feat: Add deep research use case (Python) #482
Conversation
🦋 Changeset detectedLatest commit: c2136e5 The changes in this PR will be included in the next version bump. This PR includes changesets to release 1 package
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
|
Warning Rate limit exceeded@leehuwuj has exceeded the limit for the number of commits or files that can be reviewed per hour. Please wait 5 minutes and 22 seconds before requesting another review. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. 📒 Files selected for processing (1)
WalkthroughThis pull request introduces a comprehensive enhancement for deep research functionality across multiple components. The changes span various files and technologies, adding a new "deep_research" application type with support for multi-agent document analysis workflows. The implementation includes Python backend components for research planning, frontend React components for displaying research progress, and updated type definitions to support the new use case. Changes
Sequence DiagramsequenceDiagram
participant User
participant ChatUI
participant WorkflowEngine
participant ResearchAgents
participant DocumentIndex
User->>ChatUI: Initiate Deep Research
ChatUI->>WorkflowEngine: Start Workflow
WorkflowEngine->>DocumentIndex: Retrieve Relevant Documents
DocumentIndex-->>WorkflowEngine: Return Context Nodes
WorkflowEngine->>ResearchAgents: Plan Research
ResearchAgents-->>WorkflowEngine: Generate Research Questions
WorkflowEngine->>ResearchAgents: Execute Research
ResearchAgents->>DocumentIndex: Answer Questions
ResearchAgents-->>WorkflowEngine: Compile Answers
WorkflowEngine->>ChatUI: Stream Research Progress
WorkflowEngine->>ChatUI: Generate Final Report
Possibly related PRs
Suggested reviewers
Poem
Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media? 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
Documentation and Community
|
templates/components/agents/python/blog/app/workflows/writer.py
Outdated
Show resolved
Hide resolved
templates/components/agents/python/blog/app/workflows/models.py
Outdated
Show resolved
Hide resolved
templates/components/agents/python/blog/app/workflows/models.py
Outdated
Show resolved
Hide resolved
templates/components/agents/python/blog/app/workflows/models.py
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/chat-message-content.tsx
Outdated
Show resolved
Hide resolved
templates/components/agents/python/blog/app/workflows/agents.py
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
leehuwuj
force-pushed
the
lee/deep-research
branch
from
a912773 to
acfc4ec
Compare
leehuwuj
force-pushed
the
lee/deep-research
branch
from
e288b08 to
76c20f3
Compare
leehuwuj
force-pushed
the
lee/deep-research
branch
from
81cb6c3 to
a3a09d7
Compare
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 6
🧹 Nitpick comments (8)
templates/components/agents/python/blog/app/workflows/agents.py (1)
34-57: Adjust Indentation in Multi-Line Prompt StringThe prompt string in
analyze_prompthas inconsistent indentation, which may lead to formatting issues in the generated prompt. Consider adjusting the indentation to ensure the prompt is correctly formatted.Apply this diff to fix the indentation:
analyze_prompt = PromptTemplate( - """ - You are a professor who is guiding a researcher to research a specific request/problem. - Your task is to decide on a research plan for the researcher. - The possible actions are: - + Provide a list of questions for the researcher to investigate, with the purpose of clarifying the request. - + Write a report if the researcher has already gathered enough research on the topic and can resolve the initial request. - + Cancel the research if most of the answers from researchers indicate there is insufficient information to research the request. Do not attempt more than 3 research iterations or too many questions. - The workflow should be: - + Always begin by providing some initial questions for the researcher to investigate. - + Analyze the provided answers against the initial topic/request. If the answers are insufficient to resolve the initial request, provide additional questions for the researcher to investigate. - + If the answers are sufficient to resolve the initial request, instruct the researcher to write a report. - <User request> - {user_request} - </User request> - - <Collected information> - {context_str} - </Collected information> - - <Conversation context> - {conversation_context} - </Conversation context> - """ + """ +You are a professor who is guiding a researcher to research a specific request/problem. +Your task is to decide on a research plan for the researcher. +The possible actions are: + - Provide a list of questions for the researcher to investigate, with the purpose of clarifying the request. + - Write a report if the researcher has already gathered enough research on the topic and can resolve the initial request. + - Cancel the research if most of the answers from researchers indicate there is insufficient information to research the request. Do not attempt more than 3 research iterations or too many questions. +The workflow should be: + - Always begin by providing some initial questions for the researcher to investigate. + - Analyze the provided answers against the initial topic/request. If the answers are insufficient to resolve the initial request, provide additional questions for the researcher to investigate. + - If the answers are sufficient to resolve the initial request, instruct the researcher to write a report. + +<User request> +{user_request} +</User request> + +<Collected information> +{context_str} +</Collected information> + +<Conversation context> +{conversation_context} +</Conversation context> +""" )templates/components/agents/python/blog/app/workflows/writer.py (2)
114-149: Consider Streaming the Report Generation for Better User ExperienceIn the
write_reportfunction, you have an option to stream the response based on thestreamflag. Streaming the report generation can enhance the user experience by allowing the user to receive the content incrementally.
214-249: Limit the Number of Concurrent Workers inanswerStepIn the
answerstep, you have setnum_workers=2. Ensure that this limit is appropriate for your application's performance and resource constraints. Increasing the number may improve throughput but could also lead to resource exhaustion.helpers/datasources.ts (1)
45-53: Consider making the AI reports data source more maintainable.The hardcoded URL and verbose filename could be improved:
- Consider moving the URL to a configuration file for easier updates.
- Simplify the filename to something more readable.
export const AI_REPORTS: TemplateDataSource = { type: "file", config: { url: new URL( "https://www.europarl.europa.eu/RegData/etudes/ATAG/2024/760392/EPRS_ATA(2024)760392_EN.pdf", ), - filename: "EPRS_ATA_2024_760392_EN.pdf", + filename: "ai_report_2024.pdf", }, };templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (1)
59-112: Enhance error handling in state transformation.The state transformation function could benefit from better error handling:
- Add validation for required fields.
- Handle potential edge cases.
const transformState = ( state: WriterState, event: WriterEvent, ): WriterState => { + if (!event?.data) { + return state; + } + switch (event.data.event) { case "answer": { const { id, question, answer } = event.data; if (!id || !question) return state;questions/simple.ts (2)
230-236: Consider adding document generation tools for blog use case.The blog use case might benefit from document generation tools for creating well-formatted blog posts.
blog: { template: "multiagent", useCase: "blog", - tools: [], + tools: getTools(["document_generator"]), frontend: true, dataSources: [AI_REPORTS], },
98-102: Simplify language selection logic.Consider using an array or Set for excluded app types to make the code more maintainable.
+const FASTAPI_ONLY_APPS = new Set(["extractor", "contract_review", "blog"]); + - if ( - appType !== "extractor" && - appType !== "contract_review" && - appType !== "blog" - ) { + if (!FASTAPI_ONLY_APPS.has(appType)) {templates/components/agents/python/blog/README-template.md (1)
28-28: Add missing article for better readability.Add "the" before "generate script" to improve grammar.
-or add your own documents and run generate script again. +or add your own documents and run the generate script again.🧰 Tools
🪛 LanguageTool
[uncategorized] ~28-~28: You might be missing the article “the” here.
Context: ...2024, or add your own documents and run generate script again. After starting the serve...(AI_EN_LECTOR_MISSING_DETERMINER_THE)
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
⛔ Files ignored due to path filters (1)
pnpm-lock.yamlis excluded by!**/pnpm-lock.yaml
📒 Files selected for processing (23)
helpers/datasources.ts(1 hunks)helpers/types.ts(1 hunks)package.json(1 hunks)questions/simple.ts(5 hunks)templates/components/agents/python/blog/README-template.md(2 hunks)templates/components/agents/python/blog/app/agents/choreography.py(0 hunks)templates/components/agents/python/blog/app/agents/orchestrator.py(0 hunks)templates/components/agents/python/blog/app/agents/publisher.py(0 hunks)templates/components/agents/python/blog/app/agents/researcher.py(0 hunks)templates/components/agents/python/blog/app/agents/workflow.py(0 hunks)templates/components/agents/python/blog/app/workflows/__init__.py(1 hunks)templates/components/agents/python/blog/app/workflows/agents.py(1 hunks)templates/components/agents/python/blog/app/workflows/blog.py(0 hunks)templates/components/agents/python/blog/app/workflows/models.py(1 hunks)templates/components/agents/python/blog/app/workflows/multi.py(0 hunks)templates/components/agents/python/blog/app/workflows/planner.py(0 hunks)templates/components/agents/python/blog/app/workflows/single.py(0 hunks)templates/components/agents/python/blog/app/workflows/writer.py(1 hunks)templates/components/multiagent/python/app/api/routers/chat.py(1 hunks)templates/components/multiagent/python/app/api/routers/vercel_response.py(4 hunks)templates/components/multiagent/python/app/workflows/events.py(2 hunks)templates/types/streaming/nextjs/app/components/ui/chat/chat-message-content.tsx(2 hunks)templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx(1 hunks)
💤 Files with no reviewable changes (9)
- templates/components/agents/python/blog/app/workflows/blog.py
- templates/components/agents/python/blog/app/agents/orchestrator.py
- templates/components/agents/python/blog/app/agents/choreography.py
- templates/components/agents/python/blog/app/agents/researcher.py
- templates/components/agents/python/blog/app/workflows/multi.py
- templates/components/agents/python/blog/app/agents/publisher.py
- templates/components/agents/python/blog/app/workflows/planner.py
- templates/components/agents/python/blog/app/workflows/single.py
- templates/components/agents/python/blog/app/agents/workflow.py
✅ Files skipped from review due to trivial changes (1)
- templates/components/agents/python/blog/app/workflows/init.py
🧰 Additional context used
📓 Path-based instructions (9)
templates/types/streaming/nextjs/app/components/ui/chat/chat-message-content.tsx (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/multiagent/python/app/api/routers/chat.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/multiagent/python/app/workflows/events.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/blog/README-template.md (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/blog/app/workflows/models.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/multiagent/python/app/api/routers/vercel_response.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/blog/app/workflows/agents.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/blog/app/workflows/writer.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
🪛 LanguageTool
templates/components/agents/python/blog/README-template.md
[uncategorized] ~28-~28: You might be missing the article “the” here.
Context: ...2024, or add your own documents and run generate script again. After starting the serve...
(AI_EN_LECTOR_MISSING_DETERMINER_THE)
⏰ Context from checks skipped due to timeout of 90000ms (41)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --example-file)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --no-files)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --no-files)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --example-file)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --no-files)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --no-files)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --example-file)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --no-files)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --no-files)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --llamacloud)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --example-file)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --no-files)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --llamacloud)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --example-file)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --no-files)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --example-file)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --no-files)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --example-file)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --no-files)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --example-file)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --no-files)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --llamacloud)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --no-files)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --no-files)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --example-file)
🔇 Additional comments (17)
templates/components/multiagent/python/app/api/routers/vercel_response.py (1)
91-93: Handle Unexpected Result Types in_chat_response_generatorMake sure that the variable
resultis always converted to a string before concatenation tofinal_responseand yielding. Verify thatresultis of an expected type to prevent potential runtime errors.templates/components/agents/python/blog/app/workflows/agents.py (1)
156-158: Sanitize Node IDs to Prevent Injection AttacksWhen constructing content with node IDs in
_get_text_node_content_for_citation, ensure node IDs are properly escaped or sanitized to prevent potential injection attacks if node IDs contain special characters.[security_issue]
Consider applying input sanitization:
import html node_id = html.escape(node.node.node_id)templates/components/agents/python/blog/app/workflows/writer.py (2)
264-270: Verify Correctness of Event Collection incollect_answersIn the
collect_answersmethod, you usectx.collect_eventswith an expected list of events based onnum_questions. Ensure thatnum_questionscorrectly reflects the number ofCollectAnswersEventinstances to be collected.
73-90: Initialize Memory with Correct Chat HistoryIn the
__init__method ofWriterWorkflow, ensure that the chat history is correctly initialized in theChatMemoryBuffer. This will help maintain conversation context throughout the workflow.templates/components/agents/python/blog/app/workflows/models.py (6)
9-10: LGTM! Simple marker event for planning research phase.The empty event class is appropriate for signaling the start of research planning.
13-16: LGTM! Well-structured research event with proper metadata.The event captures all necessary data for research tracking:
- Question ID for referencing
- Actual question text for readability
- Context nodes with relevance scores
19-22: LGTM! Consistent answer collection event.The event maintains symmetry with ResearchEvent for question tracking while adding the answer field.
25-26: LGTM! Simple marker event for report writing phase.The empty event class is appropriate for signaling the start of report writing.
30-35: LGTM! Well-defined event data model with type safety.Good use of:
- Literal types for event and state enums
- Optional fields for flexible data structure
- Clear state progression (pending → inprogress → done/error)
38-43: LGTM! Clean event class with proper serialization.The event properly encapsulates writer card data with a clean serialization method.
templates/components/multiagent/python/app/workflows/events.py (1)
33-45: LGTM! Well-structured source nodes event with clean serialization.The event properly:
- Handles source nodes with relevance scores
- Transforms nodes to a consistent format
- Provides a clean response structure
templates/types/streaming/nextjs/app/components/ui/chat/chat-message-content.tsx (1)
26-30: LGTM! Clean integration of WriterCard component.Based on the previous discussion in PR comments, this is a reasonable approach to integrate the writer-specific component while maintaining extensibility for future custom components.
templates/components/multiagent/python/app/api/routers/chat.py (1)
41-41: LGTM! Clean addition of background tasks support.The background tasks parameter is properly integrated into the endpoint and passed to the response handler.
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (1)
1-1: Add 'use client' directive.The component needs the 'use client' directive at the top of the file for Next.js client-side rendering.
+// @ts-expect-error - 'use client' directive +'use client'; import { Message } from "@llamaindex/chat-ui";templates/components/agents/python/blog/README-template.md (2)
1-1: LGTM! Clear and concise introduction.The introduction effectively communicates that this is a LlamaIndex multi-agents project using Workflows.
26-31: LGTM! Clear usage instructions and example.The section effectively explains the blog writer use case with a practical example.
🧰 Tools
🪛 LanguageTool
[uncategorized] ~28-~28: You might be missing the article “the” here.
Context: ...2024, or add your own documents and run generate script again. After starting the serve...(AI_EN_LECTOR_MISSING_DETERMINER_THE)
package.json (1)
46-46: LGTM! Type definitions version update.The update to @types/prompts version 2.4.2 aligns with the new blog-related type definitions.
templates/components/multiagent/python/app/api/routers/vercel_response.py
Show resolved
Hide resolved
templates/components/multiagent/python/app/api/routers/vercel_response.py
Show resolved
Hide resolved
templates/components/agents/python/blog/app/workflows/agents.py
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (3)
templates/components/agents/python/blog/README-template.md (3)
Line range hint
3-24: Add more details about required environment variables.While the setup instructions are clear, consider adding a list of required environment variables and their descriptions to help users configure their
.envfile correctly.Add a section like this after the dev-container note:
Required environment variables in `.env`: - `OPENAI_API_KEY`: Your OpenAI API key (required if using OpenAI models) - Add other required variables here...
26-31: Consider adding more diverse examples.While the AI investment example is good, providing additional example prompts would help users better understand the range of blog topics they can generate.
Add more examples like:
Example prompts: - "Write a post about AI investment in 2024" - "Create a blog post summarizing the key trends in [topic from your documents]" - "Generate an article analyzing [specific aspect] from the provided documents"
33-33: Add more details about the workflow structure.Consider expanding this section to help users understand how to customize the workflow. Include information about the
WriterWorkflowclass and its components.Add details like:
The `writer.py` file contains the `WriterWorkflow` class that orchestrates: - Information retrieval from documents - Content analysis - Blog post generation To customize the workflow, you can modify: - Research parameters - Analysis criteria - Output format
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
templates/components/agents/python/blog/README-template.md(2 hunks)
🧰 Additional context used
📓 Path-based instructions (1)
templates/components/agents/python/blog/README-template.md (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
⏰ Context from checks skipped due to timeout of 90000ms (45)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --example-file)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --no-files)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --no-files)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --example-file)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --no-files)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --no-files)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --example-file)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --no-files)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --no-files)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --llamacloud)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --example-file)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --no-files)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --llamacloud)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --example-file)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --no-files)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --example-file)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --no-files)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --llamacloud)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --example-file)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --no-files)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --no-files)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --llamacloud)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --example-file)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --no-files)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --no-files)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --llamacloud)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --example-file)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --no-files)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --no-files)
🔇 Additional comments (3)
templates/components/agents/python/blog/README-template.md (3)
1-1: LGTM! Clear and concise introduction.The introduction effectively communicates the project's purpose and core technologies with proper documentation links.
33-33: File path reference is now correct.The text and link now consistently refer to
writer.py.
Line range hint
35-47: LGTM! Comprehensive resource links.The deployment reference and learning resources are well-organized and provide valuable links to documentation.
templates/components/agents/python/blog/app/workflows/models.py
Outdated
Show resolved
Hide resolved
templates/components/agents/python/blog/app/workflows/models.py
Outdated
Show resolved
Hide resolved
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (1)
142-152:⚠️ Potential issueAdd type safety and error handling for annotations.
The type casting of annotations could lead to runtime errors.
Add error boundary and type guards:
export function WriterCard({ message, className }: WriterCardProps) { + if (!message?.annotations) { + return null; + } + const writerEvents = message.annotations as WriterEvent[] | undefined; + + // Add type guard + const isWriterEvent = (event: any): event is WriterEvent => { + return event?.type === "writer_card" && event?.data?.event; + }; + const hasWriterEvents = writerEvents?.some( - (event) => event.type === "writer_card", + isWriterEvent );
🧹 Nitpick comments (2)
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (2)
22-58: Consider using discriminated union types for better type safety.The event types could be more specific using discriminated unions. This would provide better type safety and autocompletion.
Consider this refactor:
-type WriterEvent = { +type BaseWriterEvent = { type: "writer_card"; +}; + +type RetrieveEvent = BaseWriterEvent & { + data: { + event: "retrieve"; + state: EventState; + }; +}; + +type AnalyzeEvent = BaseWriterEvent & { + data: { + event: "analyze"; + state: EventState; + }; +}; + +type AnswerEvent = BaseWriterEvent & { data: { - event: "retrieve" | "analyze" | "answer"; + event: "answer"; state: EventState; id?: string; question?: string; answer?: string | null; }; -}; +}; + +type WriterEvent = RetrieveEvent | AnalyzeEvent | AnswerEvent;
154-213: Add loading states and accessibility improvements.The UI could benefit from better loading indicators and accessibility.
Consider these improvements:
return ( <div className={cn( "rounded-lg border bg-card text-card-foreground shadow-sm p-5 space-y-6 w-full", className, )} + role="region" + aria-label="Writer Analysis" > {state.retrieve.state !== null && ( <div className="border-t pt-4"> - <h3 className="text-lg font-semibold flex items-center gap-2"> + <h3 className="text-lg font-semibold flex items-center gap-2" aria-live="polite"> <Search className="h-5 w-5" /> <span> {state.retrieve.state === "inprogress" ? "Searching..." : "Search completed"} </span> + {state.retrieve.state === "inprogress" && ( + <span className="sr-only">Loading...</span> + )} </h3> </div> )}
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
templates/components/agents/python/blog/app/workflows/models.py(1 hunks)templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (1)
- templates/components/agents/python/blog/app/workflows/models.py
🧰 Additional context used
📓 Path-based instructions (1)
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
⏰ Context from checks skipped due to timeout of 90000ms (44)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --example-file)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, express, --no-files)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, ubuntu-22.04, nextjs, --no-files)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --example-file)
- GitHub Check: typescript (20, 3.11, windows-latest, express, --no-files)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, windows-latest, nextjs, --no-files)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --llamacloud)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --example-file)
- GitHub Check: typescript (20, 3.11, macos-latest, express, --no-files)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --llamacloud)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --example-file)
- GitHub Check: typescript (20, 3.11, macos-latest, nextjs, --no-files)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --llamacloud)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --example-file)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, express, --no-files)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --llamacloud)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --example-file)
- GitHub Check: typescript (18, 3.11, ubuntu-22.04, nextjs, --no-files)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --example-file)
- GitHub Check: typescript (18, 3.11, windows-latest, express, --no-files)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --example-file)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, windows-latest, nextjs, --no-files)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --llamacloud)
- GitHub Check: python (20, 3.11, ubuntu-22.04, fastapi, --no-files)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --example-file)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, macos-latest, express, --no-files)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --example-file)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --llamacloud)
- GitHub Check: python (20, 3.11, windows-latest, fastapi, --no-files)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --example-file)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --llamacloud)
- GitHub Check: typescript (18, 3.11, macos-latest, nextjs, --no-files)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --example-file)
- GitHub Check: python (20, 3.11, macos-latest, fastapi, --no-files)
🔇 Additional comments (1)
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx (1)
1-21: LGTM! Well-structured imports and client directive.The file structure follows Next.js best practices with the "use client" directive correctly placed at the top, and imports are well-organized.
templates/types/streaming/nextjs/app/components/ui/chat/custom/writer-card.tsx
Outdated
Show resolved
Hide resolved
leehuwuj
force-pushed
the
lee/deep-research
branch
from
1162efd to
c4020cb
Compare
leehuwuj
changed the title
There was a problem hiding this comment.
Actionable comments posted: 5
🧹 Nitpick comments (6)
templates/components/agents/python/deep_research/app/workflows/models.py (2)
13-16: Consider adding input validation for question attributes.While the typing is good, consider adding validation to ensure
question_idandquestionare non-empty strings. You can use Pydantic for this:-class ResearchEvent(Event): +from pydantic import Field + +class ResearchEvent(Event): question_id: str = Field(..., min_length=1) question: str = Field(..., min_length=1) context_nodes: List[NodeWithScore]
19-22: Consider inheritance for common attributes.
ResearchEventandCollectAnswersEventshare common attributes. Consider creating a base class:+class BaseQuestionEvent(Event): + question_id: str = Field(..., min_length=1) + question: str = Field(..., min_length=1) + -class ResearchEvent(Event): +class ResearchEvent(BaseQuestionEvent): - question_id: str - question: str context_nodes: List[NodeWithScore] -class CollectAnswersEvent(Event): +class CollectAnswersEvent(BaseQuestionEvent): - question_id: str - question: str answer: strtemplates/types/streaming/nextjs/app/components/ui/chat/custom/deep-research-card.tsx (1)
145-225: Consider improving accessibility.While the implementation is solid, consider these accessibility improvements:
- <CollapsibleTrigger className="w-full"> + <CollapsibleTrigger className="w-full" aria-label={`Toggle answer for: ${question.question}`}>Also consider adding appropriate ARIA labels for the search and analysis sections.
.changeset/yellow-oranges-play.md (1)
1-5: Consider expanding the changeset description.While the version bump is appropriate, consider adding more details about:
- Key features of the deep research functionality
- Any breaking changes or dependencies
- Migration steps if applicable
🧰 Tools
🪛 LanguageTool
[style] ~5-~5: Consider an alternative adjective to strengthen your wording.
Context: --- "create-llama": patch --- Add deep research over own documents use case (P...(DEEP_PROFOUND)
templates/components/agents/python/deep_research/README-template.md (2)
5-11: Enhance setup instructions with environment requirements.The setup instructions should specify:
- Required Python version
- Link to dev container setup documentation
- Platform-specific requirements (if any)
26-35: Enhance technical documentation for usage section.Please add:
- List of supported file formats
- Default value of
TOP_K- Performance implications of different
TOP_Kvalues- Memory requirements for document processing
🧰 Tools
🪛 LanguageTool
[style] ~26-~26: Consider an alternative adjective to strengthen your wording.
Context: ...`shell poetry run dev ``` ## Use Case: Deep Research over own documents The workfl...(DEEP_PROFOUND)
[style] ~28-~28: Consider an alternative adjective to strengthen your wording.
Context: ...er own documents The workflow performs deep research by retrieving and analyzing do...(DEEP_PROFOUND)
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (11)
.changeset/yellow-oranges-play.md(1 hunks)helpers/types.ts(1 hunks)questions/simple.ts(5 hunks)templates/components/agents/python/deep_research/README-template.md(1 hunks)templates/components/agents/python/deep_research/app/workflows/__init__.py(1 hunks)templates/components/agents/python/deep_research/app/workflows/agents.py(1 hunks)templates/components/agents/python/deep_research/app/workflows/deep_research.py(1 hunks)templates/components/agents/python/deep_research/app/workflows/models.py(1 hunks)templates/types/streaming/nextjs/app/components/ui/chat/chat-message-content.tsx(2 hunks)templates/types/streaming/nextjs/app/components/ui/chat/custom/deep-research-card.tsx(1 hunks)templates/types/streaming/nextjs/package.json(1 hunks)
✅ Files skipped from review due to trivial changes (1)
- templates/types/streaming/nextjs/package.json
🚧 Files skipped from review as they are similar to previous changes (1)
- templates/types/streaming/nextjs/app/components/ui/chat/chat-message-content.tsx
🧰 Additional context used
📓 Path-based instructions (6)
templates/components/agents/python/deep_research/README-template.md (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/deep_research/app/workflows/__init__.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/deep_research/app/workflows/models.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/types/streaming/nextjs/app/components/ui/chat/custom/deep-research-card.tsx (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/deep_research/app/workflows/agents.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
templates/components/agents/python/deep_research/app/workflows/deep_research.py (1)
Pattern templates/**: For files under the templates folder, do not report 'Missing Dependencies Detected' errors.
🪛 LanguageTool
.changeset/yellow-oranges-play.md
[style] ~5-~5: Consider an alternative adjective to strengthen your wording.
Context: --- "create-llama": patch --- Add deep research over own documents use case (P...
(DEEP_PROFOUND)
templates/components/agents/python/deep_research/README-template.md
[style] ~26-~26: Consider an alternative adjective to strengthen your wording.
Context: ...`shell poetry run dev ``` ## Use Case: Deep Research over own documents The workfl...
(DEEP_PROFOUND)
[style] ~28-~28: Consider an alternative adjective to strengthen your wording.
Context: ...er own documents The workflow performs deep research by retrieving and analyzing do...
(DEEP_PROFOUND)
🔇 Additional comments (14)
templates/components/agents/python/deep_research/app/workflows/agents.py (1)
14-25:AnalysisDecisionclass implementation is correct and well-structured.The class correctly leverages Pydantic's
BaseModelto define the data model with appropriate fields, types, and default values. This facilitates structured data handling and validation.templates/components/agents/python/deep_research/app/workflows/__init__.py (1)
1-3: Public API exposure is correctly implemented.The
create_workflowfunction is properly imported and included in the__all__list, making it accessible as part of the module's public interface.templates/components/agents/python/deep_research/app/workflows/models.py (4)
9-10: LGTM! Simple marker event class.The empty class extending
Eventis appropriate for signaling the start of research planning.
25-26: LGTM! Consistent with PlanResearchEvent pattern.The empty class extending
Eventis appropriate for signaling the completion of research.
30-35: LGTM! Well-structured event data model.Good use of:
- Literal types for type safety
- Optional fields for flexibility
- Comprehensive state enumeration
38-43: LGTM! Clean event serialization.The implementation correctly handles event type safety and serialization.
helpers/types.ts (1)
55-55: LGTM! Valid addition to template use cases.The new "deep_research" use case is consistently added to the union type.
templates/types/streaming/nextjs/app/components/ui/chat/custom/deep-research-card.tsx (3)
23-58: LGTM! Well-structured type definitions.Clear separation of concerns between event types and UI state types with comprehensive type safety.
68-120: LGTM! Clean and immutable state transformation.The implementation:
- Maintains immutability
- Handles edge cases
- Clearly separates different event types
123-143: LGTM! Clean state initialization.Good use of functional programming patterns with proper null handling.
questions/simple.ts (4)
21-22: LGTM! Type definition changes are clean and well-integrated.The addition of the "deep_research" type maintains consistency with existing patterns.
39-39: Great UI/UX improvements with clear categorization!The emoji prefixes and detailed descriptions help users understand the differences between agents and workflows.
Also applies to: 42-87
98-102: Verify language selection exclusion for deep research.The exclusion of language selection for deep_research seems appropriate, but please confirm this is intentional and aligns with the implementation.
✅ Verification successful
Language selection exclusion for deep_research is correct
The exclusion is appropriate as deep_research is implemented exclusively as a Python-based multiagent template with no alternative language implementations.
🏁 Scripts executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Check if deep_research implementation has language-specific code rg -l "deep_research.*\.(py|ts|js)"Length of output: 105
Script:
#!/bin/bash # Search for deep_research related files echo "=== Deep Research Files ===" fd "deep_research" echo -e "\n=== Deep Research Implementation References ===" rg -i "deep_research" -g "!*.md" -A 3 echo -e "\n=== Language Configuration Patterns ===" rg "language.*selection.*deep_research|deep_research.*language" -g "!*.md"Length of output: 7002
230-236: Review tools configuration for deep research.Other multiagent templates (like financial_report_agent) include tools such as document_generator and interpreter. Consider if deep_research would benefit from similar tools.
templates/components/agents/python/deep_research/app/workflows/agents.py
Show resolved
Hide resolved
templates/components/agents/python/deep_research/app/workflows/deep_research.py
Outdated
Show resolved
Hide resolved
templates/components/agents/python/deep_research/app/workflows/deep_research.py
Outdated
Show resolved
Hide resolved
Summary by CodeRabbit
Release Notes
New Features
Improvements
Documentation