Intelligent documentation agent for AI coding assistants
Project description
Sensei
Intelligent documentation agent for AI coding assistants
Sensei is a hyper-focused model specialized in providing correct, accurate, actionable, thorough, cross-validated guidance with working examples. Sensei handles all the synthesis, so you stay focused on your task with minimal context pollution.
AI assistants often hallucinate documentation or rely on outdated training data. Sensei fixes that by searching multiple authoritative sources and synthesizing accurate, up-to-date answers with source attribution.
Install
claude plugins:install @alizain/sensei
That's it. No API keys, no configuration.
How It Works
Sensei searches multiple sources and synthesizes the best answer:
- Context7 - Pre-indexed official library documentation
- Scout - GitHub repository exploration (code search, file structure, repo maps)
- Tavily - AI-focused web search for docs, blogs, and discussions
- Kura - Cached previous answers for instant responses
Every response includes source attribution and confidence levels. Rate responses with the feedback tool to improve future results.
Usage
Just ask your AI assistant to use Sensei when you need documentation:
"Use Sensei to find how to set up authentication in FastAPI with OAuth2"
Sensei works best for library documentation, API references, framework guides, and "how do I do X with Y" questions.
Prerequisites
PostgreSQL 17+
Sensei requires PostgreSQL to be installed on your system. Install it using your package manager:
macOS:
brew install postgresql@17
Ubuntu/Debian:
sudo apt install postgresql-17
Windows: Download from https://www.postgresql.org/download/windows/
Note: You don't need to configure PostgreSQL or create databases manually. Sensei automatically manages a local PostgreSQL instance in ~/.sensei/pgdata/ using a Unix socket (no port conflicts with system PostgreSQL).
Development Setup
Set SENSEI_HOME=.sensei in your .env file to keep development data local to the repo instead of ~/.sensei/.
License
MIT
Built with PydanticAI and FastMCP
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file sensei_ai-1.3.0.tar.gz.
File metadata
- Download URL: sensei_ai-1.3.0.tar.gz
- Upload date:
- Size: 79.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.9.11 {"installer":{"name":"uv","version":"0.9.11"},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
6467b85a53eab73ff31910fe5b20aeb11ca13e7ccc197a40a75bb3d826186d1a
|
|
| MD5 |
5357239ebb0d959e37cfa8fdf6fe6448
|
|
| BLAKE2b-256 |
fdccb8d55d4ca1bfe38ae991fa7ce1e1ca82a50cb97bd9b91e9403cf0150856e
|
File details
Details for the file sensei_ai-1.3.0-py3-none-any.whl.
File metadata
- Download URL: sensei_ai-1.3.0-py3-none-any.whl
- Upload date:
- Size: 75.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.9.11 {"installer":{"name":"uv","version":"0.9.11"},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
5f083e0a6e1b0f0a204510a9bb914d1f9a73ccd9d5bff6d1d306cac22bea782d
|
|
| MD5 |
e7d60db27f484878ae0662ff5bfddc8e
|
|
| BLAKE2b-256 |
8992640e8a52f1f5e6976ba007e24abb98ffe41238437f2d9308f7a6fc963dad
|
