Dulus
An AI agent for your desktop and terminal, on any model you choose. Hosted models run on Fuel, paid in $DULUS.
翻訳待ち
リポジトリ情報はまだ収集していません。
No updates detected yet. GitHub releases will appear here.
Curated catalog of 344 GitHub repos for building software with AI agents, in 24 lists: coding agents, skills, MCP, orchestration, memory. Searchable, filterable GitHub Pages site plus CSV, TOML and JSON exports. Stdlib Python build, vanilla JS, no dependencies.
翻訳待ち
Loading the full README.
Read the full README ↗344 public GitHub repositories I have starred, sorted into 24 lists. 222 of them are tools for building software with AI agents: frameworks, coding harnesses, orchestration, memory, skills and evals. The rest are the terminal tools, infrastructure and references around them. Snapshot taken 2026-10-29.
Browse, search and filter the list on the site, or read it below. The same data is published as CSV, TOML and JSON for other tools to import.
Activity is set when the snapshot is taken, from the date of the last push.
| Label | Last push |
|---|---|
| 🟠 Very active | Within 31 days of the snapshot |
| 🔵 Active | Within 92 days |
| 🟡 Slowing | Within a year |
| 🔴 Dormant | More than a year before |
| ⚫ Archived | Any date; the repository is archived |
| File | Use it for |
|---|---|
starred_repos.csv | The source. Spreadsheets and anything that reads CSV. |
starred_repos.toml | Config-driven tools. One [[repos]] table per repository. |
starred_repos.json | Scripts and the site. Same keys as the TOML. |
Each file is also served from the site, for example https://thinkjones.github.io/agentic-tools/starred_repos.toml.
[[repos]]
list = "AI · Agent Frameworks & SDKs"
repo = "langchain-ai/langchain"
url = "https://github.com/langchain-ai/langchain"
stars = 147561
forks = 24753
activity = "Very active"
last_push = 2026-10-10
last_release = "langchain-core==1.6.9"
release_date = 2026-10-08
license = "MIT"
language = "Python"
homepage = "https://docs.langchain.com/langchain/"
topics = ["agents", "ai", "ai-agents"]
description = "The agent engineering platform."
starred_on = 2026-03-24
import tomllib
with open("starred_repos.toml", "rb") as f:
repos = tomllib.load(f)["repos"]
active_rust = [r["repo"] for r in repos if r.get("language") == "Rust" and r["activity"] == "Very active"]
| Key | CSV column | Type | Notes |
|---|---|---|---|
list | Proposed List | string | |
repo | Repo | string | owner/name |
url | URL | string | |
stars, forks | Stars, Forks | integer | |
activity | Activity | string | One of the five labels above |
last_push, release_date, starred_on | Last Push, Release Date, Starred On | date | TOML local date; YYYY-MM-DD string in JSON |
last_release | Last Release | string | Release tag |
license | License | string | SPDX identifier; NOASSERTION means GitHub found a license it could not classify |
language | Language | string | Primary language |
homepage | Homepage | string | |
topics | Topics | array of strings | |
description | Description | string |
This is the beginning of the README. Read the rest at the original. Read the full README ↗