Skip to main content

A local multi-agent code reviewer

Project description

Ollama Reviewer PIP LINK HERE

A local multi-agent code review tool that uses Ollama-hosted LLMs to analyze git diffs and produce structured PR reviews.

How It Works

The tool runs several specialised review agents against your staged/unstaged code changes (git diff HEAD):

  1. Code Smells & Metrics — analyses code quality, complexity, and anti-patterns
  2. Readability Review — evaluates naming, documentation, and overall clarity
  3. Security Review — checks for common vulnerabilities and insecure patterns
  4. Synthesizer — stitches the individual agent reports into a single markdown PR comment

Each agent returns a JSON score out of 100. If any agent detects a fatal block (security flaw or architecture issue), the commit is blocked with exit code 1. But users can commit whether or not they care about the pr check. After all agents is ran an md file consisting of the reviews will be available in whatever directory the command was ran.

Requirements

  • Ollama running locally with at least one model pulled
  • Python 3.10+

Setup

# Install the package
pip install -e . # for any local installs from git

# Pull a supported model or check available models at end of page (default: gemma3:12b)
ollama pull gemma3:12b

Usage

Run from the root of any git repository you want to review:

ollama-review -m AVAILABLE_MODEL -p MODEL_PARAMS

Specify a different model:

ollama-review -m "qwen3:4b"

Model parameters (temperature, top_p) are configured in providers.yml.

Output

  • Prints each agent's score as it runs
  • Writes a final synthesised markdown review to markdown_output.md to the directory in which the review command was executed
  • Exits with code 0 on approval, code 1 if a fatal issue is found

Available models in providers.yml, feel free to add your own by looking through the package

  • "qwen3:1.7b": { "temperature": 0.0, "top_p": 0.9 },
  • "gemma3:1b": { "temperature": 0.0, "top_p": 0.9 },
  • "qwen3:4b": { "temperature": 0.1, "top_p": 0.4 },
  • "gemma3:4b": { "temperature": 0.1, "top_p": 0.9 },
  • "gemma3:12b": { "temperature": 0.1, "top_p": 0.9 },
  • "mistral:7b": { "temperature": 0.1, "top_p": 0.9 }

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ollama_local_reviewer-1.0.7.tar.gz (11.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

ollama_local_reviewer-1.0.7-py3-none-any.whl (14.4 kB view details)

Uploaded Python 3

File details

Details for the file ollama_local_reviewer-1.0.7.tar.gz.

File metadata

  • Download URL: ollama_local_reviewer-1.0.7.tar.gz
  • Upload date:
  • Size: 11.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.15

File hashes

Hashes for ollama_local_reviewer-1.0.7.tar.gz
Algorithm Hash digest
SHA256 255d27ce18cd7fa1d0de0222df40401d27cd32ebd80e35e976fe9b6486ff9cea
MD5 e0840747e8f4d49fb000424a7d5ebd5e
BLAKE2b-256 cdebca5a8b59e0c5aef99b926cc6e83c880c544ef25399b99a56d7bcd72182d5

See more details on using hashes here.

File details

Details for the file ollama_local_reviewer-1.0.7-py3-none-any.whl.

File metadata

File hashes

Hashes for ollama_local_reviewer-1.0.7-py3-none-any.whl
Algorithm Hash digest
SHA256 bdac4764fcf8a21b2b471a9f01bcb840392249894223c9895678d136044e3e04
MD5 aba21fcf298f04e50551b4f2762b5e92
BLAKE2b-256 43730804d8bba86c77a933c8b1138ffc84e964b6606856307e1dbf7ed525435a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page