ai-cr 0.1.0__tar.gz → 0.4.3__tar.gz

ai_cr-0.4.3/PKG-INFO

@@ -0,0 +1,161 @@
+Metadata-Version: 2.3
+Name: ai-cr
+Version: 0.4.3
+Summary: LLM-agnostic GitHub AI Code Review Tool with integration to GitHub actions
+License: MIT
+Keywords: ai,coding,assistant,llm,code review,github,automation,devops,developer tools,github actions,workflows
+Author: Nayjest
+Author-email: mail@vitaliy.in
+Requires-Python: >=3.11,<4.0
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Software Development
+Requires-Dist: GitPython (>=3.1.44,<4.0.0)
+Requires-Dist: ai-microcore (>=4.0.0.dev11,<5.0.0)
+Requires-Dist: anthropic (>=0.49.0,<0.50.0)
+Requires-Dist: async-typer (>=0.1.8,<0.2.0)
+Requires-Dist: google-generativeai (>=0.8.5,<0.9.0)
+Requires-Dist: typer (>=0.9.4,<0.10.0)
+Requires-Dist: unidiff (>=0.7.5,<0.8.0)
+Project-URL: Homepage, https://github.com/Nayjest/github-ai-code-review
+Project-URL: Repository, https://github.com/Nayjest/github-ai-code-review
+Description-Content-Type: text/markdown
+
+<p align="right">
+<a href="https://pypi.org/project/ai-code-review/" target="_blank"><img src="https://badge.fury.io/py/ai-code-review.svg" alt="PYPI Release"></a>
+<a href="https://github.com/Nayjest/ai-code-review/actions/workflows/code-style.yml" target="_blank"><img src="https://github.com/Nayjest/ai-code-review/actions/workflows/code-style.yml/badge.svg" alt="Pylint"></a>
+<a href="https://github.com/Nayjest/ai-code-review/actions/workflows/tests.yml" target="_blank"><img src="https://github.com/Nayjest/ai-code-review/actions/workflows/tests.yml/badge.svg" alt="Tests"></a>
+<a href="https://github.com/Nayjest/ai-code-review/blob/main/LICENSE" target="_blank"><img src="https://img.shields.io/static/v1?label=license&message=MIT&color=d08aff" alt="License"></a>
+</p>
+
+# 🤖 AI Code Review Tool
+
+An AI-powered GitHub code review tool that uses LLMs to detect high-confidence, high-impact issues—such as security vulnerabilities, bugs, and maintainability concerns.
+
+## ✨ Features
+
+- Automatically reviews pull requests via GitHub Actions
+- Focuses on critical issues (e.g., bugs, security risks, design flaws)
+- Posts review results as a comment on your PR
+- Can be used locally; works with both local and remote Git repositories
+- Optional, fun AI-generated code awards 🏆
+- Easily configurable via [`.ai-code-review.toml`](https://github.com/Nayjest/ai-code-review/blob/main/ai_code_review/.ai-code-review.toml) in your repository root
+- Extremely fast, parallel LLM usage
+- Model-agnostic (OpenAI, Anthropic, Google, local PyTorch inference, etc.)
+
+See code review in action: [example](https://github.com/Nayjest/github-ai-code-review/pull/28#issuecomment-2891611236)
+
+## 🚀 Quickstart
+
+### 1. Review Pull Requests via GitHub Actions
+
+Create a `.github/workflows/ai-code-review.yml` file:
+
+```yaml
+name: AI Code Review
+on: { pull_request: { types: [opened, synchronize, reopened] } }
+jobs:
+  review:
+    runs-on: ubuntu-latest
+    permissions: { contents: read, pull-requests: write } # 'write' for leaving the summary comment
+    steps:
+    - uses: actions/checkout@v4
+      with: { fetch-depth: 0 }
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with: { python-version: "3.13" }
+    - name: Install AI Code Review tool
+      run: pip install ai-code-review==0.4.1
+    - name: Run AI code review
+      env:
+        LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+        LLM_API_TYPE: openai
+        MODEL: "gpt-4.1"
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      run: |
+        ai-code-review
+        ai-code-review github-comment --token ${{ secrets.GITHUB_TOKEN }}
+    - uses: actions/upload-artifact@v4
+      with:
+        name: ai-code-review-results
+        path: |
+          code-review-report.txt
+          code-review-report.json
+```
+
+> ⚠️ Make sure to add `LLM_API_KEY` to your repository’s GitHub secrets.
+
+💪 Done!  
+PRs to your repository will now receive AI code reviews automatically. ✨
+
+### 2. Run Locally
+
+Install and run:
+
+```bash
+# Prerequisites: Python 3.11+
+pip install ai-code-review
+
+# One-time setup using interactive wizard (saves configuration in ~/.env.ai-code-review)
+ai-code-review setup
+
+# Run review on committed changes in current branch vs main
+ai-code-review
+```
+
+To review a remote repository:
+
+```bash
+ai-code-review remote --url https://github.com/owner/repo --branch feature-branch
+```
+
+## 🔧 Configuration
+
+Change behavior via `.ai-code-review.toml`:
+
+- Prompt templates, filtering and post-processing using Python code snippets
+- Tagging, severity, and confidence settings
+- Custom AI awards for developer brilliance
+- Output customization
+
+You can override the default config by placing `.ai-code-review.toml` in your repo root.
+
+
+See default configuration [here](ai_code_review/.ai-code-review.toml).
+
+## 💻 Development Setup
+
+Install dependencies:
+
+```bash
+make install
+```
+
+Format code and check style:
+
+```bash
+make black
+make cs
+```
+
+Run tests:
+
+```bash
+pytest
+```
+
+## 🤝 Contributing
+
+We ❤️ contributions! See [CONTRIBUTING.md](CONTRIBUTING.md).
+
+## 📝 License
+
+Licensed under the [MIT License](LICENSE).
+
+© 2025 [Vitalii Stepanenko](mailto:mail@vitaliy.in)
+

ai_cr-0.4.3/README.md

@@ -0,0 +1,132 @@
+<p align="right">
+<a href="https://pypi.org/project/ai-code-review/" target="_blank"><img src="https://badge.fury.io/py/ai-code-review.svg" alt="PYPI Release"></a>
+<a href="https://github.com/Nayjest/ai-code-review/actions/workflows/code-style.yml" target="_blank"><img src="https://github.com/Nayjest/ai-code-review/actions/workflows/code-style.yml/badge.svg" alt="Pylint"></a>
+<a href="https://github.com/Nayjest/ai-code-review/actions/workflows/tests.yml" target="_blank"><img src="https://github.com/Nayjest/ai-code-review/actions/workflows/tests.yml/badge.svg" alt="Tests"></a>
+<a href="https://github.com/Nayjest/ai-code-review/blob/main/LICENSE" target="_blank"><img src="https://img.shields.io/static/v1?label=license&message=MIT&color=d08aff" alt="License"></a>
+</p>
+
+# 🤖 AI Code Review Tool
+
+An AI-powered GitHub code review tool that uses LLMs to detect high-confidence, high-impact issues—such as security vulnerabilities, bugs, and maintainability concerns.
+
+## ✨ Features
+
+- Automatically reviews pull requests via GitHub Actions
+- Focuses on critical issues (e.g., bugs, security risks, design flaws)
+- Posts review results as a comment on your PR
+- Can be used locally; works with both local and remote Git repositories
+- Optional, fun AI-generated code awards 🏆
+- Easily configurable via [`.ai-code-review.toml`](https://github.com/Nayjest/ai-code-review/blob/main/ai_code_review/.ai-code-review.toml) in your repository root
+- Extremely fast, parallel LLM usage
+- Model-agnostic (OpenAI, Anthropic, Google, local PyTorch inference, etc.)
+
+See code review in action: [example](https://github.com/Nayjest/github-ai-code-review/pull/28#issuecomment-2891611236)
+
+## 🚀 Quickstart
+
+### 1. Review Pull Requests via GitHub Actions
+
+Create a `.github/workflows/ai-code-review.yml` file:
+
+```yaml
+name: AI Code Review
+on: { pull_request: { types: [opened, synchronize, reopened] } }
+jobs:
+  review:
+    runs-on: ubuntu-latest
+    permissions: { contents: read, pull-requests: write } # 'write' for leaving the summary comment
+    steps:
+    - uses: actions/checkout@v4
+      with: { fetch-depth: 0 }
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with: { python-version: "3.13" }
+    - name: Install AI Code Review tool
+      run: pip install ai-code-review==0.4.1
+    - name: Run AI code review
+      env:
+        LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+        LLM_API_TYPE: openai
+        MODEL: "gpt-4.1"
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      run: |
+        ai-code-review
+        ai-code-review github-comment --token ${{ secrets.GITHUB_TOKEN }}
+    - uses: actions/upload-artifact@v4
+      with:
+        name: ai-code-review-results
+        path: |
+          code-review-report.txt
+          code-review-report.json
+```
+
+> ⚠️ Make sure to add `LLM_API_KEY` to your repository’s GitHub secrets.
+
+💪 Done!  
+PRs to your repository will now receive AI code reviews automatically. ✨
+
+### 2. Run Locally
+
+Install and run:
+
+```bash
+# Prerequisites: Python 3.11+
+pip install ai-code-review
+
+# One-time setup using interactive wizard (saves configuration in ~/.env.ai-code-review)
+ai-code-review setup
+
+# Run review on committed changes in current branch vs main
+ai-code-review
+```
+
+To review a remote repository:
+
+```bash
+ai-code-review remote --url https://github.com/owner/repo --branch feature-branch
+```
+
+## 🔧 Configuration
+
+Change behavior via `.ai-code-review.toml`:
+
+- Prompt templates, filtering and post-processing using Python code snippets
+- Tagging, severity, and confidence settings
+- Custom AI awards for developer brilliance
+- Output customization
+
+You can override the default config by placing `.ai-code-review.toml` in your repo root.
+
+
+See default configuration [here](ai_code_review/.ai-code-review.toml).
+
+## 💻 Development Setup
+
+Install dependencies:
+
+```bash
+make install
+```
+
+Format code and check style:
+
+```bash
+make black
+make cs
+```
+
+Run tests:
+
+```bash
+pytest
+```
+
+## 🤝 Contributing
+
+We ❤️ contributions! See [CONTRIBUTING.md](CONTRIBUTING.md).
+
+## 📝 License
+
+Licensed under the [MIT License](LICENSE).
+
+© 2025 [Vitalii Stepanenko](mailto:mail@vitaliy.in)

ai_cr-0.4.3/ai_code_review/.ai-code-review.toml

@@ -0,0 +1,305 @@
+report_template_md = """
+# 🤖 I've Reviewed the Code
+
+{% if report.summary -%}
+    {{ report.summary }}
+{%- endif %}
+
+**{%- if report.total_issues > 0 %}⚠️{% endif -%}
+Total issues: `{{ report.total_issues }}`
+{{- ' ' -}}
+in `{{ report.number_of_processed_files }}` files**
+
+{%- for issue in report.plain_issues -%}
+    {{"\n"}}## `#{{ issue.id}}`  {{ issue.title -}}
+    {{ "\n"}}[{{ issue.file }}{{' '}}
+
+    {%- if issue.affected_lines -%}
+        {%- for i in issue.affected_lines -%}
+        L{{ i.start_line }}{%- if i.end_line != i.start_line -%}-L{{ i.end_line }}{%- endif -%}
+        {%- if loop.last == false -%}, {%- endif -%}
+        {%- endfor -%}
+    {%- endif -%}
+    ]({{ issue.github_code_link(github_env) }})
+
+    {{"\n"}}{{ issue.details -}}
+    {{"\n"}}**Tags: {{ ', '.join(issue.tags) }}**
+    {%- for i in issue.affected_lines -%}
+        {%- if i.affected_code %}\n**Affected code:**\n```{{ i.syntax_hint }}\n{{ i.affected_code }}\n```{%- endif -%}
+        {%- if i.proposal %}\n**Proposed change:**\n```{{ i.syntax_hint }}\n{{ i.proposal }}\n```{%- endif -%}
+    {%- endfor -%}
+    {{ "\n" }}
+{%- endfor -%}
+
+"""
+retries = 3
+prompt = """
+{{ self_id }}
+----TASK----
+Review the provided code diff carefully and identify *only* highly confident issues which are relevant to any code context.
+
+----CODE----
+{{ input }}
+--------
+
+{% if file_lines -%}
+----ADDITIONAL CONTEXT----
+{{ file_lines }}
+{%- endif %}
+
+----TASK GUIDELINES----
+- Only report issues you are **100% confident** are relevant to any context.
+- Only include issues that are **significantly valuable** to the maintainers (e.g., bugs, security flaws, or clear maintainability concerns).
+- Do **not** report vague, theoretical, or overly generic advice.
+- Do **not** report anything with medium or lower confidence.
+{{ json_requirements }}
+
+Respond with a valid JSON array of issues in the following format:
+[
+    {
+        "title": "<issue_title>",
+        "details": "<issue_description>",
+        "tags": ["<issue_tag1>", "<issue_tag2>"],
+        "severity": <issue_severity>,
+        "confidence": <confidence_score>
+        "affected_lines": [ // optional; list of affected lines
+            {
+                "start_line": <start_line:int>,
+                "end_line": <end_line:int>,
+                "proposal": "<proposed code to replace the affected lines (optional)>"
+            },
+            ...
+        ]
+    },
+    ...
+]
+Available issue tags:
+- bug
+- security
+- performance
+- readability
+- maintainability
+- overcomplexity
+- language
+- architecture
+- compatibility
+- deprecation
+- anti-pattern
+- naming
+- code-style
+
+Issue severity scale:
+- 1 — Critical
+- 2 — Major
+- 3 — Minor
+- 4 — Trivial
+- 5 — Suggestion
+
+Confidence scale:
+- 1 — Highest, 100% confidence that code requires changes in any context
+- 2 — Very High
+- 3 — High
+- 4 — Medium - Should not be reported
+
+(!) - If no issues found according to the criterias, respond with empty list: []
+"""
+# Remove issues with confidence + severity > 3
+post_process = """
+for fn in issues:
+    issues[fn] = [
+        i for i in issues[fn]
+        if i["confidence"] == 1 and i["severity"] <= 2
+    ]
+"""
+summary_prompt = """
+{{ self_id }}
+Summarize the code review in one sentence.
+--Reviewed Changes--
+{% for part in diff %}{{ part }}\n{% endfor %}
+--Detected Issues--
+{{ issues | tojson(indent=2) }}
+---
+If code changs contains exceptional achievements, you may additionally present the award in summary text.
+--Available Awards--
+{{ awards }}
+---
+Your response will be parsed programmatically, so do not include any additional text.
+Use Markdown formatting in your response.
+"""
+
+[prompt_vars]
+self_id = """
+You are a subsystem of an AI-powered software platform, specifically tasked with performing expert code reviews.
+Act as a senior, highly experienced software engineer.
+"""
+json_requirements = """
+- ⚠️ IMPORTANT: RESPOND ONLY WITH VALID JSON, YOUR RESPONSE WILL BE PARSED PROGRAMMATICALLY.
+- Do not include any additional text or explanation outside the specified format.
+"""
+awards = """
+## 🧙‍♂️ "Refactoring Archmage"
+**For:** Elegantly transforming complex code into simple code without losing functionality.
+
+**Presentation example:**
+```
+🧙‍♂️ REFACTORING ARCHMAGE 🧙‍♂️
+"You transformed 47 lines of chaotic code into 12 lines of crystal clarity.
+Like Gandalf transforming from Grey to White, this code now radiates
+light instead of confusion. The coding magic school gives a standing ovation."
+```
+
+## 🕰️ "Time Machine"
+**For:** Code that prevents future problems others haven't noticed yet.
+
+**Presentation example:**
+```
+🕰️ TIME MACHINE 🕰️
+"Your edge case handler just saved the company from a dark
+alternative timeline where at 3:00 AM next month
+the DevOps team goes crazy from incomprehensible errors. History has changed,
+the future is no longer what it was."
+```
+
+## 🎭 "Shakespearean Playwright"
+**For:** Exceptionally expressive variable and function names that tell a story.
+
+**Presentation example:**
+```
+🎭 SHAKESPEAREAN PLAYWRIGHT 🎭
+"'processUserInputAndValidateBeforeSending' — a whole act of drama in one
+function name! Such clarity of intent, such drama! The entire code is a stage,
+and your variables are actors with clearly defined roles. The audience is thrilled."
+```
+
+## 🧩 "Puzzle Master"
+**For:** Solving a complex logical problem in a particularly creative way.
+
+**Presentation example:**
+```
+🧩 PUZZLE MASTER 🧩
+"Where others saw impassable thickets of conditions, you paved an elegant algorithmic
+path. Your solution looks so natural that now it seems like there could never have been
+another way. Rubik applauds."
+```
+
+## 🐛 "Ghostbuster"
+**For:** Detecting and fixing elusive bugs or potential issues.
+
+**Presentation example:**
+```
+🐛 GHOSTBUSTER 🐛
+"This elusive bug was hiding in the shadows for five sprints, feeding on developers'
+souls and sowing chaos. 'Who are you?' it screamed when you dragged it into the light
+with your precise fix. Paranormal activity eliminated."
+```
+
+## 🏛️ "Architectural Virtuoso"
+**For:** Code structuring that promotes extensibility and flexibility.
+
+**Presentation example:**
+```
+🏛️ ARCHITECTURAL VIRTUOSO 🏛️
+"Your architecture is like the Parthenon of modern code: proportional, harmonious, and seems
+to withstand the pressure of time and changing requirements. Vitruvius records your patterns
+for future generations."
+```
+
+## 🧬 "Code Geneticist"
+**For:** Successful use of inheritance/composition or other complex OOP concepts.
+
+**Presentation example:**
+```
+🧬 CODE GENETICIST 🧬
+"Your elegant inheritance chain has mutated the code into a new life form — more
+adaptive, more evolutionarily stable. Natural selection kindly approved these changes,
+while unacceptable complexity remains in the paleontological past of development."
+```
+
+## 🔄 "Zen of Loops"
+**For:** Writing particularly efficient and understandable loops/iterations.
+
+**Presentation example:**
+```
+🔄 ZEN OF LOOPS 🔄
+"Your loop impresses with its laconic wisdom. Nothing extra, nothing forgotten,
+perfect balance between readability and performance. 'Before writing a loop,
+think whether it's needed at all,' says the master. Your loop — is needed."
+```
+
+## 🛡️ "Gate Guardian"
+**For:** Excellent input validation and protection against edge cases.
+
+**Presentation example:**
+```
+🛡️ GATE GUARDIAN 🛡️
+"No bad data shall pass your vigilant defense. Users may enter
+the most bizarre combinations, but your code stands firm, like a sentinel at the gates
+of the data city. 'You shall not pass!' it speaks to invalid format."
+```
+
+## 🎨 "Readability Impressionist"
+**For:** Code that reads like well-written prose.
+
+**Presentation example:**
+```
+🎨 READABILITY IMPRESSIONIST 🎨
+"Reading your code, you feel sunlight falling on the water lilies of clarity,
+like a breeze playing in the willows of logic. Each line is a brush stroke,
+and together they create a picture that can be understood at first glance."
+```
+
+## 🚀 "Optimization Pioneer"
+**For:** Significant performance improvement without sacrificing readability.
+
+**Presentation example:**
+```
+🚀 OPTIMIZATION PIONEER 🚀
+"Oh! Your algorithm now flies at the speed of light! If it used to crawl
+like a snail through O(n²) sand, now it races down the O(log n) highway.
+The passengers of this code won't even notice how they arrive at their destination!"
+```
+
+## 📚 "Code Chronicler"
+**For:** Exceptionally useful and informative comments.
+
+**Presentation example:**
+```
+📚 CODE CHRONICLER 📚
+"Your comments are like an ancient manuscript revealing the secrets of forgotten civilizations.
+'And there was light,' you said, and indeed the light bulb of understanding lit up above the heads
+of all who will read this code in the future."
+```
+
+## 🧪 "Testing Alchemist"
+**For:** Writing particularly creative and thorough tests.
+
+**Presentation example:**
+```
+🧪 TESTING ALCHEMIST 🧪
+"In your testing crucible you mixed reagents of edge cases, catalyst
+of boundary conditions and essence of unit tests. The philosopher's stone of quality was born —
+your code is now immortal in the face of regression!"
+```
+
+## 🗿 "Ancient Artifact Decoder"
+**For:** Successfully working with complex legacy code.
+
+**Presentation example:**
+```
+🗿 ANCIENT ARTIFACT DECODER 🗿
+"You stand among the ruins of code written five years ago in the forgotten language of the ancients.
+Like Champollion with the Rosetta Stone, you deciphered the hieroglyphs of functions,
+restored lost knowledge and now bestow it upon a new generation."
+```
+
+## 🎵 "Pattern Composer"
+**For:** Sophisticated application or combination of multiple design patterns.
+
+**Presentation example:**
+```
+🎵 PATTERN COMPOSER 🎵
+"Your symphony of patterns sounds magnificent! Factory method opens the first movement,
+decorators add depth and texture, and observer masterfully completes the composition.
+The Gang of Four gives a standing ovation from the stalls."
+```
+"""

ai_cr-0.4.3/ai_code_review/bootstrap.py

@@ -0,0 +1,37 @@
+import logging
+from datetime import datetime
+
+import microcore as mc
+
+from .constants import ENV_CONFIG_FILE
+
+
+def setup_logging():
+    class CustomFormatter(logging.Formatter):
+        def format(self, record):
+            dt = datetime.fromtimestamp(record.created).strftime("%Y-%m-%d %H:%M:%S")
+            message, level_name = record.getMessage(), record.levelname
+            if record.levelno == logging.WARNING:
+                message = mc.ui.yellow(message)
+                level_name = mc.ui.yellow(level_name)
+            if record.levelno >= logging.ERROR:
+                message = mc.ui.red(message)
+                level_name = mc.ui.red(level_name)
+            return f"{dt} {level_name}: {message}"
+
+    handler = logging.StreamHandler()
+    handler.setFormatter(CustomFormatter())
+    logging.basicConfig(level=logging.INFO, handlers=[handler])
+
+
+def bootstrap():
+    """Bootstrap the application with the environment configuration."""
+    setup_logging()
+    logging.info("Bootstrapping...")
+    mc.configure(
+        DOT_ENV_FILE=ENV_CONFIG_FILE,
+        VALIDATE_CONFIG=False,
+        USE_LOGGING=True,
+        EMBEDDING_DB_TYPE=mc.EmbeddingDbType.NONE,
+    )
+    mc.logging.LoggingConfig.STRIP_REQUEST_LINES = [100, 15]