xiaxinhang/pr-agent
1
0
Fork 0
mirror of https://github.com/qodo-ai/pr-agent.git synced 2025-07-10 07:40:39 +08:00
Code Issues Packages Projects Releases Wiki Activity

docs: improve /implement tool documentation and update news section

Browse Source
This commit is contained in:
mrT23
2025-01-02 16:49:59 +02:00
parent f6b80174b5
commit b570c758aa
3 changed files with 25 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
docs/docs/tools/implement.md
View File

@ -1,12 +1,12 @@
## Overview
The `implement` tool automatically generates implementation code based on PR review suggestions.
It combines LLM capabilities with PR review suggestions to help developers implement code changes quickly and with confidence.
The `implement` tool converts human code review discussions and feedback into ready-to-commit code changes.
It leverages LLM technology to transform PR comments and review suggestions into concrete implementation code, helping developers quickly turn feedback into working solutions.
## Usage Scenarios
### 1. For Reviewers
### For Reviewers
Reviewers can request code changes by: <br>
1. Selecting the code block to be modified. <br>
@ -15,10 +15,10 @@ Reviewers can request code changes by: <br>
/implement <code-change-description>
```
![implement1](https://codium.ai/images/pr_agent/implement1.png){width=512}
![implement1](https://codium.ai/images/pr_agent/implement1.png){width=640}
### 2. For PR Authors
### For PR Authors
PR authors can implement suggested changes by replying to a review comment using either: <br>
1. Add specific implementation details as described above
@ -30,16 +30,16 @@ PR authors can implement suggested changes by replying to a review comment using
/implement
```
![implement2](https://codium.ai/images/pr_agent/implement2.png){width=512}
![implement2](https://codium.ai/images/pr_agent/implement2.png){width=640}
### 3. For Referencing Comments
### For Referencing Comments
You can reference and implement changes from any comment by:
```
/implement <link-to-review-comment>
```
![implement3](https://codium.ai/images/pr_agent/implement3.png){width=512}
![implement3](https://codium.ai/images/pr_agent/implement3.png){width=640}
Note that the implementation will occur within the review discussion thread.

7
docs/docs/usage-guide/changing_a_model.md
View File

@ -40,7 +40,7 @@ E.g. to use a new model locally via Ollama, set in `.secrets.toml` or in a confi
model = "ollama/qwen2.5-coder:32b"
fallback_models=["ollama/qwen2.5-coder:32b"]
custom_model_max_tokens=128000 # set the maximal input tokens for the model
duplicate_examples=true # will duplicate the examples in the prompt, to help the model to output structured output
duplicate_examples=true # will duplicate the examples in the prompt, to help the model to generate structured output
[ollama]
api_base = "http://localhost:11434" # or whatever port you're running Ollama on
@ -48,11 +48,14 @@ api_base = "http://localhost:11434" # or whatever port you're running Ollama on
!!! note "Local models vs commercial models"
Qodo Merge is compatible with almost any AI model, but analyzing complex code repositories and pull requests requires a model specifically optimized for code analysis.
Commercial models such as GPT-4, Claude Sonnet, and Gemini have demonstrated robust capabilities in generating structured output for code analysis tasks with large input. In contrast, most open-source models currently available (as of January 2025) face challenges with these complex tasks.
Based on our testing, local open-source models are suitable for experimentation and learning purposes, but they are not suitable for production-level code analysis tasks.
Hence, for production workflows and real-world usage, we recommend using commercial models.
### Hugging Face Inference Endpoints
### Hugging Face
To use a new model with Hugging Face Inference Endpoints, for example, set:
```