refactor(mistral): enhance example with dynamic(pattern1) multi-tool support #28

YueTwo · 2025-07-04T08:39:14Z

Key improvements:

Converted single tool example to multi-tool demonstration
- switched to an open model which is able to achieve expected output
- updated task description to showcase multiple tool capabilities
Modified the system prompt
- ensured a more stable performance

Note: Upgrading to paid premium models can deliver more stable performance, and the current free open model is sensitive to the prompt.

Comparison of runtime results:

Before improvement:

OR
After improvement:

...

Summary by CodeRabbit

New Features
- Enhanced initial instructions to include starring a specific GitHub repository after searching for information.
Improvements
- Clarified prompt instructions for function usage and tool list management.
- Updated chat completion model to a newer version for improved performance.

…support

coderabbitai · 2025-07-04T08:39:20Z

Walkthrough

The script for an agent using dynamic tool discovery was updated to clarify prompt instructions, switch the Mistral chat model from "large" to "medium", and add a new user task involving starring a GitHub repository after performing a web search.

Changes

File(s)	Change Summary
examples/mistral/agent_with_dynamic_tool_discovery_pattern_1.py	Updated prompt for clarity, changed model to "mistral-medium-latest", and added a new user instruction.

Possibly related PRs

Added mistral #16: Refactored the same script by introducing initial prompt and model selection logic, which is directly refined in this PR.

Suggested reviewers

alex-aipolabs

Poem

A bunny hops through code so neat,
Tweaks the prompt and swaps the seat,
From large to medium, Mistral flows,
Now starring repos as it goes!
Discovery tools, instructions clear—
The agent’s smarter, give a cheer!
🐇✨

📜 Recent review details

Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between fffa0e2 and 2fb4717.

📒 Files selected for processing (1)

examples/mistral/agent_with_dynamic_tool_discovery_pattern_1.py (2 hunks)

🔇 Additional comments (3)

examples/mistral/agent_with_dynamic_tool_discovery_pattern_1.py (3)

25-26: LGTM! Improved prompt clarity for better LLM understanding.

The clarifications enhance the prompt by:

Specifying that ACI_SEARCH_FUNCTIONS takes the complete user input as a parameter

Correcting "appended" to "refreshed" which better describes the tools list management behavior

These changes should improve the LLM's understanding of how to use the meta function effectively.

45-45: Model downgrade aligns with PR objectives.

The change from "mistral-large-latest" to "mistral-medium-latest" supports the PR's goal of using a free open model. The enhanced prompt clarity should help compensate for any potential performance differences.

Consider monitoring the model's performance with the new multi-tool tasks to ensure it meets the expected output quality mentioned in the PR objectives.

53-53: Excellent multi-tool demonstration enhancement.

The extended user task effectively showcases the dynamic multi-tool capabilities by:

Combining web search (Brave) and GitHub operations in a single request

Creating a logical flow from information gathering to action execution

Demonstrating the agent's ability to chain multiple tools together

This aligns perfectly with the PR objective of converting from a single-tool to multi-tool demonstration.

✨ Finishing Touches

📝 Generate Docstrings

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.

❤️ Share

🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

Review comments: Directly reply to a review comment made by CodeRabbit. Example:
- I pushed a fix in commit <commit_id>, please review it.
- Explain this complex logic.
- Open a follow-up GitHub issue for this discussion.
Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
- @coderabbitai explain this code block.
- @coderabbitai modularize this function.
PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
- @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
- @coderabbitai read src/utils.ts and explain its main purpose.
- @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
- @coderabbitai help me debug CodeRabbit configuration file.

Support

Need help? Create a ticket on our support page for assistance with any issues or questions.

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)

@coderabbitai pause to pause the reviews on a PR.
@coderabbitai resume to resume the paused reviews.
@coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
@coderabbitai full review to do a full review from scratch and review all the files again.
@coderabbitai summary to regenerate the summary of the PR.
@coderabbitai generate docstrings to generate docstrings for this PR.
@coderabbitai generate sequence diagram to generate a sequence diagram of the changes in this PR.
@coderabbitai resolve resolve all the CodeRabbit review comments.
@coderabbitai configuration to show the current CodeRabbit configuration for the repository.
@coderabbitai help to get help.

Other keywords and placeholders

Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
Add @coderabbitai anywhere in the PR title to generate the title automatically.

CodeRabbit Configuration File (`.coderabbit.yaml`)

You can programmatically configure CodeRabbit by adding a .coderabbit.yaml file to the root of your repository.
Please see the configuration documentation for more information.
If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

Documentation and Community

Visit our Documentation for detailed information on how to use CodeRabbit.
Join our Discord Community to get help, request features, and share feedback.
Follow us on X/Twitter for updates and announcements.

recurseml · 2025-07-14T11:16:12Z

✨ No issues found! Your code is sparkling clean! ✨

^{Need help? Join our Discord for support!
https://discord.gg/qEjHQk64Z9}

thisisfixer · 2025-07-23T19:29:10Z

examples/mistral/agent_with_dynamic_tool_discovery_pattern_1.py

    "ACI_SEARCH_FUNCTIONS"
-    "You can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."
-    "Once you have identified the functions you need to use, you can append them to the tools list and use them in future tool calls."
+    "Taking complete user input as parameter value, you can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."


Suggested change

"Taking complete user input as parameter value, you can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."

"Based on user intent, you can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."

Hey, Could you please check if the changes work as expected? I’ve updated the prompt to better handle complex queries involving multiple tools.
Previously, If just "based on the user query", the search tool didn’t always return all the necessary tools then the model will output "can't ……" as the ouput.

I think this was because the model couldn’t generate suitable values for the search tool, so I updated the prompt to take the full user input as the parameter value.

thisisfixer · 2025-07-23T19:30:12Z

examples/mistral/agent_with_dynamic_tool_discovery_pattern_1.py

-    "You can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."
-    "Once you have identified the functions you need to use, you can append them to the tools list and use them in future tool calls."
+    "Taking complete user input as parameter value, you can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."
+    "Once you have identified the functions you need to use, you can refresh the tools list and use them in future tool calls."


"Once you have identified the functions you need to use, you can refresh the tools list and use them in future tool calls." this should not be part of the prompt, as it is orchestrated by you (dev) not the agent/LLM directly.

examples/mistral/agent_with_dynamic_tool_discovery_pattern_1.py

YueTwo added 2 commits July 3, 2025 19:46

refactor(mistral): enhance example with dynamic(pattern1) multi-tool …

d5bb001

…support

modified the system promt and changed the model

2fb4717

thisisfixer reviewed Jul 23, 2025

View reviewed changes

thisisfixer requested changes Jul 23, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

refactor(mistral): enhance example with dynamic(pattern1) multi-tool support #28

refactor(mistral): enhance example with dynamic(pattern1) multi-tool support #28

Uh oh!

YueTwo commented Jul 4, 2025 •

edited by coderabbitai bot

Loading

Uh oh!

coderabbitai bot commented Jul 4, 2025 •

edited

Loading

Chat

Support

CodeRabbit Commands (Invoked using PR comments)

Other keywords and placeholders

CodeRabbit Configuration File (`.coderabbit.yaml`)

Documentation and Community

Uh oh!

recurseml bot commented Jul 14, 2025

Uh oh!

thisisfixer Jul 23, 2025

Uh oh!

YueTwo Jul 25, 2025

Uh oh!

YueTwo Jul 25, 2025

Uh oh!

thisisfixer Jul 23, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	"Taking complete user input as parameter value, you can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."
	"Based on user intent, you can use ACI_SEARCH_FUNCTIONS to find relevant functions across all apps."

refactor(mistral): enhance example with dynamic(pattern1) multi-tool support #28

Are you sure you want to change the base?

refactor(mistral): enhance example with dynamic(pattern1) multi-tool support #28

Uh oh!

Conversation

YueTwo commented Jul 4, 2025 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Uh oh!

coderabbitai bot commented Jul 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Possibly related PRs

Suggested reviewers

Poem

Chat

Support

CodeRabbit Commands (Invoked using PR comments)

Other keywords and placeholders

CodeRabbit Configuration File (.coderabbit.yaml)

Documentation and Community

Uh oh!

recurseml bot commented Jul 14, 2025

Uh oh!

thisisfixer Jul 23, 2025

Choose a reason for hiding this comment

Uh oh!

YueTwo Jul 25, 2025

Choose a reason for hiding this comment

Uh oh!

YueTwo Jul 25, 2025

Choose a reason for hiding this comment

Uh oh!

thisisfixer Jul 23, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

YueTwo commented Jul 4, 2025 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Jul 4, 2025 •

edited

Loading

CodeRabbit Configuration File (`.coderabbit.yaml`)