code-review-ai-cli 1.0.0__tar.gz → 1.0.1__tar.gz

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-review-ai-cli
-Version: 1.0.0
+Version: 1.0.1
 Summary: Automated AI-powered code review CLI for Azure DevOps / TFS Pull Requests
 License: MIT
 Keywords: code-review,ai,azure-devops,tfs,pull-request,llm
@@ -78,22 +78,26 @@ pip install "code-review-ai-cli[dev]"
 
 ## Configuration
 
-After installing the package, generate a ready-to-edit `config.yaml` in your working directory:
+After installing the package, generate ready-to-edit configuration files in your working directory:
 
 ```bash
 ai-review init
 ```
 
-This copies the bundled template with all available options and inline documentation:
+This copies two bundled templates:
 
-```
+- **`config.yaml`** — all available options with inline documentation
+- **`review_prompt.md`** — default review style rules, injected into every LLM prompt
+
+```text
 ✅ config.yaml created at: /home/user/my-project/config.yaml
-   Edit it to add your credentials and preferences.
+✅ review_prompt.md created at: /home/user/my-project/review_prompt.md
+   Edit them to add your credentials, preferences and review rules.
 ```
 
-If a `config.yaml` already exists you will be prompted before it is overwritten:
+If either file already exists you will be prompted individually before it is overwritten:
 
-```
+```text
 config.yaml already exists in the current directory.
 Overwrite? [y/N]
 ```
@@ -165,23 +169,22 @@ review:
 
 ### Markdown-Customizable Prompt
 
-You can adjust review rules, context, and examples in `review_prompt.md`.
-This file is loaded automatically and injected into LLM instructions on each run.
-
-Example in `config.yaml`:
-
-```yaml
-review:
-  custom_prompt_file: review_prompt.md
-```
+`ai-review init` creates a `review_prompt.md` alongside `config.yaml`. This file is loaded automatically and injected into LLM instructions on every run.
 
-Suggested usage for this file:
+Edit it to tailor the review to your team:
 
 - Define comment tone and format
 - Add mandatory validation rules
 - Include business/architecture context
 - Add examples of good/bad comments
 
+The path is configurable in `config.yaml` (default: `review_prompt.md` in the current directory):
+
+```yaml
+review:
+  custom_prompt_file: review_prompt.md
+```
+
 ### Bedrock Example
 
 ```yaml
@@ -201,50 +204,23 @@ tfs:
   project: ProjectName
   pat: xxxxxxxxx
 ```
+
 Bedrock notes:
 
 - You can use `profile` or explicit credentials in YAML.
 - If explicit credentials are not defined, the AWS SDK uses the default credentials chain.
 
-## Review Flow
-
-The main `pr-review` workflow is:
-
-1. load and validate configuration
-2. fetch Pull Request metadata from Azure DevOps/TFS
-3. fetch the PR diff
-4. filter and truncate the diff according to configuration
-5. request textual analysis and structured comments from the LLM provider
-6. display a preview in the terminal
-7. post inline or general PR comments when applicable
-
-## Tests
-
-The project's functional coverage is reflected in the `tests/` folder, including:
-
-- `tests/test_ai_review.py` for the CLI and main workflow
-- `tests/test_config.py` for configuration and validation
-- `tests/test_formatter.py` for terminal/markdown/json rendering
-- `tests/test_git_utils.py` for diffs and Git utilities
-- `tests/test_llm_client.py` for prompts, parsing, and LLM providers
-- `tests/test_tfs_client.py` for TFS/Azure DevOps integration
-
-Run the suite:
-
-```bash
-python -m pytest --cov=src --cov-report=term
-```
-
 ## CLI Usage
 
 ### Help
+
 ```bash
-python src/ai_review.py --help
+ai-review pr-review --help
 ```
 
 ### Bootstrap configuration
 
-Generate a `config.yaml` template in the current directory:
+Generate `config.yaml` and `review_prompt.md` templates in the current directory:
 
 ```bash
 ai-review init
@@ -253,7 +229,7 @@ ai-review init
 ### Interactive Mode
 
 ```bash
-python ai_review.py
+ ai-review pr-review
 ```
 
 ### Pull Request Review
@@ -261,51 +237,51 @@ python ai_review.py
 List PRs and select interactively:
 
 ```bash
-python ai_review.py pr-review
+ ai-review pr-review pr-review
 ```
 
 Review a specific PR:
 
 ```bash
-python ai_review.py pr-review 42
+ ai-review pr-review pr-review 42
 ```
 
 Dry-run:
 
 ```bash
-python ai_review.py pr-review 42 --dry-run
+ ai-review pr-review pr-review 42 --dry-run
 ```
 
 Full review of changed files (in addition to diff-focused review):
 
 ```bash
-python ai_review.py pr-review 42 --review-scope full_code
+ ai-review pr-review pr-review 42 --review-scope full_code
 ```
 
 Automatic posting (without confirmation):
 
 ```bash
-python ai_review.py pr-review 42 --auto-post
+ ai-review pr-review pr-review 42 --auto-post
 ```
 
 Filter PRs in interactive selection:
 
 ```bash
-python ai_review.py pr-review --author "John Smith" --target-branch main
+ ai-review pr-review pr-review --author "John Smith" --target-branch main
 ```
 
 Choose provider/model via CLI:
 
 ```bash
-python ai_review.py pr-review 42 --provider bedrock --model anthropic.claude-3-5-sonnet-20240620-v1:0
+ ai-review pr-review pr-review 42 --provider bedrock --model anthropic.claude-3-5-sonnet-20240620-v1:0
 ```
 
 ### List Pull Requests
 
 ```bash
-python ai_review.py list-prs
-python ai_review.py list-prs --status completed
-python ai_review.py list-prs --repo-name backend --author "John"
+ ai-review pr-review list-prs
+ ai-review pr-review list-prs --status completed
+ ai-review pr-review list-prs --repo-name backend --author "John"
 ```
 
 ## Execution Flow
@@ -314,7 +290,7 @@ The diagram below summarizes how the review application moves from CLI entry to
 
 ```mermaid
 flowchart TD
-  A[Start: python ai_review.py] --> B{Arguments provided?}
+  A[Start:  ai-review pr-review] --> B{Arguments provided?}
   B -->|No| C[Interactive mode]
   B -->|Yes| D[Parse CLI command]
 
@@ -364,7 +340,7 @@ flowchart TD
 ### `pr-review`
 
 ```bash
-python ai_review.py pr-review [pr_id]
+ ai-review pr-review pr-review [pr_id]
 ```
 
 Options:
@@ -388,7 +364,7 @@ Options:
 ### `list-prs`
 
 ```bash
-python ai_review.py list-prs
+ ai-review pr-review list-prs
 ```
 
 Options:
@@ -399,10 +375,10 @@ Options:
 
 ## Available VS Code Tasks
 
-- `🌟 AI Review: Pull Request (Interactive)`
-- `🌟 AI Review: PR (Dry-Run)`
-- `📋 AI Review: List Active PRs`
-- `🤖 AI Review: Interactive Mode`
+- `AI Review: Pull Request (Interactive)`
+- `AI Review: PR (Dry-Run)`
+- `AI Review: List Active PRs`
+- `AI Review: Interactive Mode`
 
 ## Troubleshooting
 
@@ -424,18 +400,19 @@ Avoid `verify_ssl: false` except for temporary troubleshooting.
 - Confirm `llm.model` with a valid Bedrock model ID in the chosen region.
 - Validate AWS credentials (`profile` or explicit keys).
 
-## Architecture
+## Tests
 
-```text
-ai_code_review_script/
-├── src/
-│   ├── config.py         # YAML-only configuration
-│   ├── ai_review.py      # Main PR CLI and workflow
-│   ├── llm_client.py     # LLM provider integration (includes Bedrock)
-│   ├── tfs_client.py     # Azure DevOps/TFS (PRs and comments)
-│   ├── formatter.py      # Output formatting
-│   └── git_utils.py      # Utilities for diff parsing/truncation
-├── config.yaml           # Single configuration file
-├── requirements.txt      # Python dependencies
-└── .vscode/tasks.json    # PR review tasks
+The project's functional coverage is reflected in the `tests/` folder, including:
+
+- `tests/test_ai_review.py` for the CLI and main workflow
+- `tests/test_config.py` for configuration and validation
+- `tests/test_formatter.py` for terminal/markdown/json rendering
+- `tests/test_git_utils.py` for diffs and Git utilities
+- `tests/test_llm_client.py` for prompts, parsing, and LLM providers
+- `tests/test_tfs_client.py` for TFS/Azure DevOps integration
+
+Run the suite:
+
+```bash
+python -m pytest --cov=src --cov-report=term
 ```

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/README.md

@@ -41,22 +41,26 @@ pip install "code-review-ai-cli[dev]"
 
 ## Configuration
 
-After installing the package, generate a ready-to-edit `config.yaml` in your working directory:
+After installing the package, generate ready-to-edit configuration files in your working directory:
 
 ```bash
 ai-review init
 ```
 
-This copies the bundled template with all available options and inline documentation:
+This copies two bundled templates:
 
-```
+- **`config.yaml`** — all available options with inline documentation
+- **`review_prompt.md`** — default review style rules, injected into every LLM prompt
+
+```text
 ✅ config.yaml created at: /home/user/my-project/config.yaml
-   Edit it to add your credentials and preferences.
+✅ review_prompt.md created at: /home/user/my-project/review_prompt.md
+   Edit them to add your credentials, preferences and review rules.
 ```
 
-If a `config.yaml` already exists you will be prompted before it is overwritten:
+If either file already exists you will be prompted individually before it is overwritten:
 
-```
+```text
 config.yaml already exists in the current directory.
 Overwrite? [y/N]
 ```
@@ -128,23 +132,22 @@ review:
 
 ### Markdown-Customizable Prompt
 
-You can adjust review rules, context, and examples in `review_prompt.md`.
-This file is loaded automatically and injected into LLM instructions on each run.
-
-Example in `config.yaml`:
-
-```yaml
-review:
-  custom_prompt_file: review_prompt.md
-```
+`ai-review init` creates a `review_prompt.md` alongside `config.yaml`. This file is loaded automatically and injected into LLM instructions on every run.
 
-Suggested usage for this file:
+Edit it to tailor the review to your team:
 
 - Define comment tone and format
 - Add mandatory validation rules
 - Include business/architecture context
 - Add examples of good/bad comments
 
+The path is configurable in `config.yaml` (default: `review_prompt.md` in the current directory):
+
+```yaml
+review:
+  custom_prompt_file: review_prompt.md
+```
+
 ### Bedrock Example
 
 ```yaml
@@ -164,50 +167,23 @@ tfs:
   project: ProjectName
   pat: xxxxxxxxx
 ```
+
 Bedrock notes:
 
 - You can use `profile` or explicit credentials in YAML.
 - If explicit credentials are not defined, the AWS SDK uses the default credentials chain.
 
-## Review Flow
-
-The main `pr-review` workflow is:
-
-1. load and validate configuration
-2. fetch Pull Request metadata from Azure DevOps/TFS
-3. fetch the PR diff
-4. filter and truncate the diff according to configuration
-5. request textual analysis and structured comments from the LLM provider
-6. display a preview in the terminal
-7. post inline or general PR comments when applicable
-
-## Tests
-
-The project's functional coverage is reflected in the `tests/` folder, including:
-
-- `tests/test_ai_review.py` for the CLI and main workflow
-- `tests/test_config.py` for configuration and validation
-- `tests/test_formatter.py` for terminal/markdown/json rendering
-- `tests/test_git_utils.py` for diffs and Git utilities
-- `tests/test_llm_client.py` for prompts, parsing, and LLM providers
-- `tests/test_tfs_client.py` for TFS/Azure DevOps integration
-
-Run the suite:
-
-```bash
-python -m pytest --cov=src --cov-report=term
-```
-
 ## CLI Usage
 
 ### Help
+
 ```bash
-python src/ai_review.py --help
+ai-review pr-review --help
 ```
 
 ### Bootstrap configuration
 
-Generate a `config.yaml` template in the current directory:
+Generate `config.yaml` and `review_prompt.md` templates in the current directory:
 
 ```bash
 ai-review init
@@ -216,7 +192,7 @@ ai-review init
 ### Interactive Mode
 
 ```bash
-python ai_review.py
+ ai-review pr-review
 ```
 
 ### Pull Request Review
@@ -224,51 +200,51 @@ python ai_review.py
 List PRs and select interactively:
 
 ```bash
-python ai_review.py pr-review
+ ai-review pr-review pr-review
 ```
 
 Review a specific PR:
 
 ```bash
-python ai_review.py pr-review 42
+ ai-review pr-review pr-review 42
 ```
 
 Dry-run:
 
 ```bash
-python ai_review.py pr-review 42 --dry-run
+ ai-review pr-review pr-review 42 --dry-run
 ```
 
 Full review of changed files (in addition to diff-focused review):
 
 ```bash
-python ai_review.py pr-review 42 --review-scope full_code
+ ai-review pr-review pr-review 42 --review-scope full_code
 ```
 
 Automatic posting (without confirmation):
 
 ```bash
-python ai_review.py pr-review 42 --auto-post
+ ai-review pr-review pr-review 42 --auto-post
 ```
 
 Filter PRs in interactive selection:
 
 ```bash
-python ai_review.py pr-review --author "John Smith" --target-branch main
+ ai-review pr-review pr-review --author "John Smith" --target-branch main
 ```
 
 Choose provider/model via CLI:
 
 ```bash
-python ai_review.py pr-review 42 --provider bedrock --model anthropic.claude-3-5-sonnet-20240620-v1:0
+ ai-review pr-review pr-review 42 --provider bedrock --model anthropic.claude-3-5-sonnet-20240620-v1:0
 ```
 
 ### List Pull Requests
 
 ```bash
-python ai_review.py list-prs
-python ai_review.py list-prs --status completed
-python ai_review.py list-prs --repo-name backend --author "John"
+ ai-review pr-review list-prs
+ ai-review pr-review list-prs --status completed
+ ai-review pr-review list-prs --repo-name backend --author "John"
 ```
 
 ## Execution Flow
@@ -277,7 +253,7 @@ The diagram below summarizes how the review application moves from CLI entry to
 
 ```mermaid
 flowchart TD
-  A[Start: python ai_review.py] --> B{Arguments provided?}
+  A[Start:  ai-review pr-review] --> B{Arguments provided?}
   B -->|No| C[Interactive mode]
   B -->|Yes| D[Parse CLI command]
 
@@ -327,7 +303,7 @@ flowchart TD
 ### `pr-review`
 
 ```bash
-python ai_review.py pr-review [pr_id]
+ ai-review pr-review pr-review [pr_id]
 ```
 
 Options:
@@ -351,7 +327,7 @@ Options:
 ### `list-prs`
 
 ```bash
-python ai_review.py list-prs
+ ai-review pr-review list-prs
 ```
 
 Options:
@@ -362,10 +338,10 @@ Options:
 
 ## Available VS Code Tasks
 
-- `🌟 AI Review: Pull Request (Interactive)`
-- `🌟 AI Review: PR (Dry-Run)`
-- `📋 AI Review: List Active PRs`
-- `🤖 AI Review: Interactive Mode`
+- `AI Review: Pull Request (Interactive)`
+- `AI Review: PR (Dry-Run)`
+- `AI Review: List Active PRs`
+- `AI Review: Interactive Mode`
 
 ## Troubleshooting
 
@@ -387,18 +363,19 @@ Avoid `verify_ssl: false` except for temporary troubleshooting.
 - Confirm `llm.model` with a valid Bedrock model ID in the chosen region.
 - Validate AWS credentials (`profile` or explicit keys).
 
-## Architecture
+## Tests
 
-```text
-ai_code_review_script/
-├── src/
-│   ├── config.py         # YAML-only configuration
-│   ├── ai_review.py      # Main PR CLI and workflow
-│   ├── llm_client.py     # LLM provider integration (includes Bedrock)
-│   ├── tfs_client.py     # Azure DevOps/TFS (PRs and comments)
-│   ├── formatter.py      # Output formatting
-│   └── git_utils.py      # Utilities for diff parsing/truncation
-├── config.yaml           # Single configuration file
-├── requirements.txt      # Python dependencies
-└── .vscode/tasks.json    # PR review tasks
+The project's functional coverage is reflected in the `tests/` folder, including:
+
+- `tests/test_ai_review.py` for the CLI and main workflow
+- `tests/test_config.py` for configuration and validation
+- `tests/test_formatter.py` for terminal/markdown/json rendering
+- `tests/test_git_utils.py` for diffs and Git utilities
+- `tests/test_llm_client.py` for prompts, parsing, and LLM providers
+- `tests/test_tfs_client.py` for TFS/Azure DevOps integration
+
+Run the suite:
+
+```bash
+python -m pytest --cov=src --cov-report=term
 ```

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/code_review_ai_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-review-ai-cli
-Version: 1.0.0
+Version: 1.0.1
 Summary: Automated AI-powered code review CLI for Azure DevOps / TFS Pull Requests
 License: MIT
 Keywords: code-review,ai,azure-devops,tfs,pull-request,llm
@@ -78,22 +78,26 @@ pip install "code-review-ai-cli[dev]"
 
 ## Configuration
 
-After installing the package, generate a ready-to-edit `config.yaml` in your working directory:
+After installing the package, generate ready-to-edit configuration files in your working directory:
 
 ```bash
 ai-review init
 ```
 
-This copies the bundled template with all available options and inline documentation:
+This copies two bundled templates:
 
-```
+- **`config.yaml`** — all available options with inline documentation
+- **`review_prompt.md`** — default review style rules, injected into every LLM prompt
+
+```text
 ✅ config.yaml created at: /home/user/my-project/config.yaml
-   Edit it to add your credentials and preferences.
+✅ review_prompt.md created at: /home/user/my-project/review_prompt.md
+   Edit them to add your credentials, preferences and review rules.
 ```
 
-If a `config.yaml` already exists you will be prompted before it is overwritten:
+If either file already exists you will be prompted individually before it is overwritten:
 
-```
+```text
 config.yaml already exists in the current directory.
 Overwrite? [y/N]
 ```
@@ -165,23 +169,22 @@ review:
 
 ### Markdown-Customizable Prompt
 
-You can adjust review rules, context, and examples in `review_prompt.md`.
-This file is loaded automatically and injected into LLM instructions on each run.
-
-Example in `config.yaml`:
-
-```yaml
-review:
-  custom_prompt_file: review_prompt.md
-```
+`ai-review init` creates a `review_prompt.md` alongside `config.yaml`. This file is loaded automatically and injected into LLM instructions on every run.
 
-Suggested usage for this file:
+Edit it to tailor the review to your team:
 
 - Define comment tone and format
 - Add mandatory validation rules
 - Include business/architecture context
 - Add examples of good/bad comments
 
+The path is configurable in `config.yaml` (default: `review_prompt.md` in the current directory):
+
+```yaml
+review:
+  custom_prompt_file: review_prompt.md
+```
+
 ### Bedrock Example
 
 ```yaml
@@ -201,50 +204,23 @@ tfs:
   project: ProjectName
   pat: xxxxxxxxx
 ```
+
 Bedrock notes:
 
 - You can use `profile` or explicit credentials in YAML.
 - If explicit credentials are not defined, the AWS SDK uses the default credentials chain.
 
-## Review Flow
-
-The main `pr-review` workflow is:
-
-1. load and validate configuration
-2. fetch Pull Request metadata from Azure DevOps/TFS
-3. fetch the PR diff
-4. filter and truncate the diff according to configuration
-5. request textual analysis and structured comments from the LLM provider
-6. display a preview in the terminal
-7. post inline or general PR comments when applicable
-
-## Tests
-
-The project's functional coverage is reflected in the `tests/` folder, including:
-
-- `tests/test_ai_review.py` for the CLI and main workflow
-- `tests/test_config.py` for configuration and validation
-- `tests/test_formatter.py` for terminal/markdown/json rendering
-- `tests/test_git_utils.py` for diffs and Git utilities
-- `tests/test_llm_client.py` for prompts, parsing, and LLM providers
-- `tests/test_tfs_client.py` for TFS/Azure DevOps integration
-
-Run the suite:
-
-```bash
-python -m pytest --cov=src --cov-report=term
-```
-
 ## CLI Usage
 
 ### Help
+
 ```bash
-python src/ai_review.py --help
+ai-review pr-review --help
 ```
 
 ### Bootstrap configuration
 
-Generate a `config.yaml` template in the current directory:
+Generate `config.yaml` and `review_prompt.md` templates in the current directory:
 
 ```bash
 ai-review init
@@ -253,7 +229,7 @@ ai-review init
 ### Interactive Mode
 
 ```bash
-python ai_review.py
+ ai-review pr-review
 ```
 
 ### Pull Request Review
@@ -261,51 +237,51 @@ python ai_review.py
 List PRs and select interactively:
 
 ```bash
-python ai_review.py pr-review
+ ai-review pr-review pr-review
 ```
 
 Review a specific PR:
 
 ```bash
-python ai_review.py pr-review 42
+ ai-review pr-review pr-review 42
 ```
 
 Dry-run:
 
 ```bash
-python ai_review.py pr-review 42 --dry-run
+ ai-review pr-review pr-review 42 --dry-run
 ```
 
 Full review of changed files (in addition to diff-focused review):
 
 ```bash
-python ai_review.py pr-review 42 --review-scope full_code
+ ai-review pr-review pr-review 42 --review-scope full_code
 ```
 
 Automatic posting (without confirmation):
 
 ```bash
-python ai_review.py pr-review 42 --auto-post
+ ai-review pr-review pr-review 42 --auto-post
 ```
 
 Filter PRs in interactive selection:
 
 ```bash
-python ai_review.py pr-review --author "John Smith" --target-branch main
+ ai-review pr-review pr-review --author "John Smith" --target-branch main
 ```
 
 Choose provider/model via CLI:
 
 ```bash
-python ai_review.py pr-review 42 --provider bedrock --model anthropic.claude-3-5-sonnet-20240620-v1:0
+ ai-review pr-review pr-review 42 --provider bedrock --model anthropic.claude-3-5-sonnet-20240620-v1:0
 ```
 
 ### List Pull Requests
 
 ```bash
-python ai_review.py list-prs
-python ai_review.py list-prs --status completed
-python ai_review.py list-prs --repo-name backend --author "John"
+ ai-review pr-review list-prs
+ ai-review pr-review list-prs --status completed
+ ai-review pr-review list-prs --repo-name backend --author "John"
 ```
 
 ## Execution Flow
@@ -314,7 +290,7 @@ The diagram below summarizes how the review application moves from CLI entry to
 
 ```mermaid
 flowchart TD
-  A[Start: python ai_review.py] --> B{Arguments provided?}
+  A[Start:  ai-review pr-review] --> B{Arguments provided?}
   B -->|No| C[Interactive mode]
   B -->|Yes| D[Parse CLI command]
 
@@ -364,7 +340,7 @@ flowchart TD
 ### `pr-review`
 
 ```bash
-python ai_review.py pr-review [pr_id]
+ ai-review pr-review pr-review [pr_id]
 ```
 
 Options:
@@ -388,7 +364,7 @@ Options:
 ### `list-prs`
 
 ```bash
-python ai_review.py list-prs
+ ai-review pr-review list-prs
 ```
 
 Options:
@@ -399,10 +375,10 @@ Options:
 
 ## Available VS Code Tasks
 
-- `🌟 AI Review: Pull Request (Interactive)`
-- `🌟 AI Review: PR (Dry-Run)`
-- `📋 AI Review: List Active PRs`
-- `🤖 AI Review: Interactive Mode`
+- `AI Review: Pull Request (Interactive)`
+- `AI Review: PR (Dry-Run)`
+- `AI Review: List Active PRs`
+- `AI Review: Interactive Mode`
 
 ## Troubleshooting
 
@@ -424,18 +400,19 @@ Avoid `verify_ssl: false` except for temporary troubleshooting.
 - Confirm `llm.model` with a valid Bedrock model ID in the chosen region.
 - Validate AWS credentials (`profile` or explicit keys).
 
-## Architecture
+## Tests
 
-```text
-ai_code_review_script/
-├── src/
-│   ├── config.py         # YAML-only configuration
-│   ├── ai_review.py      # Main PR CLI and workflow
-│   ├── llm_client.py     # LLM provider integration (includes Bedrock)
-│   ├── tfs_client.py     # Azure DevOps/TFS (PRs and comments)
-│   ├── formatter.py      # Output formatting
-│   └── git_utils.py      # Utilities for diff parsing/truncation
-├── config.yaml           # Single configuration file
-├── requirements.txt      # Python dependencies
-└── .vscode/tasks.json    # PR review tasks
+The project's functional coverage is reflected in the `tests/` folder, including:
+
+- `tests/test_ai_review.py` for the CLI and main workflow
+- `tests/test_config.py` for configuration and validation
+- `tests/test_formatter.py` for terminal/markdown/json rendering
+- `tests/test_git_utils.py` for diffs and Git utilities
+- `tests/test_llm_client.py` for prompts, parsing, and LLM providers
+- `tests/test_tfs_client.py` for TFS/Azure DevOps integration
+
+Run the suite:
+
+```bash
+python -m pytest --cov=src --cov-report=term
 ```

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/code_review_ai_cli.egg-info/SOURCES.txt

@@ -14,6 +14,7 @@ src/git_utils.py
 src/llm_client.py
 src/tfs_client.py
 src/prompts/config.yaml.template
+src/prompts/review_prompt.md.template
 tests/test_ai_review.py
 tests/test_config.py
 tests/test_formatter.py

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "code-review-ai-cli"
-version = "1.0.0"
+version = "1.0.1"
 description = "Automated AI-powered code review CLI for Azure DevOps / TFS Pull Requests"
 readme = "README.md"
 requires-python = ">=3.10"

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/src/ai_review.py

@@ -40,6 +40,7 @@ Version: see pyproject.toml
 """
 
 import argparse
+import importlib.resources
 import os
 import sys
 import time
@@ -872,16 +873,16 @@ def _show_config(config: ReviewConfig) -> None:
 # Init command
 # ---------------------------------------------------------------------------
 def cmd_init() -> int:
-    """Copies a config.yaml template to the current working directory.
+    """Copies a config.yaml template and review_prompt.md to the current working directory.
 
     Creates a ``config.yaml`` file pre-populated with all available options
-    and inline documentation. If the file already exists in the current
-    directory the user is prompted for confirmation before overwriting.
+    and inline documentation, and a ``review_prompt.md`` file with default
+    review style rules. If either file already exists in the current directory
+    the user is prompted for confirmation before overwriting.
 
-    The template is bundled with the package at
-    ``src/prompts/config.yaml.template`` and is resolved at runtime via
-    :mod:`importlib.resources`, so it works regardless of how the package
-    was installed.
+    Both files are bundled with the package at ``src/prompts/`` and are
+    resolved at runtime via :mod:`importlib.resources`, so they work
+    regardless of how the package was installed.
 
     Returns:
         int: Exit code. ``0`` on success or user cancellation, ``1`` on error.
@@ -891,14 +892,15 @@ def cmd_init() -> int:
 
             $ ai-review init
             ✅ config.yaml created at: /home/user/my-project/config.yaml
-               Edit it to add your credentials and preferences.
+            ✅ review_prompt.md created at: /home/user/my-project/review_prompt.md
+               Edit them to add your credentials, preferences and review rules.
     """
-    import importlib.resources as pkg_resources
-
-    dest = os.path.join(os.getcwd(), "config.yaml")
+    cwd = os.getcwd()
     c = Colors()
 
-    if os.path.exists(dest):
+    # --- config.yaml ---
+    config_dest = os.path.join(cwd, "config.yaml")
+    if os.path.exists(config_dest):
         print(f"{c.YELLOW}config.yaml already exists in the current directory.{c.RESET}")
         answer = input("Overwrite? [y/N] ").strip().lower()
         if answer != "y":
@@ -906,17 +908,38 @@ def cmd_init() -> int:
             return 0
 
     try:
-        ref = pkg_resources.files("src.prompts").joinpath("config.yaml.template")
-        template_content = ref.read_text(encoding="utf-8")
+        ref = importlib.resources.files("src.prompts").joinpath("config.yaml.template")
+        config_content = ref.read_text(encoding="utf-8")
+    except (FileNotFoundError, TypeError) as exc:
+        print(f"{c.RED}Error: could not locate config template: {exc}{c.RESET}")
+        return 1
+
+    with open(config_dest, "w", encoding="utf-8") as fh:
+        fh.write(config_content)
+
+    # --- review_prompt.md ---
+    prompt_dest = os.path.join(cwd, "review_prompt.md")
+    if os.path.exists(prompt_dest):
+        print(f"{c.YELLOW}review_prompt.md already exists in the current directory.{c.RESET}")
+        answer = input("Overwrite? [y/N] ").strip().lower()
+        if answer != "y":
+            print(f"{c.GREEN}✅ config.yaml created at:{c.RESET} {config_dest}")
+            print("   Skipped review_prompt.md (kept existing).")
+            return 0
+
+    try:
+        ref = importlib.resources.files("src.prompts").joinpath("review_prompt.md.template")
+        prompt_content = ref.read_text(encoding="utf-8")
     except (FileNotFoundError, TypeError) as exc:
-        print(f"{c.RED}Error: could not locate template file: {exc}{c.RESET}")
+        print(f"{c.RED}Error: could not locate review_prompt template: {exc}{c.RESET}")
         return 1
 
-    with open(dest, "w", encoding="utf-8") as fh:
-        fh.write(template_content)
+    with open(prompt_dest, "w", encoding="utf-8") as fh:
+        fh.write(prompt_content)
 
-    print(f"{c.GREEN}✅ config.yaml created at:{c.RESET} {dest}")
-    print(f"   Edit it to add your credentials and preferences.")
+    print(f"{c.GREEN}✅ config.yaml created at:{c.RESET} {config_dest}")
+    print(f"{c.GREEN}✅ review_prompt.md created at:{c.RESET} {prompt_dest}")
+    print("   Edit them to add your credentials, preferences and review rules.")
     return 0
 
 

code_review_ai_cli-1.0.1/src/prompts/review_prompt.md.template

@@ -0,0 +1,42 @@
+## Style Rules
+- Be direct and concise.
+- No greetings.
+- No emojis.
+- Include file and line whenever possible.
+
+## Project Context
+- Focus on practical findings that can be posted as PR comments.
+
+## Team Preferences
+- Prioritize actionable comments over long explanations.
+- Prefer concrete fix suggestions.
+- Avoid style-only comments unless they affect maintainability.
+
+## Good Comment Examples
+- "[src/service.py:128] Missing timeout in external HTTP call. Add timeout and handle exceptions to avoid blocking requests."
+- "[src/auth.py:57] Token is logged in plain text. Remove sensitive data from logs."
+
+## Bad Comment Examples
+- "Great work team!"
+- "As a senior reviewer, I think..."
+- "Maybe improve this part"
+
+## Universal Rules (all languages)
+
+### Security
+- No hardcoded secrets, credentials, or tokens
+- Validate and sanitize all external inputs
+- No SQL/command injection vectors
+- Least privilege principle applied
+
+### Code Quality
+- Single Responsibility Principle per function/class
+- No dead code, commented-out blocks, or TODOs without tickets
+- Meaningful names (no abbreviations unless domain-standard)
+- Max function length: \~30 lines; max file: \~300 lines
+- DRY: no duplicated logic
+
+### Error Handling
+- Errors must be caught, logged, and propagated correctly
+- No silent catches (`catch {}` or `catch (e) {}`)
+- Use typed/domain exceptions where applicable

{code_review_ai_cli-1.0.0 → code_review_ai_cli-1.0.1}/tests/test_ai_review.py

@@ -390,4 +390,165 @@ def test_main_prints_help_when_no_command_is_parsed(mocker) -> None:
     mocker.patch("src.ai_review.build_parser", return_value=parser)
 
     assert ai_review.main() == 0
-    parser.print_help.assert_called_once_with()
+    parser.print_help.assert_called_once_with()
+
+
+# ---------------------------------------------------------------------------
+# cmd_init
+# ---------------------------------------------------------------------------
+
+def test_cmd_init_creates_both_files(mocker, tmp_path) -> None:
+    """It should create config.yaml and review_prompt.md in the current directory."""
+    mocker.patch("src.ai_review.os.getcwd", return_value=str(tmp_path))
+    mocker.patch(
+        "src.ai_review.importlib.resources.files",
+        side_effect=_fake_pkg_resources,
+    )
+    mocker.patch("builtins.print")
+
+    result = ai_review.cmd_init()
+
+    assert result == 0
+    assert (tmp_path / "config.yaml").read_text() == "config-template"
+    assert (tmp_path / "review_prompt.md").read_text() == "prompt-template"
+
+
+def test_cmd_init_aborts_when_user_declines_config_overwrite(mocker, tmp_path) -> None:
+    """It should abort without writing either file when user declines config overwrite."""
+    (tmp_path / "config.yaml").write_text("existing")
+    mocker.patch("src.ai_review.os.getcwd", return_value=str(tmp_path))
+    mocker.patch("builtins.input", return_value="n")
+    mocker.patch("builtins.print")
+
+    result = ai_review.cmd_init()
+
+    assert result == 0
+    assert (tmp_path / "config.yaml").read_text() == "existing"
+    assert not (tmp_path / "review_prompt.md").exists()
+
+
+def test_cmd_init_skips_prompt_overwrite_when_user_declines(mocker, tmp_path) -> None:
+    """It should write config.yaml but keep existing review_prompt.md when user declines."""
+    (tmp_path / "review_prompt.md").write_text("existing-prompt")
+    mocker.patch("src.ai_review.os.getcwd", return_value=str(tmp_path))
+    mocker.patch(
+        "src.ai_review.importlib.resources.files",
+        side_effect=_fake_pkg_resources,
+    )
+    mocker.patch("builtins.input", return_value="n")
+    mocker.patch("builtins.print")
+
+    result = ai_review.cmd_init()
+
+    assert result == 0
+    assert (tmp_path / "config.yaml").read_text() == "config-template"
+    assert (tmp_path / "review_prompt.md").read_text() == "existing-prompt"
+
+
+def test_cmd_init_returns_error_when_template_missing(mocker, tmp_path) -> None:
+    """It should return 1 when the bundled config template cannot be found."""
+    mocker.patch("src.ai_review.os.getcwd", return_value=str(tmp_path))
+
+    def _raise(*args: object, **kwargs: object) -> None:
+        raise FileNotFoundError("missing")
+
+    mocker.patch("src.ai_review.importlib.resources.files", side_effect=_raise)
+    mocker.patch("builtins.print")
+
+    result = ai_review.cmd_init()
+
+    assert result == 1
+
+
+def test_cmd_init_returns_error_when_prompt_template_missing(mocker, tmp_path) -> None:
+    """It should return 1 when the review_prompt.md template cannot be found.
+
+    config.yaml is already written at that point; the function still surfaces the error.
+    """
+    mocker.patch("src.ai_review.os.getcwd", return_value=str(tmp_path))
+
+    class _ConfigOnlyResources(_FakeResource):
+        def __init__(self) -> None:
+            super().__init__("")
+
+        def joinpath(self, name: str) -> "_FakeResource":
+            if name == "config.yaml.template":
+                return _FakeResource("config-template")
+            raise FileNotFoundError(name)
+
+    mocker.patch(
+        "src.ai_review.importlib.resources.files",
+        return_value=_ConfigOnlyResources(),
+    )
+    mocker.patch("builtins.print")
+
+    result = ai_review.cmd_init()
+
+    assert result == 1
+    # config.yaml was already written before the error
+    assert (tmp_path / "config.yaml").read_text() == "config-template"
+    assert not (tmp_path / "review_prompt.md").exists()
+
+
+def test_cmd_init_overwrites_both_files_when_user_accepts(mocker, tmp_path) -> None:
+    """It should overwrite both existing files when the user confirms both prompts."""
+    (tmp_path / "config.yaml").write_text("old-config")
+    (tmp_path / "review_prompt.md").write_text("old-prompt")
+    mocker.patch("src.ai_review.os.getcwd", return_value=str(tmp_path))
+    mocker.patch(
+        "src.ai_review.importlib.resources.files",
+        side_effect=_fake_pkg_resources,
+    )
+    mocker.patch("builtins.input", return_value="y")
+    mocker.patch("builtins.print")
+
+    result = ai_review.cmd_init()
+
+    assert result == 0
+    assert (tmp_path / "config.yaml").read_text() == "config-template"
+    assert (tmp_path / "review_prompt.md").read_text() == "prompt-template"
+
+
+def test_main_dispatches_to_cmd_init(mocker) -> None:
+    """It should call cmd_init when the init subcommand is provided."""
+    mocker.patch("src.ai_review.sys.argv", ["ai_review.py", "init"])
+    cmd_init_mock = mocker.patch("src.ai_review.cmd_init", return_value=0)
+
+    result = ai_review.main()
+
+    assert result == 0
+    cmd_init_mock.assert_called_once_with()
+
+
+# ---------------------------------------------------------------------------
+# Helpers for cmd_init tests
+# ---------------------------------------------------------------------------
+
+class _FakeResource:
+    """Minimal stand-in for an importlib.resources path object."""
+
+    def __init__(self, text: str) -> None:
+        self._text = text
+
+    def joinpath(self, name: str) -> "_FakeResource":
+        return self
+
+    def read_text(self, encoding: str = "utf-8") -> str:
+        return self._text
+
+
+def _fake_pkg_resources(package: str) -> _FakeResource:
+    """Return a fake resource root whose joinpath distinguishes template names."""
+
+    class _Dispatcher(_FakeResource):
+        def __init__(self) -> None:
+            super().__init__("")
+
+        def joinpath(self, name: str) -> _FakeResource:
+            if name == "config.yaml.template":
+                return _FakeResource("config-template")
+            if name == "review_prompt.md.template":
+                return _FakeResource("prompt-template")
+            raise FileNotFoundError(name)
+
+    return _Dispatcher()