Restructre SWE Agent

[kaavee315]

Description

• Enhanced the SubmitPatchRequest and SubmitPatch class descriptions to include details about the diff format.
• Simplified CoderAgentArgs initialization and updated method calls across multiple files.
• Introduced BaseSWEAgent abstract class.
• Added REVIEWER_BACKSTORY_TMPL template for reviewer agent.
• Refactored CoderAgent to inherit from BaseSWEAgent and added reviewer agent to the workflow.
• Added utility functions for logger setup and LLM retrieval.
• Updated README to include additional setup steps and environment variables.
• Changed the patch generated to be static instead of agentic.

---

Changes walkthrough 📝

	Relevant files
Enhancement	8 files submit_patch.py Improve patch submission descriptions and examples              python/composio/local_tools/local_workspace/submit_patch/actions/submit_patch.py Enhanced the SubmitPatchRequest and SubmitPatch class descriptions to include details about the diff format. +11/-2    run_evaluation.py Simplify CoderAgentArgs initialization and method call      python/composio_swe/benchmark/run_evaluation.py Simplified CoderAgentArgs initialization by removing redundant parameters. Updated coder.run method to accept issue_config. +2/-6      base_swe_agent.py Add BaseSWEAgent abstract class                                                    python/composio_swe/composio_swe/agent/base_swe_agent.py Introduced BaseSWEAgent abstract class with an abstract run method. +12/-0    prompts.py Add reviewer backstory template                                                    python/composio_swe/composio_swe/agent/prompts.py - Added `REVIEWER_BACKSTORY_TMPL` template for reviewer agent. +12/-0    swe.py Refactor CoderAgent and add reviewer agent                              python/composio_swe/composio_swe/agent/swe.py Refactored CoderAgent to inherit from BaseSWEAgent. Moved logger setup and LLM retrieval to utils.py. Added reviewer agent and task to the workflow. +59/-80  swe_run.py Update CoderAgentArgs initialization and method call          python/composio_swe/composio_swe/agent/swe_run.py Updated CoderAgentArgs initialization to remove redundant parameters. Updated coder_agent.run method to accept issue_config. +4/-5      utils.py Add utility functions for logger and LLM                                  python/composio_swe/composio_swe/agent/utils.py - Added utility functions for logger setup and LLM retrieval. +34/-0    cli.py Simplify solve command initialization and method call        python/composio_swe/composio_swe/cli/cli.py Updated solve command to simplify CoderAgentArgs initialization and method call. +5/-6
Documentation	1 files README.md Update README with additional setup steps                                python/composio_swe/composio_swe/agent/README.md Updated README to include additional setup steps and environment variables. +14/-6

---

💡 PR-Agent usage: Comment /help on the PR to get a list of all available PR-Agent tools and their descriptions

[codiumai-pr-agent-pro[bot]]

PR Reviewer Guide 🔍

⏱️ Estimated effort to review [1-5]	4
🧪 Relevant tests	No
🔒 Security concerns	No
⚡ Key issues to review	Possible Bug: The refactoring in run_evaluation.py and swe_run.py simplifies the initialization of CoderAgentArgs but removes the passing of issue_config directly. This could potentially lead to issues if the issue_config is required in the initialization process and not just in the run method.
	Code Clarity: The changes in submit_patch.py enhance the description of patch_code but might benefit from further examples or clearer explanation of what constitutes a "valid patch in diff format."
	Refactoring Impact: The introduction of the BaseSWEAgent class and changes in inheritance for CoderAgent are significant. It's important to ensure that all functionalities of the original CoderAgent are preserved or appropriately modified to fit the new structure.

[codiumai-pr-agent-pro[bot]]

PR Code Suggestions ✨

Category	Suggestion	Score
Possible bug	Add validation to ensure issue_config is not None before using it in the run method ___ **The run method should validate that issue_config is not None before proceeding, to prevent potential runtime errors.** [python/composio_swe/composio_swe/agent/swe.py [119-121]](https://github.com/ComposioHQ/composio/pull/204/files#diff-bbb8fcd0bfdb219cf4e20ffd41e3695abb277d9a5efc2c5a315e478976655540R119-R121) ```diff def run(self, issue_config: IssueConfig): + if issue_config is None: + raise ValueError("issue_config must be provided") llm = get_llm() repo_name = issue_config.repo_name ``` - [ ] **Apply this suggestion** Suggestion importance[1-10]: 8 Why: This is a valid suggestion to prevent runtime errors by ensuring that `issue_config` is provided before proceeding with the method execution.	8
Enhancement	Log an error message before raising a ValueError in the get_llm function ___ **The get_llm function should log an error message before raising a ValueError when no model is found, to aid in debugging.** [python/composio_swe/composio_swe/agent/utils.py [34]](https://github.com/ComposioHQ/composio/pull/204/files#diff-ebba6012e3781136b15802f8496f063323a2d11aba51c645d1d8a5819486dab2R34-R34) ```diff +logger.error("No valid API key found for LLM configuration") raise ValueError("no model is found") ``` - [ ] **Apply this suggestion** Suggestion importance[1-10]: 7 Why: Adding an error log before raising an exception is a good practice for better debugging and understanding the error context, making this a useful enhancement.	7
Best practice	Add a try-except block to the run method to handle and log exceptions ___ **The run method should include a try-except block around the main logic to handle potential exceptions gracefully and log them for debugging purposes.** [python/composio_swe/composio_swe/agent/swe.py [119-123]](https://github.com/ComposioHQ/composio/pull/204/files#diff-bbb8fcd0bfdb219cf4e20ffd41e3695abb277d9a5efc2c5a315e478976655540R119-R123) ```diff def run(self, issue_config: IssueConfig): - llm = get_llm() - repo_name = issue_config.repo_name - if not repo_name: - raise ValueError("no repo-name configuration is found") + try: + llm = get_llm() + repo_name = issue_config.repo_name + if not repo_name: + raise ValueError("no repo-name configuration is found") ``` - [ ] **Apply this suggestion** Suggestion importance[1-10]: 6 Why: Implementing a try-except block for exception handling in the `run` method is a good practice to manage exceptions and maintain the robustness of the application.	6