20T17:08:08.042Z", | ||
| "expiresAt": "2026-02-21T16:45:49.261Z", | ||
| "createdAt": "2026-02-21T15:55:49.261Z", | ||
| "usageCount": 0, | ||
| "maxUsageCount": 50, | ||
| "errorScore": 0 | ||
| }, | ||
| { | ||
| "id": "session_NH0pwVJBnf", | ||
| "id": "session_u5j8hh1ZIe", | ||
| "cookieJar": { | ||
| "version": "tough-cookie@6.0.0", | ||
| "storeType": "MemoryCookieStore", | ||
| "rejectPublicSuffixes": true, | ||
| "enableLooseMode": false, | ||
| "allowSpecialUseDomain": true, | ||
| "prefixSecurity": "silent", | ||
| "cookies": [] | ||
| }, | ||
| "userData": {}, | ||
| "maxErrorScore": 3, | ||
| "errorScoreDecrement": 0.5, | ||
| "expiresAt": "2026-02-20T17:58:08.797Z", | ||
| "createdAt": "2026-02-20T17:08:08.797Z", | ||
| "expiresAt": "2026-02-21T16:45:49.997Z", | ||
| "createdAt": "2026-02-21T15:55:49.997Z", | ||
| "usageCount": 1, | ||
| "maxUsageCount": 50, | ||
| "errorScore": 1 | ||
| } |
There was a problem hiding this comment.
These storage JSON files are runtime artifacts from Mastra Crawlee that should not be committed to version control. According to .gitignore lines 143-148, files matching this pattern should already be ignored. These changes should be removed from the PR as they represent temporary runtime state, not source code changes.
Comment on lines
1
to
49
| { | ||
| "requestsFinished": 0, | ||
| "requestsFailed": 1, | ||
| "requestsRetries": 1, | ||
| "requestsFailedPerMinute": 9, | ||
| "requestsFinishedPerMinute": 0, | ||
| "requestMinDurationMillis": null, | ||
| "requestMaxDurationMillis": 0, | ||
| "requestTotalFailedDurationMillis": 135, | ||
| "requestTotalFailedDurationMillis": 60, | ||
| "requestTotalFinishedDurationMillis": 0, | ||
| "crawlerStartedAt": "2026-02-20T17:08:02.592Z", | ||
| "crawlerFinishedAt": "2026-02-20T17:08:08.954Z", | ||
| "statsPersistedAt": "2026-02-20T17:08:08.954Z", | ||
| "crawlerRuntimeMillis": 6466, | ||
| "crawlerLastStartTimestamp": 1771607282488, | ||
| "crawlerStartedAt": "2026-02-21T15:55:43.489Z", | ||
| "crawlerFinishedAt": "2026-02-21T15:55:50.080Z", | ||
| "statsPersistedAt": "2026-02-21T15:55:50.080Z", | ||
| "crawlerRuntimeMillis": 6638, | ||
| "crawlerLastStartTimestamp": 1771689343442, | ||
| "requestRetryHistogram": [ | ||
| null, | ||
| null, | ||
| 1 | ||
| ], | ||
| "statsId": 0, | ||
| "requestAvgFailedDurationMillis": 135, | ||
| "requestAvgFailedDurationMillis": 60, | ||
| "requestAvgFinishedDurationMillis": null, | ||
| "requestTotalDurationMillis": 135, | ||
| "requestTotalDurationMillis": 60, | ||
| "requestsTotal": 1, | ||
| "requestsWithStatusCode": { | ||
| "404": 3 | ||
| }, | ||
| "errors": { | ||
| "file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/cd6ed769-7c2f-48f3-b92e-04b1554bb55e.mjs:602:21": { | ||
| "file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/63e94f59-8415-41f8-b936-2b2d519f5000.mjs:602:21": { | ||
| "HTTP_ERROR": { | ||
| "ScrapingError": { | ||
| "HTTP 404: Not Found (code=HTTP_ERROR)": { | ||
| "count": 1 | ||
| } | ||
| } | ||
| } | ||
| } | ||
| }, | ||
| "retryErrors": { | ||
| "file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/cd6ed769-7c2f-48f3-b92e-04b1554bb55e.mjs:602:21": { | ||
| "file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/63e94f59-8415-41f8-b936-2b2d519f5000.mjs:602:21": { | ||
| "HTTP_ERROR": { | ||
| "ScrapingError": { | ||
| "HTTP 404: Not Found (code=HTTP_ERROR)": { | ||
| "count": 2 | ||
| } | ||
| } | ||
| } | ||
| } |
There was a problem hiding this comment.
This file contains runtime crawler statistics that should not be committed to version control. According to .gitignore line 148, this file should be ignored. Remove this file from the PR as it represents temporary runtime state.
| fetchTools(), | ||
| fetchWorkflows(), | ||
| fetchTraces(), | ||
| fetchThreads(), |
There was a problem hiding this comment.
The fetchThreads() call is made without the required resourceId parameter, which causes the function to always return an empty array (see sidebar-actions.ts line 152). This means threads will never be displayed in the sidebar even if they exist. Either pass a valid resourceId to fetchThreads() or remove the threads state and related UI if this functionality is not yet ready.
Comment on lines
+234
to
+438
| <SidebarGroup> | ||
| <SidebarGroupLabel>Resources</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <WorkflowIcon className="mr-2 size-4" /> | ||
| <span>Workflows</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {workflows.map((wf) => ( | ||
| <SidebarMenuSubItem key={wf.id}> | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {wf.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <NetworkIcon className="mr-2 size-4" /> | ||
| <span>Networks</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {networks.map((net) => ( | ||
| <SidebarMenuSubItem | ||
| key={net.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {net.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <CpuIcon className="mr-2 size-4" /> | ||
| <span>Tools</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {tools.map((tool) => ( | ||
| <SidebarMenuSubItem | ||
| key={tool.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {tool.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> | ||
|
|
||
| <SidebarGroup> | ||
| <SidebarGroupLabel>Observability</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <ActivityIcon className="mr-2 size-4" /> | ||
| <span>Recent Traces</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {traces.map((trace) => ( | ||
| <SidebarMenuSubItem | ||
| key={trace.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {trace.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> | ||
|
|
||
| <SidebarGroup> | ||
| <SidebarGroupLabel>Resources</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <WorkflowIcon className="mr-2 size-4" /> | ||
| <span>Workflows</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {workflows.map((wf) => ( | ||
| <SidebarMenuSubItem key={wf.id}> | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {wf.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <NetworkIcon className="mr-2 size-4" /> | ||
| <span>Networks</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {networks.map((net) => ( | ||
| <SidebarMenuSubItem | ||
| key={net.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {net.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <CpuIcon className="mr-2 size-4" /> | ||
| <span>Tools</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {tools.map((tool) => ( | ||
| <SidebarMenuSubItem | ||
| key={tool.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {tool.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> |
There was a problem hiding this comment.
The "Resources" section (lines 234-318) and "Observability" section (lines 320-352) are duplicated identically at lines 354-438 and 440-472 respectively. This code duplication makes the component harder to maintain and increases the file size unnecessarily. Consider extracting these sections into reusable components or removing the duplicate sections entirely.
Comment on lines
+478
to
+487
| <SidebarMenuButton | ||
| variant="default" | ||
| className="w-full justify-center bg-primary text-primary-foreground hover:bg-primary/90" | ||
| onClick={() => { | ||
| handleAgentClick(selectedAgent) | ||
| }} | ||
| > | ||
| <PlusIcon className="mr-2 size-4" /> | ||
| New Thread | ||
| </SidebarMenuButton> |
There was a problem hiding this comment.
The "New Thread" button attempts to create a new thread for the currently selected agent, but if no agent is selected (selectedAgent is undefined or empty), this will fail or behave unexpectedly. Add a guard condition to disable the button or show an error when no agent is selected.
Comment on lines
147
to
150
| export async function fetchThreads( | ||
| resourceId?: string | ||
| resourceId?: string, | ||
| agentId?: string | ||
| ): Promise<SidebarThread[]> { |
There was a problem hiding this comment.
The agentId parameter is added to the function signature but never used in the implementation. Either utilize this parameter in the listMemoryThreads call (if supported by the API) or remove it from the function signature to avoid confusion.
Comment on lines
130
to
+157
| @@ -141,13 +145,16 @@ export async function fetchTraces(limit = 20): Promise<SidebarTrace[]> { | |||
| } | |||
|
|
|||
| export async function fetchThreads( | |||
| resourceId?: string | |||
| resourceId?: string, | |||
| agentId?: string | |||
| ): Promise<SidebarThread[]> { | |||
| try { | |||
| const result = await mastraClient.listMemoryThreads( | |||
| resourceId ? { resourceId } : undefined | |||
| ) | |||
| return result.threads.map((t) => ({ | |||
| if (!resourceId) return [] | |||
| const result = await mastraClient.listMemoryThreads({ | |||
| resourceId, | |||
| }) | |||
| const threadsList = (result as any).threads || (result as any) || [] | |||
| return threadsList.map((t: any) => ({ | |||
There was a problem hiding this comment.
Multiple uses of 'as any' type assertions throughout the file bypass TypeScript's type safety. This can hide potential runtime errors. Consider defining proper types for API responses or using more specific type guards to handle different response shapes safely.
| } | ||
|
|
||
| // Convert single prompt to messages array with proper UIMessage shape | ||
| const messageId = `msg-${Date.now()}` |
There was a problem hiding this comment.
Using Date.now() for message IDs can lead to collisions if multiple messages are created in rapid succession (within the same millisecond). Consider using a more robust ID generation method like crypto.randomUUID() or the generateId function from the 'ai' package that's already imported elsewhere in the codebase.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 20
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (8)
src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json (1)
1-51:⚠️ Potential issue | 🟡 MinorThis file is already in
.gitignorebut needs to be removed from git history.The
.gitignorefile already contains explicit rules for this path:src/mastra/public/storage/key_value_stores/default/*.json src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.jsonHowever, since the file was committed before these rules were added, it remains in git history. Use
git rm --cached src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.jsonto stop tracking it.Also note: The error entries contain local machine metadata (
C:/Users/ssdsk/AgentStack/...), which violates the guideline against including PII or machine-specific identifiers.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json` around lines 1 - 51, The JSON file src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json was committed before it was added to .gitignore and contains local machine paths (PII); stop tracking it with git rm --cached SDK_CRAWLER_STATISTICS_0.json, commit that change, then rewrite history to remove the file entirely (use git filter-repo or BFG to purge the path from all commits) and force-push the cleaned branch; also sanitize the file contents (remove or replace the local path strings found under the "errors" and "retryErrors" keys) or regenerate the file without machine-specific identifiers before re-adding any non-sensitive artifact.src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.json (2)
1-1:⚠️ Potential issue | 🟡 MinorFile name violates the kebab-case naming convention.
SDK_SESSION_POOL_STATE.jsonuses SCREAMING_SNAKE_CASE. Per coding guidelines, all JSON files must use kebab-case. The correct name would besdk-session-pool-state.json.As per coding guidelines: "Use kebab-case for file names (e.g.,
weather-tool.ts,user-profile.tsx)" for**/*.{ts,tsx,js,jsx,json,md}.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.json` at line 1, Rename the file named SDK_SESSION_POOL_STATE.json to kebab-case 'sdk-session-pool-state.json' and update all references to it across the codebase (imports, require calls, config entries, test fixtures, build scripts) to use the new name; search for the symbol SDK_SESSION_POOL_STATE.json and replace occurrences with sdk-session-pool-state.json to ensure nothing breaks.
1-86:⚠️ Potential issue | 🟠 MajorRemove this committed runtime state file from the repository.
This file is a Crawlee/Apify
SessionPoolstate artifact that should never be committed. The path is already listed in.gitignore, but the file was committed anyway. Two issues apply:
- Session token leakage: The
idfields (session_m2f8Zbu3Hq, etc.) are session identifiers. Once committed, they are permanently in git history even after expiry. Coding guidelines forbid tokens in code/files.- Runtime state in version control: This state will diverge between environments and cause confusing merge conflicts.
Action: Use
git rm --cached src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.jsonto remove the file from tracking. The.gitignoreentry already prevents future commits of this path.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.json` around lines 1 - 86, Remove the committed Crawlee/Apify SessionPool runtime state (SDK_SESSION_POOL_STATE.json) from version control and purge it from future commits: run git rm --cached src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.json and commit the removal, ensure the existing .gitignore entry prevents re-adding, and if you need to remove the sensitive session IDs from history use a history-rewrite tool (git filter-repo or BFG) to strip that file from past commits; do not reintroduce the file content into the repo.app/chat/actions/sidebar-actions.ts (5)
5-7:⚠️ Potential issue | 🟠 MajorDuplicate
MastraClientinstantiation missing retry/backoff configuration.A shared client already exists in
lib/mastra-client.tswithretries: 3,backoffMs: 300,maxBackoffMs: 5000, an abort signal, andcredentials: 'same-origin'. This local instance has none of those, degrading resilience for all sidebar actions. Import the shared client instead.Proposed fix
-import { MastraClient } from '@mastra/client-js' - -const mastraClient = new MastraClient({ - baseUrl: process.env.NEXT_PUBLIC_MASTRA_API_URL || 'http://localhost:4111', -}) +import { mastraClient } from '@/lib/mastra-client'Also, Line 6 uses
||instead of??for the fallback, which would coerce empty strings to the default. As per coding guidelines, "Use nullish coalescing??for defaults instead of||to properly handle falsy values".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@app/chat/actions/sidebar-actions.ts` around lines 5 - 7, Replace the local MastraClient instantiation in sidebar-actions.ts with the shared exported client from lib/mastra-client.ts so you reuse its retries, backoffMs, maxBackoffMs, abort signal handling and credentials: 'same-origin' configuration; remove the new MastraClient({...}) created in this file and import the existing client instead (refer to the shared client export in lib/mastra-client.ts and the local instantiation symbol new MastraClient in sidebar-actions.ts), and while doing so change the fallback operator for the base URL from || to ?? to avoid coercing empty strings to the default.
199-232:⚠️ Potential issue | 🟠 Major
transportIdparameter is accepted but ignored.Line 200 declares
transportId?: string, but Line 204 hardcodes'MastraLogger'. Callers passing a value will get silently ignored results.Either use the parameter or remove it:
Proposed fix
export async function fetchLogs( transportId?: string ): Promise<SidebarLogEntry[]> { try { const logs = await mastraClient.listLogs({ - transportId: 'MastraLogger', + transportId: transportId ?? 'MastraLogger', })🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@app/chat/actions/sidebar-actions.ts` around lines 199 - 232, The fetchLogs function currently ignores its transportId parameter and always passes the hardcoded 'MastraLogger' to mastraClient.listLogs; update the mastraClient.listLogs call inside fetchLogs to use the incoming transportId when provided (e.g. pass transportId ?? 'MastraLogger') so callers' values are respected, and keep the existing fallback to 'MastraLogger' if transportId is undefined; ensure the change affects the call site where mastraClient.listLogs is invoked and no other logic needs modification.
67-92: 🧹 Nitpick | 🔵 TrivialPervasive
as any/as unknown as Record<string, unknown>casts throughout the file.Nearly every fetch function casts the API response through
as anyoras unknown as Record<string, unknown>, defeating TypeScript's safety. This is a systemic pattern acrossfetchAgents,fetchTools,fetchTraces,fetchThreads,fetchVectors,fetchLogs,fetchProcessors, andfetchScorers.Consider creating a small set of type-narrowing helpers (e.g.,
isRecord(v): v is Record<string, unknown>) or defining expected response shapes from@mastra/client-jsto reduce theanysurface. As per coding guidelines, "Do not use implicitanytypes — useunknownor proper types instead".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@app/chat/actions/sidebar-actions.ts` around lines 67 - 92, fetchAgents (and sibling functions like fetchTools, fetchTraces, fetchThreads, fetchVectors, fetchLogs, fetchProcessors, fetchScorers) currently force-unwrap API results using "as unknown as Record<string, unknown>" which defeats TS safety; replace these casts by declaring minimal response interfaces for the expected shapes (e.g., AgentRecord with modelId|model|name|description) and/or add small type-narrowing helpers (e.g., isRecord(v): v is Record<string, unknown>) and use them to safely extract fields from the mastraClient.listAgents() return value inside fetchAgents (and analogous code in the other fetch* functions) so you validate types at runtime and remove all "as any"/"as unknown as Record..." casts.
88-91: 🧹 Nitpick | 🔵 TrivialReplace
console.errorwith structured logging utility.Every catch block in this file uses
console.error. The coding guidelines specify using thelogutility fromsrc/mastra/config/loggerfor structured logging. This applies to all 14 error handlers in this file.As per coding guidelines, "Use
logutility fromsrc/mastra/config/loggerfor structured logging instead ofconsole.log".Also applies to: 102-105, 116-119, 141-144, 163-166, 193-196, 228-231, 240-243, 269-272, 284-287, 302-305, 313-316, 333-336, 359-362
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@app/chat/actions/sidebar-actions.ts` around lines 88 - 91, Replace all console.error usages in this file's catch blocks with the structured log utility: import log from "src/mastra/config/logger" at the top, then in each catch (for example in the fetchAgents function and the other 13 catch blocks referenced) call log.error with a clear message and the caught error object (e.g., log.error('fetchAgents error', { error }) ), preserving the original return behavior; do this for every catch block listed (lines around 88-91, 102-105, 116-119, 141-144, 163-166, 193-196, 228-231, 240-243, 269-272, 284-287, 302-305, 313-316, 333-336, 359-362).
147-167:⚠️ Potential issue | 🟠 MajorUnused
agentIdparameter and unsafe fallback onthreadsList.Two issues:
agentId(Line 149) is accepted but never referenced in the function body. This is misleading for callers.- Line 156:
(result as any).threads || (result as any) || []— ifresultis a truthy non-array object (e.g.{}), the fallback resolves to that object and.map()on Line 157 will throwTypeError: threadsList.map is not a function.Proposed fix
export async function fetchThreads( - resourceId?: string, - agentId?: string + resourceId?: string ): Promise<SidebarThread[]> { try { if (!resourceId) return [] const result = await mastraClient.listMemoryThreads({ resourceId, }) - const threadsList = (result as any).threads || (result as any) || [] + const raw = (result as Record<string, unknown>).threads ?? result + const threadsList = Array.isArray(raw) ? raw : [] return threadsList.map((t: any) => ({🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@app/chat/actions/sidebar-actions.ts` around lines 147 - 167, fetchThreads currently accepts an unused agentId and unsafely falls back to a non-array result, causing .map to throw; update the function to include agentId in the mastraClient.listMemoryThreads call when provided (e.g., pass { resourceId, agentId } or only include agentId if defined) and replace the unsafe fallback with a robust array check such as: const threadsList = Array.isArray((result as any).threads) ? (result as any).threads : Array.isArray(result) ? (result as any) : []; then continue mapping as before so .map is only called on an actual array.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.agents/skills/ai-sdk-ui-master/ai-elements-reference.md:
- Line 235: The heading level jumps from "# Part 3" (h1) to "### Backend →
stream toolName → UI mapping" (h3), violating the MD001 rule; update the "###
Backend → stream toolName → UI mapping" heading to "## Backend → stream toolName
→ UI mapping" or insert an intermediate "##" heading between "Part 3" and that
heading so the document follows proper incremental heading order (refer to the
"Part 3" and "Backend → stream toolName → UI mapping" headings to locate the
change).
- Around line 209-223: The markdown table header "File | Exports | Purpose |
Section" has only a 3-column separator row; update the separator to match four
columns (e.g., add a fourth |-----| cell) and ensure every data row (e.g., rows
listing loader.tsx, context.tsx, sources.tsx, snippet.tsx, commit.tsx,
package-info.tsx) contains four pipe-separated columns or move trailing inline
comments (like "This is for Web" / "This is for Coding") out of the table cells;
specifically edit the header/separator around the "File | Exports | Purpose |
Section" row and the separator row below it so the table columns align.
In @.agents/skills/ai-sdk-ui-master/reference.md:
- Line 115: Remove the informal developer note on the agentId line and replace
it with a concise, accurate comment (or remove it entirely) that reflects the
actual behavior of the agent selection expression `data?.agentId ?? id ??
'weatherAgent'`; reference the `agentId` binding and the fallback
`'weatherAgent'` so readers understand this implements dynamic agent selection
via `data?.agentId` or `id` with a default, and do not include personal
commentary or inaccurate claims.
In @.gitignore:
- Around line 165-173: Consolidate the repetitive .gitignore entries by
replacing the multiple specific files under
.kiro/specs/comprehensive-codebase-spec and the many CONTINUITY session files
with two glob/directory patterns: one that ignores the entire
comprehensive-codebase-spec directory and one that matches all CONTINUITY_ses_*
.md files (e.g., CONTINUITY_ses_*.md) under the thoughts/ledgers directory;
remove the individual file lines so future session files are automatically
ignored.
In `@app/api/AGENTS.md`:
- Line 161: The fenced code block in AGENTS.md (the directory tree starting with
"app/api/") is missing a language specifier and triggers MD040; update that code
fence to include a language identifier (use "text" or "plaintext") so the block
begins with ```text (or ```plaintext) before the directory listing to satisfy
markdownlint.
In `@app/api/chat-extra/route.ts`:
- Around line 1-93: This file duplicates agent selection
(agentId/mastra.getAgent), RequestContext setup (new RequestContext,
MASTRA_RESOURCE_ID_KEY/MASTRA_THREAD_ID_KEY and merging data), streamOptions
construction, and the ReadableStream/AsyncIterable bridge used inside
createUIMessageStream.execute; replace that duplicated logic by using the
reusable helper in lib/client-stream-to-ai-sdk.ts. Concretely: remove the inline
agent selection/RequestContext/streamOptions and the toAISdkStream
reader/for-await bridge inside POST, import and call the helper from
lib/client-stream-to-ai-sdk.ts (alongside
createUIMessageStream/createUIMessageStreamResponse) to obtain the ai-to-ui
streaming behavior, and keep POST focused on extracting body/messages and
passing agent, messages, and the constructed streamOptions to the helper; mirror
the same change in the other routes (app/api/chat/route.ts and
app/api/completion/route.ts) so all three reuse the helper.
- Around line 11-13: The POST handler currently calls await req.json() without
try/catch and destructures messages directly (used later as UIMessage[] for
agent.stream()), so add input validation: wrap req.json() in try/catch to handle
malformed JSON and validate the parsed body using a Zod schema (e.g., schema
with messages: array().min(1) and required fields id/data) and use
schema.safeParse to return a clear 400 error when validation fails; ensure you
only call agent.stream() with the validated messages array (not the raw cast
UIMessage[]) and return meaningful error responses on parse/validation failure.
In `@app/api/completion/route.ts`:
- Around line 86-98: The code uses "as any" to access getReader on aiStream and
to iterate, which bypasses typing; add a proper type-narrowing guard (e.g., an
isReadableStream(value: unknown): value is ReadableStream<Uint8Array> or
similar) that checks typeof value === 'object' && value !== null && typeof
(value as any).getReader === 'function', then treat aiStream under that guard as
a ReadableStream and call getReader without "as any"; likewise use a separate
guard like isAsyncIterable(value: unknown): value is AsyncIterable<Uint8Array>
(checking Symbol.asyncIterator in the value) before using for-await to avoid
implicit any and keep the reader/releaseLock and writer.write types correct
(reference symbols/idents: aiStream, getReader, reader, releaseLock, writer,
Symbol.asyncIterator).
- Around line 11-13: Wrap parsing of req.json() in a try/catch in the POST
handler and validate the extracted fields: ensure prompt is a string, trim it,
check it's non-empty and <= 10000 characters, and validate agentId/data shapes
as needed before proceeding; if validation fails return a 400 response with a
clear error message. After validation, use the sanitized prompt when
constructing the UIMessage (the value passed to agent.stream) to avoid undefined
messages. Ensure any parsing or runtime errors are caught and return a 500
response (or appropriate error) rather than allowing exceptions to bubble up.
- Around line 74-103: Wrap the agent.stream(...) call in a try/catch that logs
structured error details (include agentId from the current context,
error.message and error.stack) and returns/propagates a proper error UI response
instead of allowing an unhandled 500; then remove the inline
ReadableStream/AsyncIterable bridge in execute of createUIMessageStream and use
the shared helper from lib/client-stream-to-ai-sdk.ts (the function that
converts the model stream to the AI SDK stream) to consume the stream, passing
the original stream returned by agent.stream, so replace the custom loop around
toAISdkStream with the reusable bridge and ensure errors from the bridge are
also caught and logged.
In `@app/chat/actions/sidebar-actions.ts`:
- Around line 130-139: The mapping that builds tracesList should not substitute
the current time when span.startTime is missing; update the timestamp assignment
inside the tracesList.map (the object produced for each span in
sidebar-actions.ts) to return undefined (or an empty string) instead of new
Date().toISOString() when span.startTime is falsy — e.g. set timestamp to
span.startTime ? String(span.startTime) : undefined so the UI can detect/mask
missing timestamps; keep the rest of the id/name/status/duration mapping
unchanged.
- Around line 339-362: The saveMessage function currently accepts messages:
any[] and uses || defaults that incorrectly override valid falsy values; define
a proper input type/interface (e.g., MessageInput) and change the parameter to
messages: MessageInput[] (or messages: unknown[] validated/cast to
MessageInput[]), then replace all uses of || with nullish coalescing (??) when
building formattedMessages inside saveMessage (reference formattedMessages, m.id
generation, m.resourceId, m.createdAt, m.format), keeping the existing id
fallback logic via optional chaining (crypto.randomUUID?.()) but using ?? for
defaults, and pass the typed formattedMessages to
mastraClient.saveMessageToMemory to restore type safety.
- Around line 290-337: Add explicit return-type annotations to the public action
functions so callers don't have to infer them: declare createThread,
deleteThread and updateWorkingMemory to return a uniform type (e.g. Promise<{
success: true; data: T } | { success: false; error: string }>) or a shared alias
like MemoryActionResult, then update the function signatures for
createThread(agentId: string, title?: string, resourceId?: string):
Promise<MemoryActionResult>, deleteThread(agentId: string, threadId: string):
Promise<MemoryActionResult>, and updateWorkingMemory(agentId: string, threadId:
string, workingMemory: string, resourceId?: string): Promise<MemoryActionResult>
(or the concrete generic) and add the MemoryActionResult type definition in the
module.
In `@app/chat/components/main-sidebar.tsx`:
- Around line 234-472: The file contains duplicated SidebarGroup blocks for
"Resources" and "Observability" (duplicate JSX rendering
Workflows/Networks/Tools and Recent Traces), causing two identical sidebar
sections; remove one of the duplicate groups (either the first or second
occurrence) so only a single SidebarGroup for Resources (with the
Workflows/Networks/Tools Collapsible items) and a single SidebarGroup for
Observability (with the Recent Traces Collapsible) remain—look for the repeated
SidebarGroup JSX that references workflows.map, networks.map, tools.map and
traces.map and delete the redundant block.
- Line 90: The networks state is currently typed as any[] and never populated so
the Networks collapsible sections never render; either remove the networks UI or
implement fetching: define a proper Network interface (avoid implicit any),
change const [networks, setNetworks] from any[] to Network[], add an async
fetchNetworks function and call it inside the existing useEffect (or an
appropriate effect) to setNetworks with the fetched data, and ensure error
handling and typing use unknown where appropriate before narrowing;
alternatively, remove the Networks-related JSX if you opt not to support
networks.
- Around line 99-133: The current useEffect + loadData pattern fetches
agents/tools/workflows/traces/threads/vectors only on mount so the sidebar never
refreshes; replace this with client-side data fetching via SWR or React Query
(e.g., create a useSidebarData hook or individual useSWR/useQuery hooks for
fetchAgents, fetchTools, fetchWorkflows, fetchTraces, fetchThreads,
fetchVectors) to enable caching, automatic revalidation, retry on failure, and
centralized loading/error state, and expose a manual refresh/trigger
(mutate/refetch) that the sidebar UI can call; ensure you update state setters
(setAgents, setTools, setWorkflows, setTraces, setThreads, setVectors) to
consume the SWR/React Query results and remove the mount-only useEffect and
loadData.
- Around line 54-80: The PR contains dead code: remove or wire up the unused
SidebarNavItem type, NAV_ITEMS constant, and handleNavClick function to
eliminate dead code and reduce bundle size; either delete the SidebarNavItem
declaration, the NAV_ITEMS array, and the handleNavClick implementation, or
integrate them into the component render by using NAV_ITEMS to generate the
sidebar links and replacing existing hardcoded nav markup with a map that calls
handleNavClick (or uses hrefs) and ensure the component imports/uses
WorkflowIcon, CpuIcon, ActivityIcon where needed so no referenced symbols remain
unused.
- Around line 82-133: MainSidebar is too large and mixes fetching/state with
rendering; extract the data-loading logic (current useEffect + loadData and
state variables agents, tools, workflows, traces, threads, vectors, loading,
error) into a custom hook (e.g., useSidebarData) that exposes {agents, tools,
workflows, traces, threads, vectors, loading, error, refetch} and move the
Promise.all calls/fetchAgents(), fetchTools(), fetchWorkflows(), fetchTraces(),
fetchThreads(), fetchVectors() into that hook (or replace with SWR/React Query
for caching/revalidation). Also extract repeated collapsible UI patterns from
MainSidebar into a presentational CollapsibleMenuSection component (used for
Agents, Tools, Workflows, Traces, Threads, Vectors) so MainSidebar only
orchestrates hook data, selection handlers (selectAgent, setThreadId) and
renders those components. Ensure unique symbols referenced are MainSidebar,
loadData/useEffect, useSidebarData, CollapsibleMenuSection, and the fetch*
functions when relocating behavior.
- Around line 475-490: The "New Thread" Footer button currently calls
handleAgentClick(selectedAgent) even when selectedAgent is empty; update the
onClick handler (or handleAgentClick) to guard against empty/invalid
selectedAgent by checking selectedAgent !== '' (or null/undefined) before
calling selectAgent and setThreadId, and alternatively disable the
SidebarMenuButton when selectedAgent is falsy so the click cannot fire;
reference SidebarMenuButton onClick, handleAgentClick, selectAgent, setThreadId,
and selectedAgent when making the change.
In
`@src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json`:
- Line 30: The JSON file
src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json
contains a Windows file URI exposing a local username
("file:///C:/Users/ssdsk/...") in the object keys; remove the sensitive file and
scrub history by first deleting it from the index (git rm --cached
src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json),
update the working copy to replace the offending keys (remove or sanitize the
"file:///C:/Users/ssdsk/..." entries) and commit that change, then purge the
file and any matching path strings from repository history using a
history-rewrite tool (git filter-repo or BFG) to remove all occurrences of that
file/PII, and ensure the file remains listed in .gitignore afterwards.
---
Outside diff comments:
In `@app/chat/actions/sidebar-actions.ts`:
- Around line 5-7: Replace the local MastraClient instantiation in
sidebar-actions.ts with the shared exported client from lib/mastra-client.ts so
you reuse its retries, backoffMs, maxBackoffMs, abort signal handling and
credentials: 'same-origin' configuration; remove the new MastraClient({...})
created in this file and import the existing client instead (refer to the shared
client export in lib/mastra-client.ts and the local instantiation symbol new
MastraClient in sidebar-actions.ts), and while doing so change the fallback
operator for the base URL from || to ?? to avoid coercing empty strings to the
default.
- Around line 199-232: The fetchLogs function currently ignores its transportId
parameter and always passes the hardcoded 'MastraLogger' to
mastraClient.listLogs; update the mastraClient.listLogs call inside fetchLogs to
use the incoming transportId when provided (e.g. pass transportId ??
'MastraLogger') so callers' values are respected, and keep the existing fallback
to 'MastraLogger' if transportId is undefined; ensure the change affects the
call site where mastraClient.listLogs is invoked and no other logic needs
modification.
- Around line 67-92: fetchAgents (and sibling functions like fetchTools,
fetchTraces, fetchThreads, fetchVectors, fetchLogs, fetchProcessors,
fetchScorers) currently force-unwrap API results using "as unknown as
Record<string, unknown>" which defeats TS safety; replace these casts by
declaring minimal response interfaces for the expected shapes (e.g., AgentRecord
with modelId|model|name|description) and/or add small type-narrowing helpers
(e.g., isRecord(v): v is Record<string, unknown>) and use them to safely extract
fields from the mastraClient.listAgents() return value inside fetchAgents (and
analogous code in the other fetch* functions) so you validate types at runtime
and remove all "as any"/"as unknown as Record..." casts.
- Around line 88-91: Replace all console.error usages in this file's catch
blocks with the structured log utility: import log from
"src/mastra/config/logger" at the top, then in each catch (for example in the
fetchAgents function and the other 13 catch blocks referenced) call log.error
with a clear message and the caught error object (e.g., log.error('fetchAgents
error', { error }) ), preserving the original return behavior; do this for every
catch block listed (lines around 88-91, 102-105, 116-119, 141-144, 163-166,
193-196, 228-231, 240-243, 269-272, 284-287, 302-305, 313-316, 333-336,
359-362).
- Around line 147-167: fetchThreads currently accepts an unused agentId and
unsafely falls back to a non-array result, causing .map to throw; update the
function to include agentId in the mastraClient.listMemoryThreads call when
provided (e.g., pass { resourceId, agentId } or only include agentId if defined)
and replace the unsafe fallback with a robust array check such as: const
threadsList = Array.isArray((result as any).threads) ? (result as any).threads :
Array.isArray(result) ? (result as any) : []; then continue mapping as before so
.map is only called on an actual array.
In
`@src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json`:
- Around line 1-51: The JSON file
src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json
was committed before it was added to .gitignore and contains local machine paths
(PII); stop tracking it with git rm --cached SDK_CRAWLER_STATISTICS_0.json,
commit that change, then rewrite history to remove the file entirely (use git
filter-repo or BFG to purge the path from all commits) and force-push the
cleaned branch; also sanitize the file contents (remove or replace the local
path strings found under the "errors" and "retryErrors" keys) or regenerate the
file without machine-specific identifiers before re-adding any non-sensitive
artifact.
In
`@src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.json`:
- Line 1: Rename the file named SDK_SESSION_POOL_STATE.json to kebab-case
'sdk-session-pool-state.json' and update all references to it across the
codebase (imports, require calls, config entries, test fixtures, build scripts)
to use the new name; search for the symbol SDK_SESSION_POOL_STATE.json and
replace occurrences with sdk-session-pool-state.json to ensure nothing breaks.
- Around line 1-86: Remove the committed Crawlee/Apify SessionPool runtime state
(SDK_SESSION_POOL_STATE.json) from version control and purge it from future
commits: run git rm --cached
src/mastra/public/storage/key_value_stores/default/SDK_SESSION_POOL_STATE.json
and commit the removal, ensure the existing .gitignore entry prevents re-adding,
and if you need to remove the sensitive session IDs from history use a
history-rewrite tool (git filter-repo or BFG) to strip that file from past
commits; do not reintroduce the file content into the repo.
Comment on lines
+209
to
+223
| | File | Exports | Purpose | Section | | ||
|
|
||
| |------|---------|---------| | ||
| | `loader.tsx` | Loader | Loading indicator | | ||
| | `context.tsx` | Context, ContextContent, ContextItem | Token usage display | | ||
| | `sources.tsx` | Sources, SourcesTrigger, SourcesContent, Source | Citations | **This is for Web** | | ||
| | `plan.tsx` | Plan, PlanHeader, PlanContent, PlanItem | Execution plans | | ||
| | `queue.tsx` | Queue, QueueItem, QueueSection | Task queue | | ||
| | `task.tsx` | Task, TaskTrigger, TaskContent | Task display | | ||
| | `suggestion.tsx` | Suggestions, Suggestion | Input suggestions | | ||
| | `confirmation.tsx` | Confirmation | Confirmation dialog | | ||
| | `shimmer.tsx` | Shimmer | Loading skeleton | | ||
| | `snippet.tsx` | Snippet | Code snippet | **This is for Coding** | | ||
| | `commit.tsx` | Commit, CommitHeader, CommitContent | Git commit display | **This is for Coding** | | ||
| | `package-info.tsx` | PackageInfo | Package metadata | **This is for Coding** | |
There was a problem hiding this comment.
Broken markdown table: header row has 4 columns but separator row has only 3.
Line 209 defines a 4-column header (File | Exports | Purpose | Section), but the separator on Line 211 only has 3 columns. This causes the table to render incorrectly in most markdown parsers — the Section column data on lines 214, 221–223 will be silently dropped or misaligned.
Proposed fix
-| File | Exports | Purpose | Section |
-
-|------|---------|---------|
+| File | Exports | Purpose | Section |
+|------|---------|---------|---------|Also, Line 133 has 4 pipe-separated cells but the table header at Line 128 only has 3 columns. Either add a 4th header column or move the trailing comment out of the table cell:
-| `test-results.tsx` | TestResults, TestSuite, Test, TestStatus, TestName, TestDuration, TestError | Test output | This is for coding....
+| `test-results.tsx` | TestResults, TestSuite, Test, TestStatus, TestName, TestDuration, TestError | Test output |🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.agents/skills/ai-sdk-ui-master/ai-elements-reference.md around lines 209 -
223, The markdown table header "File | Exports | Purpose | Section" has only a
3-column separator row; update the separator to match four columns (e.g., add a
fourth |-----| cell) and ensure every data row (e.g., rows listing loader.tsx,
context.tsx, sources.tsx, snippet.tsx, commit.tsx, package-info.tsx) contains
four pipe-separated columns or move trailing inline comments (like "This is for
Web" / "This is for Coding") out of the table cells; specifically edit the
header/separator around the "File | Exports | Purpose | Section" row and the
separator row below it so the table columns align.
|
|
||
| These are **not** base ai-elements. They are our app-specific components that render Mastra tool results in chat. Each uses **base** ai-elements (e.g. `Tool`, `ToolHeader`, `ToolContent` from `tool.tsx`) and is typed from Mastra tools via `tools/types.ts` (`InferUITool<typeof mastraTool>`). Backend tools live in `src/mastra/tools/`. | ||
|
|
||
| ### Backend → stream toolName → UI mapping |
There was a problem hiding this comment.
Heading level skips from h1 to h3.
Line 229 is # Part 3 (h1), and Line 235 jumps to ### Backend → … (h3), skipping h2. This violates the MD001 heading-increment rule and can break document outline / accessibility navigation.
Change to ## or add an intermediate ## … heading.
🧰 Tools
🪛 markdownlint-cli2 (0.21.0)
[warning] 235-235: Heading levels should only increment by one level at a time
Expected: h2; Actual: h3
(MD001, heading-increment)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.agents/skills/ai-sdk-ui-master/ai-elements-reference.md at line 235, The
heading level jumps from "# Part 3" (h1) to "### Backend → stream toolName → UI
mapping" (h3), violating the MD001 rule; update the "### Backend → stream
toolName → UI mapping" heading to "## Backend → stream toolName → UI mapping" or
insert an intermediate "##" heading between "Part 3" and that heading so the
document follows proper incremental heading order (refer to the "Part 3" and
"Backend → stream toolName → UI mapping" headings to locate the change).
| export async function POST(req: Request) { | ||
| const body = await req.json() | ||
| const { messages, data, id } = body | ||
| const agentId = data?.agentId ?? id ?? 'weatherAgent' // This will only display weatherAgent so this is kinda useless since my mastra/index uses :agentid which makes the route dynamic and since its dynamic u can change agent by putting my other agents.... This would be a static port it must have a string for agent it wont just pick it up. |
There was a problem hiding this comment.
Remove informal developer note from reference documentation.
Line 115 contains a casual comment ("This will only display weatherAgent so this is kinda useless since my mastra/index uses :agentid…") that doesn't belong in reference documentation. It's also inaccurate — the actual app/api/chat/route.ts implementation (shown in the relevant snippets) does implement dynamic agent selection via data?.agentId ?? id ?? 'weatherAgent'.
Replace it with a concise, accurate comment or remove it entirely.
📝 Proposed fix
- const agentId = data?.agentId ?? id ?? 'weatherAgent' // This will only display weatherAgent so this is kinda useless since my mastra/index uses :agentid which makes the route dynamic and since its dynamic u can change agent by putting my other agents.... This would be a static port it must have a string for agent it wont just pick it up.
+ const agentId = data?.agentId ?? id ?? 'weatherAgent' // Dynamic agent selection with fallback📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| const agentId = data?.agentId ?? id ?? 'weatherAgent' // This will only display weatherAgent so this is kinda useless since my mastra/index uses :agentid which makes the route dynamic and since its dynamic u can change agent by putting my other agents.... This would be a static port it must have a string for agent it wont just pick it up. | |
| const agentId = data?.agentId ?? id ?? 'weatherAgent' // Dynamic agent selection with fallback |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.agents/skills/ai-sdk-ui-master/reference.md at line 115, Remove the
informal developer note on the agentId line and replace it with a concise,
accurate comment (or remove it entirely) that reflects the actual behavior of
the agent selection expression `data?.agentId ?? id ?? 'weatherAgent'`;
reference the `agentId` binding and the fallback `'weatherAgent'` so readers
understand this implements dynamic agent selection via `data?.agentId` or `id`
with a default, and do not include personal commentary or inaccurate claims.
Comment on lines
+165
to
+173
| .kiro/specs/comprehensive-codebase-spec/.config.kiro | ||
| .github/observability-setup.md | ||
| .github/testing-strategies.md | ||
| .kiro/specs/comprehensive-codebase-spec/tasks.md | ||
| .kiro/specs/comprehensive-codebase-spec/requirements.md | ||
| .kiro/specs/comprehensive-codebase-spec/design.md | ||
| thoughts/ledgers/CONTINUITY_ses_37ef.md | ||
| thoughts/ledgers/CONTINUITY_ses_37f3.md | ||
| thoughts/ledgers/CONTINUITY_ses_37f1.md |
There was a problem hiding this comment.
🛠️ Refactor suggestion | 🟠 Major
Use glob/directory patterns to replace the growing list of individual file entries.
Two separate patterns are being extended one entry at a time:
.kiro/specs/comprehensive-codebase-spec/entries (lines 165, 168-170): Four individual files under the same directory. A single directory entry covers all of them now and in the future.thoughts/ledgers/CONTINUITY_ses_*.mdentries (lines 171-173): There are now ~17 individually listed session files across the file (lines 149-164 and 171-173). Every new session appends yet another line here.
♻️ Proposed consolidation
Replace lines 165–173 (and the pre-existing session ledger entries at lines 149–164) with:
-thoughts/ledgers/CONTINUITY_ses_399c.md
-thoughts/ledgers/CONTINUITY_ses_3997.md
-thoughts/ledgers/CONTINUITY_ses_394b.md
-thoughts/ledgers/CONTINUITY_ses_394d.md
-thoughts/ledgers/CONTINUITY_ses_394f.md
-thoughts/ledgers/CONTINUITY_ses_3943.md
-thoughts/ledgers/CONTINUITY_ses_3944.md
-thoughts/ledgers/CONTINUITY_ses_38c6.md
-thoughts/ledgers/CONTINUITY_ses_38e5.md
-thoughts/ledgers/CONTINUITY_ses_388a.md
-thoughts/ledgers/CONTINUITY_ses_3885.md
-thoughts/ledgers/CONTINUITY_ses_3886.md
-thoughts/ledgers/CONTINUITY_ses_3889.md
-thoughts/ledgers/CONTINUITY_ses_37ef.md
-thoughts/ledgers/CONTINUITY_ses_37f3.md
-thoughts/ledgers/CONTINUITY_ses_37f1.md
+thoughts/ledgers/CONTINUITY_ses_*.md-.kiro/specs/comprehensive-codebase-spec/.config.kiro
-.kiro/specs/comprehensive-codebase-spec/tasks.md
-.kiro/specs/comprehensive-codebase-spec/requirements.md
-.kiro/specs/comprehensive-codebase-spec/design.md
+.kiro/specs/comprehensive-codebase-spec/🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.gitignore around lines 165 - 173, Consolidate the repetitive .gitignore
entries by replacing the multiple specific files under
.kiro/specs/comprehensive-codebase-spec and the many CONTINUITY session files
with two glob/directory patterns: one that ignores the entire
comprehensive-codebase-spec directory and one that matches all CONTINUITY_ses_*
.md files (e.g., CONTINUITY_ses_*.md) under the thoughts/ledgers directory;
remove the individual file lines so future session files are automatically
ignored.
| interface V0Request { | ||
| prompt: string | ||
| } | ||
| ``` |
There was a problem hiding this comment.
Add language identifier to the file structure code block.
The fenced code block starting at line 161 is missing a language specifier, which triggers the MD040 markdownlint rule. Use text or plaintext for directory tree listings.
📝 Proposed fix
-```
+```text
app/api/📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| ``` |
🧰 Tools
🪛 markdownlint-cli2 (0.21.0)
[warning] 161-161: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@app/api/AGENTS.md` at line 161, The fenced code block in AGENTS.md (the
directory tree starting with "app/api/") is missing a language specifier and
triggers MD040; update that code fence to include a language identifier (use
"text" or "plaintext") so the block begins with ```text (or ```plaintext) before
the directory listing to satisfy markdownlint.
| const [agents, setAgents] = useState<SidebarAgent[]>([]) | ||
| const [tools, setTools] = useState<SidebarTool[]>([]) | ||
| const [workflows, setWorkflows] = useState<SidebarWorkflow[]>([]) | ||
| const [networks, setNetworks] = useState<any[]>([]) |
There was a problem hiding this comment.
networks state is typed any[] and never populated.
networks is initialized as any[] and no fetchNetworks call exists in the useEffect at lines 99–133, so networks will always be an empty array. The Networks collapsible sections (lines 274, 394) will never render any items.
Either add a fetchNetworks call (and a proper interface) or remove the networks section from the sidebar.
As per coding guidelines, "Do not use implicit any types — use unknown or proper types instead".
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@app/chat/components/main-sidebar.tsx` at line 90, The networks state is
currently typed as any[] and never populated so the Networks collapsible
sections never render; either remove the networks UI or implement fetching:
define a proper Network interface (avoid implicit any), change const [networks,
setNetworks] from any[] to Network[], add an async fetchNetworks function and
call it inside the existing useEffect (or an appropriate effect) to setNetworks
with the fetched data, and ensure error handling and typing use unknown where
appropriate before narrowing; alternatively, remove the Networks-related JSX if
you opt not to support networks.
Comment on lines
+99
to
+133
| useEffect(() => { | ||
| const loadData = async () => { | ||
| setLoading(true) | ||
| setError(null) | ||
| try { | ||
| const [ | ||
| agentsData, | ||
| toolsData, | ||
| workflowsData, | ||
| tracesData, | ||
| threadsData, | ||
| vectorsData, | ||
| ] = await Promise.all([ | ||
| fetchAgents(), | ||
| fetchTools(), | ||
| fetchWorkflows(), | ||
| fetchTraces(), | ||
| fetchThreads(), | ||
| fetchVectors(), | ||
| ]) | ||
| setAgents(agentsData) | ||
| setTools(toolsData) | ||
| setWorkflows(workflowsData) | ||
| setTraces(tracesData) | ||
| setThreads(threadsData) | ||
| setVectors(vectorsData) | ||
| } catch (err) { | ||
| setError('Failed to load sidebar data') | ||
| console.error('loadData error:', err) | ||
| } finally { | ||
| setLoading(false) | ||
| } | ||
| } | ||
| loadData() | ||
| }, []) |
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Data fetched on mount only — sidebar never refreshes.
All six fetch calls run once on mount (empty dependency array). If agents, workflows, or traces change during the session (or if a network error occurs on initial load), the sidebar shows stale or empty data indefinitely. Consider using SWR/React Query for automatic revalidation, or at minimum expose a manual refresh mechanism.
As per coding guidelines, "Use SWR or React Query for client-side data fetching and caching".
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@app/chat/components/main-sidebar.tsx` around lines 99 - 133, The current
useEffect + loadData pattern fetches
agents/tools/workflows/traces/threads/vectors only on mount so the sidebar never
refreshes; replace this with client-side data fetching via SWR or React Query
(e.g., create a useSidebarData hook or individual useSWR/useQuery hooks for
fetchAgents, fetchTools, fetchWorkflows, fetchTraces, fetchThreads,
fetchVectors) to enable caching, automatic revalidation, retry on failure, and
centralized loading/error state, and expose a manual refresh/trigger
(mutate/refetch) that the sidebar UI can call; ensure you update state setters
(setAgents, setTools, setWorkflows, setTraces, setThreads, setVectors) to
consume the SWR/React Query results and remove the mount-only useEffect and
loadData.
Comment on lines
+234
to
+472
| <SidebarGroup> | ||
| <SidebarGroupLabel>Resources</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <WorkflowIcon className="mr-2 size-4" /> | ||
| <span>Workflows</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {workflows.map((wf) => ( | ||
| <SidebarMenuSubItem key={wf.id}> | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {wf.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <NetworkIcon className="mr-2 size-4" /> | ||
| <span>Networks</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {networks.map((net) => ( | ||
| <SidebarMenuSubItem | ||
| key={net.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {net.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <CpuIcon className="mr-2 size-4" /> | ||
| <span>Tools</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {tools.map((tool) => ( | ||
| <SidebarMenuSubItem | ||
| key={tool.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {tool.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> | ||
|
|
||
| <SidebarGroup> | ||
| <SidebarGroupLabel>Observability</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <ActivityIcon className="mr-2 size-4" /> | ||
| <span>Recent Traces</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {traces.map((trace) => ( | ||
| <SidebarMenuSubItem | ||
| key={trace.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {trace.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> | ||
|
|
||
| <SidebarGroup> | ||
| <SidebarGroupLabel>Resources</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <WorkflowIcon className="mr-2 size-4" /> | ||
| <span>Workflows</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {workflows.map((wf) => ( | ||
| <SidebarMenuSubItem key={wf.id}> | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {wf.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <NetworkIcon className="mr-2 size-4" /> | ||
| <span>Networks</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {networks.map((net) => ( | ||
| <SidebarMenuSubItem | ||
| key={net.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {net.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <CpuIcon className="mr-2 size-4" /> | ||
| <span>Tools</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {tools.map((tool) => ( | ||
| <SidebarMenuSubItem | ||
| key={tool.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {tool.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> | ||
|
|
||
| <SidebarGroup> | ||
| <SidebarGroupLabel>Observability</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <ActivityIcon className="mr-2 size-4" /> | ||
| <span>Recent Traces</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {traces.map((trace) => ( | ||
| <SidebarMenuSubItem | ||
| key={trace.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {trace.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> |
There was a problem hiding this comment.
Resources and Observability sections are fully duplicated — likely a copy-paste bug.
Lines 234–352 (Resources group with Workflows/Networks/Tools + Observability group with Recent Traces) are repeated verbatim at lines 354–472. This renders two identical sidebar sections to the user.
Remove the duplicate block (either lines 234–352 or lines 354–472).
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@app/chat/components/main-sidebar.tsx` around lines 234 - 472, The file
contains duplicated SidebarGroup blocks for "Resources" and "Observability"
(duplicate JSX rendering Workflows/Networks/Tools and Recent Traces), causing
two identical sidebar sections; remove one of the duplicate groups (either the
first or second occurrence) so only a single SidebarGroup for Resources (with
the Workflows/Networks/Tools Collapsible items) and a single SidebarGroup for
Observability (with the Recent Traces Collapsible) remain—look for the repeated
SidebarGroup JSX that references workflows.map, networks.map, tools.map and
traces.map and delete the redundant block.
Comment on lines
+475
to
+490
| <SidebarFooter className="border-t p-4"> | ||
| <SidebarMenu> | ||
| <SidebarMenuItem> | ||
| <SidebarMenuButton | ||
| variant="default" | ||
| className="w-full justify-center bg-primary text-primary-foreground hover:bg-primary/90" | ||
| onClick={() => { | ||
| handleAgentClick(selectedAgent) | ||
| }} | ||
| > | ||
| <PlusIcon className="mr-2 size-4" /> | ||
| New Thread | ||
| </SidebarMenuButton> | ||
| </SidebarMenuItem> | ||
| </SidebarMenu> | ||
| </SidebarFooter> |
There was a problem hiding this comment.
Footer "New Thread" button may fire with an empty selectedAgent.
If no agent has been selected yet, selectedAgent could be an empty string or a default value. handleAgentClick('') would call selectAgent('') and setThreadId(...) with no valid agent, leading to an invalid URL /chat?agent=.
🛡️ Proposed fix
<SidebarMenuButton
variant="default"
className="w-full justify-center bg-primary text-primary-foreground hover:bg-primary/90"
+ disabled={!selectedAgent}
onClick={() => {
- handleAgentClick(selectedAgent)
+ if (selectedAgent) {
+ handleAgentClick(selectedAgent)
+ }
}}
>📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| <SidebarFooter className="border-t p-4"> | |
| <SidebarMenu> | |
| <SidebarMenuItem> | |
| <SidebarMenuButton | |
| variant="default" | |
| className="w-full justify-center bg-primary text-primary-foreground hover:bg-primary/90" | |
| onClick={() => { | |
| handleAgentClick(selectedAgent) | |
| }} | |
| > | |
| <PlusIcon className="mr-2 size-4" /> | |
| New Thread | |
| </SidebarMenuButton> | |
| </SidebarMenuItem> | |
| </SidebarMenu> | |
| </SidebarFooter> | |
| <SidebarFooter className="border-t p-4"> | |
| <SidebarMenu> | |
| <SidebarMenuItem> | |
| <SidebarMenuButton | |
| variant="default" | |
| className="w-full justify-center bg-primary text-primary-foreground hover:bg-primary/90" | |
| disabled={!selectedAgent} | |
| onClick={() => { | |
| if (selectedAgent) { | |
| handleAgentClick(selectedAgent) | |
| } | |
| }} | |
| > | |
| <PlusIcon className="mr-2 size-4" /> | |
| New Thread | |
| </SidebarMenuButton> | |
| </SidebarMenuItem> | |
| </SidebarMenu> | |
| </SidebarFooter> |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@app/chat/components/main-sidebar.tsx` around lines 475 - 490, The "New
Thread" Footer button currently calls handleAgentClick(selectedAgent) even when
selectedAgent is empty; update the onClick handler (or handleAgentClick) to
guard against empty/invalid selectedAgent by checking selectedAgent !== '' (or
null/undefined) before calling selectAgent and setThreadId, and alternatively
disable the SidebarMenuButton when selectedAgent is falsy so the click cannot
fire; reference SidebarMenuButton onClick, handleAgentClick, selectAgent,
setThreadId, and selectedAgent when making the change.
| }, | ||
| "errors": { | ||
| "file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/cd6ed769-7c2f-48f3-b92e-04b1554bb55e.mjs:602:21": { | ||
| "file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/63e94f59-8415-41f8-b936-2b2d519f5000.mjs:602:21": { |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
cat -n "src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json"Repository: ssdeanx/AgentStack
Length of output: 1797
🏁 Script executed:
# Check if this path/directory is in .gitignore
if [ -f .gitignore ]; then
echo "=== Checking .gitignore for SDK_CRAWLER_STATISTICS_0.json or related paths ==="
grep -E "(SDK_CRAWLER_STATISTICS|key_value_stores|\.mastra)" .gitignore || echo "No matches found"
else
echo ".gitignore not found"
fiRepository: ssdeanx/AgentStack
Length of output: 445
🏁 Script executed:
# Search for similar local file paths in committed JSON files as suggested in the review
rg -n 'file:///[A-Za-z]:/Users/' --type json
rg -n 'file:///home/' --type jsonRepository: ssdeanx/AgentStack
Length of output: 44
Local filesystem path exposes developer username — remove from git history.
The error keys on lines 30 and 41 contain a full Windows file URI embedding a local OS username:
file:///C:/Users/ssdsk/AgentStack/.mastra/output/tools/63e94f59-8415-41f8-b936-2b2d519f5000.mjs:602:21
The path segment C:/Users/ssdsk/ exposes PII, violating the coding guideline that JSON files must never include personally identifiable information.
Although this file is already listed in .gitignore, it is present in the git history. Remove it using git rm --cached src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json, then commit the removal to purge it from the repository.
Also applies to: 41-41
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In
`@src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json`
at line 30, The JSON file
src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json
contains a Windows file URI exposing a local username
("file:///C:/Users/ssdsk/...") in the object keys; remove the sensitive file and
scrub history by first deleting it from the index (git rm --cached
src/mastra/public/storage/key_value_stores/default/SDK_CRAWLER_STATISTICS_0.json),
update the working copy to replace the offending keys (remove or sanitize the
"file:///C:/Users/ssdsk/..." entries) and commit that change, then purge the
file and any matching path strings from repository history using a
history-rewrite tool (git filter-repo or BFG) to remove all occurrences of that
file/PII, and ensure the file remains listed in .gitignore afterwards.
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request introduces extensive documentation for the AI SDK UI and Mastra backend, which is a great addition for developers. It also adds a new main sidebar component to improve navigation and user experience. My review focuses on several areas for improvement. I've identified some issues in the new main-sidebar.tsx component, including duplicated UI elements and missing data fetching for networks, which should be addressed. There's also some code duplication in the API routes (chat-extra and completion) that could be refactored for better maintainability. Additionally, I've found some minor issues in the new documentation files and action handlers, such as developer comments, magic numbers, and unused parameters. Addressing these points will enhance the quality and robustness of the new features.
Comment on lines
+90
to
+124
| const [networks, setNetworks] = useState<any[]>([]) | ||
| const [traces, setTraces] = useState<SidebarTrace[]>([]) | ||
| const [threads, setThreads] = useState<SidebarThread[]>([]) | ||
| const [vectors, setVectors] = useState<SidebarVectorIndex[]>([]) | ||
|
|
||
| const [loading, setLoading] = useState(false) | ||
| const [error, setError] = useState<string | null>(null) | ||
|
|
||
| // Load sidebar data on mount | ||
| useEffect(() => { | ||
| const loadData = async () => { | ||
| setLoading(true) | ||
| setError(null) | ||
| try { | ||
| const [ | ||
| agentsData, | ||
| toolsData, | ||
| workflowsData, | ||
| tracesData, | ||
| threadsData, | ||
| vectorsData, | ||
| ] = await Promise.all([ | ||
| fetchAgents(), | ||
| fetchTools(), | ||
| fetchWorkflows(), | ||
| fetchTraces(), | ||
| fetchThreads(), | ||
| fetchVectors(), | ||
| ]) | ||
| setAgents(agentsData) | ||
| setTools(toolsData) | ||
| setWorkflows(workflowsData) | ||
| setTraces(tracesData) | ||
| setThreads(threadsData) | ||
| setVectors(vectorsData) |
Contributor
There was a problem hiding this comment.
The networks state is initialized but is never populated with data, as there is no corresponding fetch call in the loadData function. This causes the 'Networks' section in the sidebar to always be empty. The data fetching for networks should be implemented.
Comment on lines
+354
to
+472
| <SidebarGroup> | ||
| <SidebarGroupLabel>Resources</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <WorkflowIcon className="mr-2 size-4" /> | ||
| <span>Workflows</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {workflows.map((wf) => ( | ||
| <SidebarMenuSubItem key={wf.id}> | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {wf.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <NetworkIcon className="mr-2 size-4" /> | ||
| <span>Networks</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {networks.map((net) => ( | ||
| <SidebarMenuSubItem | ||
| key={net.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {net.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
|
|
||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <CpuIcon className="mr-2 size-4" /> | ||
| <span>Tools</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {tools.map((tool) => ( | ||
| <SidebarMenuSubItem | ||
| key={tool.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {tool.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> | ||
|
|
||
| <SidebarGroup> | ||
| <SidebarGroupLabel>Observability</SidebarGroupLabel> | ||
| <SidebarGroupContent> | ||
| <SidebarMenu> | ||
| <Collapsible className="group/collapsible"> | ||
| <SidebarMenuItem> | ||
| <CollapsibleTrigger asChild> | ||
| <SidebarMenuButton> | ||
| <ActivityIcon className="mr-2 size-4" /> | ||
| <span>Recent Traces</span> | ||
| <ChevronRightIcon className="ml-auto size-4 transition-transform group-data-[state=open]/collapsible:rotate-90" /> | ||
| </SidebarMenuButton> | ||
| </CollapsibleTrigger> | ||
| <CollapsibleContent> | ||
| <SidebarMenuSub> | ||
| {traces.map((trace) => ( | ||
| <SidebarMenuSubItem | ||
| key={trace.id} | ||
| > | ||
| <SidebarMenuSubButton className="w-full text-xs cursor-default"> | ||
| <span className="truncate"> | ||
| {trace.name} | ||
| </span> | ||
| </SidebarMenuSubButton> | ||
| </SidebarMenuSubItem> | ||
| ))} | ||
| </SidebarMenuSub> | ||
| </CollapsibleContent> | ||
| </SidebarMenuItem> | ||
| </Collapsible> | ||
| </SidebarMenu> | ||
| </SidebarGroupContent> | ||
| </SidebarGroup> |
Contributor
There was a problem hiding this comment.
There is a duplicated block of UI code. The SidebarGroup for 'Resources' and 'Observability' is rendered twice. This appears to be a copy-paste error and the second block should be removed to fix the UI.
Comment on lines
+209
to
+211
| | File | Exports | Purpose | Section | | ||
|
|
||
| |------|---------|---------| |
Contributor
There was a problem hiding this comment.
The markdown table header is broken here, with the header row split across two lines. Additionally, the last column contains informal notes like **This is for Web** and **This is for Coding** which should be integrated into the 'Purpose' column for better clarity and a more consistent documentation style.
| export async function POST(req: Request) { | ||
| const body = await req.json() | ||
| const { messages, data, id } = body | ||
| const agentId = data?.agentId ?? id ?? 'weatherAgent' // This will only display weatherAgent so this is kinda useless since my mastra/index uses :agentid which makes the route dynamic and since its dynamic u can change agent by putting my other agents.... This would be a static port it must have a string for agent it wont just pick it up. |
Contributor
There was a problem hiding this comment.
This line contains a developer comment that appears to have been left in the documentation unintentionally. It should be removed to maintain the clarity and professionalism of the reference material.
Suggested change
| const agentId = data?.agentId ?? id ?? 'weatherAgent' // This will only display weatherAgent so this is kinda useless since my mastra/index uses :agentid which makes the route dynamic and since its dynamic u can change agent by putting my other agents.... This would be a static port it must have a string for agent it wont just pick it up. | |
| const agentId = data?.agentId ?? id ?? 'weatherAgent' | |
Comment on lines
11
to
93
| export async function POST(req: Request) { | ||
| const { messages, data } = await req.json() | ||
| const myAgent = mastra.getAgent('weatherAgent') | ||
| const body = await req.json() | ||
| const { messages, data, id } = body | ||
|
|
||
| // Dynamic agent selection - prioritize data.agentId, fallback to id, then weatherAgent | ||
| const agentId = | ||
| (typeof data?.agentId === 'string' && data.agentId.length > 0 | ||
| ? data.agentId | ||
| : undefined) ?? | ||
| (typeof id === 'string' && id.length > 0 ? id : undefined) ?? | ||
| 'weatherAgent' | ||
|
|
||
| const agent = mastra.getAgent(agentId) | ||
|
|
||
| if (!agent) { | ||
| return Response.json( | ||
| { error: `Agent "${agentId}" not found` }, | ||
| { status: 404 } | ||
| ) | ||
| } | ||
|
|
||
| // Extract multi-tenancy IDs | ||
| const threadId = body.threadId ?? data?.threadId | ||
| const resourceId = body.resourceId ?? data?.resourceId | ||
|
|
||
| // Create RequestContext for multi-tenancy isolation | ||
| const requestContext = new RequestContext() | ||
| if (resourceId) requestContext.set(MASTRA_RESOURCE_ID_KEY, resourceId) | ||
| if (threadId) requestContext.set(MASTRA_THREAD_ID_KEY, threadId) | ||
|
|
||
| if (data) { | ||
| // Merge other data into context if present | ||
| if (data && typeof data === 'object') { | ||
| for (const [key, value] of Object.entries(data)) { | ||
| requestContext.set(key, value) | ||
| if ( | ||
| key !== 'agentId' && | ||
| key !== 'threadId' && | ||
| key !== 'resourceId' | ||
| ) { | ||
| requestContext.set(key, value) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| const stream = await myAgent.stream(messages, { | ||
| // Prepare stream options | ||
| const streamOptions = { | ||
| threadId, | ||
| resourceId, | ||
| memory: body.memory ?? data?.memory, | ||
| requestContext, | ||
| format: 'aisdk', | ||
| } | ||
|
|
||
| const stream = await agent.stream(messages, streamOptions) | ||
|
|
||
| const uiStream = createUIMessageStream({ | ||
| originalMessages: messages as UIMessage[], | ||
| execute: async ({ writer }) => { | ||
| const aiStream = toAISdkStream(stream, { | ||
| from: 'agent', | ||
| sendReasoning: true, | ||
| sendSources: true, | ||
| }) | ||
|
|
||
| if (aiStream && typeof (aiStream as any).getReader === 'function') { | ||
| const reader = (aiStream as ReadableStream<any>).getReader() | ||
| try { | ||
| while (true) { | ||
| const { value, done } = await reader.read() | ||
| if (done) break | ||
| await writer.write(value) | ||
| } | ||
| } finally { | ||
| reader.releaseLock?.() | ||
| } | ||
| } else if (aiStream && Symbol.asyncIterator in aiStream) { | ||
| for await (const part of aiStream as AsyncIterable<any>) { | ||
| await writer.write(part) | ||
| } | ||
| } | ||
| }, | ||
| }) | ||
| return stream.toUIMessageStreamResponse() | ||
|
|
||
| return createUIMessageStreamResponse({ stream: uiStream }) | ||
| } |
Contributor
There was a problem hiding this comment.
This API route handler duplicates a significant amount of logic for creating a streaming response, which is also present in other routes like app/api/chat/route.ts. To improve maintainability and reduce redundancy, this logic should be extracted into a shared helper function. The project's documentation already suggests using a helper like createAgentStreamResponse for this purpose.
Comment on lines
11
to
106
| export async function POST(req: Request) { | ||
| const { prompt } = await req.json() | ||
| const myAgent = mastra.getAgent('weatherAgent') | ||
| const stream = await myAgent.stream([{ role: 'user', content: prompt }], { | ||
| format: 'aisdk', | ||
| }) | ||
| const body = await req.json() | ||
| const { prompt, agentId, data } = body | ||
|
|
||
| // Dynamic agent selection - prioritize agentId param, fallback to weatherAgent | ||
| const selectedAgentId = | ||
| (typeof agentId === 'string' && agentId.length > 0 | ||
| ? agentId | ||
| : undefined) ?? | ||
| (typeof data?.agentId === 'string' && data.agentId.length > 0 | ||
| ? data.agentId | ||
| : undefined) ?? | ||
| 'weatherAgent' | ||
|
|
||
| const agent = mastra.getAgent(selectedAgentId) | ||
|
|
||
| if (!agent) { | ||
| return Response.json( | ||
| { error: `Agent "${selectedAgentId}" not found` }, | ||
| { status: 404 } | ||
| ) | ||
| } | ||
|
|
||
| // Extract multi-tenancy IDs | ||
| const threadId = body.threadId ?? data?.threadId | ||
| const resourceId = body.resourceId ?? data?.resourceId | ||
|
|
||
| // Create RequestContext for multi-tenancy isolation | ||
| const requestContext = new RequestContext() | ||
| if (resourceId) requestContext.set(MASTRA_RESOURCE_ID_KEY, resourceId) | ||
| if (threadId) requestContext.set(MASTRA_THREAD_ID_KEY, threadId) | ||
|
|
||
| // Merge other data into context if present | ||
| if (data && typeof data === 'object') { | ||
| for (const [key, value] of Object.entries(data)) { | ||
| if ( | ||
| key !== 'agentId' && | ||
| key !== 'threadId' && | ||
| key !== 'resourceId' | ||
| ) { | ||
| requestContext.set(key, value) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| // Convert single prompt to messages array with proper UIMessage shape | ||
| const messageId = `msg-${Date.now()}` | ||
| const messages: UIMessage[] = [ | ||
| { | ||
| id: messageId, | ||
| role: 'user', | ||
| parts: [{ type: 'text', text: prompt }], | ||
| }, | ||
| ] | ||
|
|
||
| // Prepare stream options | ||
| const streamOptions = { | ||
| threadId, | ||
| resourceId, | ||
| memory: body.memory ?? data?.memory, | ||
| requestContext, | ||
| } | ||
|
|
||
| const stream = await agent.stream(messages, streamOptions) | ||
|
|
||
| const uiStream = createUIMessageStream({ | ||
| originalMessages: messages, | ||
| execute: async ({ writer }) => { | ||
| const aiStream = toAISdkStream(stream, { | ||
| from: 'agent', | ||
| sendReasoning: true, | ||
| sendSources: true, | ||
| }) | ||
|
|
||
| // Support both ReadableStream and AsyncIterable (robust bridge) | ||
| if (aiStream && typeof (aiStream as any).getReader === 'function') { | ||
| const reader = (aiStream as ReadableStream<any>).getReader() | ||
| try { | ||
| while (true) { | ||
| const { value, done } = await reader.read() | ||
| if (done) break | ||
| await writer.write(value) | ||
| } | ||
| } finally { | ||
| reader.releaseLock?.() | ||
| } | ||
| } else if (aiStream && Symbol.asyncIterator in aiStream) { | ||
| for await (const part of aiStream as AsyncIterable<any>) { | ||
| await writer.write(part) | ||
| } | ||
| } | ||
| }, | ||
| }) | ||
|
|
||
| return stream.toUIMessageStreamResponse() | ||
| return createUIMessageStreamResponse({ stream: uiStream }) | ||
| } |
Contributor
There was a problem hiding this comment.
This route handler reimplements complex streaming logic, which is also found in other API routes. This code duplication increases the maintenance burden. It is recommended to refactor this to use a centralized helper function for creating streaming responses, as suggested in the project's documentation, to ensure consistency and simplify future updates.
Comment on lines
147
to
150
| export async function fetchThreads( | ||
| resourceId?: string | ||
| resourceId?: string, | ||
| agentId?: string | ||
| ): Promise<SidebarThread[]> { |
Contributor
There was a problem hiding this comment.
The fetchThreads function accepts an agentId parameter that is not used within the function's body. This can be misleading for developers. If the parameter is intended for future use, a // TODO comment would be helpful. Otherwise, it should be removed to clean up the function signature.
export async function fetchThreads(
resourceId?: string
): Promise<SidebarThread[]> {| const response = await mastraClient.createMemoryThread({ | ||
| agentId, | ||
| title: title ?? 'New Thread', | ||
| resourceId: resourceId ?? 'default-resource', |
Contributor
There was a problem hiding this comment.
The string 'default-resource' is hardcoded as a fallback resourceId in createThread, updateWorkingMemory (line 330), and saveMessage (line 350). Using hardcoded magic strings makes the code harder to maintain and update. This value should be extracted into a named constant and reused across these functions.
| threadId, | ||
| resourceId: m.resourceId || 'default-resource', | ||
| createdAt: m.createdAt || new Date(), | ||
| format: m.format || 2, |
Contributor
There was a problem hiding this comment.
The format property of a message is assigned a fallback value of 2. This magic number lacks context and makes the code harder to understand. To improve readability and maintainability, it should be replaced with a named constant or an enum that describes its meaning (e.g., MESSAGE_FORMAT_V2).
Comment on lines
+61
to
+80
| const NAV_ITEMS: SidebarNavItem[] = [ | ||
| { | ||
| id: 'workflows', | ||
| label: 'Workflows', | ||
| icon: <WorkflowIcon className="size-4" />, | ||
| href: '/workflows', | ||
| }, | ||
| { | ||
| id: 'tools', | ||
| label: 'Tools', | ||
| icon: <CpuIcon className="size-4" />, | ||
| href: '/tools', | ||
| }, | ||
| { | ||
| id: 'observability', | ||
| label: 'Observability', | ||
| icon: <ActivityIcon className="size-4" />, | ||
| href: '/dashboard/observability', | ||
| }, | ||
| ] |
Contributor
There was a problem hiding this comment.
The NAV_ITEMS constant and the associated handleNavClick function (lines 155-159) are defined but are not used anywhere in the component. This unused code should be removed to improve code clarity and maintainability.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
mastra-reference.mdto document the Mastra backend structure, including agents, tools, networks, workflows, configuration, A2A coordination, MCP server, evaluations, services, data processing, and policy definitions.reference.mdto provide an extended reference for AI elements, agent configuration schema, and API routes for the dashboard, linking to the newly createdmastra-reference.md.MainSidebarcomponent inmain-sidebar.tsxto enhance the user interface with navigation for agents, tools, workflows, and observability features.Summary by Sourcery
Add a primary AI dashboard sidebar and extend API/chat integrations with Mastra-aware streaming and multi-tenant support while documenting AI elements and Mastra backend structure.
New Features:
Enhancements:
Documentation: