Python projects — indie software
-
comandos
Mission control for running many Claude Code agents in parallel — live dashboard, actionable popups with full markdown, voice alerts, tabbed terminals, one-key answers, Telegram remote control. Linux, no cloud, survives reboots.
-
kimi-quota-statusline
Kimi Code CLI 底部状态栏增强:官方额度条(5h/7d)+ token 消耗金额 + swarm 特效 | Quota & usage statusline for Kimi Code CLI
-
astro-tasks
Astro-based task management and productivity tool
-
hypertaks-agent
Hypertaks = "A founder who owns a library of more than 1,400 books, knows exactly which book to consult for this issue, and has a strict SOP: intake → formulation → implementation → integration → documentation."
-
laya
Laya is an open-source, local-first AI notification command center that aggregates Slack, Gmail, GitHub, Jira, Notion, Outlook, Calendar (and more) notifications using local LLMs via Ollama and LM Studio. Supports cloud models via BYOK.
-
bitbox
A community-built collection of tiny Python CLI tools. One file. One function. One contributor.
-
session-recall
Shared memory for Claude Code and Codex. Your agent recalls decisions from weeks ago — including what the other engine worked out. Local semantic search over both histories, as MCP tools.
-
agent-engineering-toolkit
Proof-carrying workflows for coding agents: bind tests to exact code, detect stale evidence, review scope, and hand proof across agents.
-
obsidian-agent
AI agent that lives in your Obsidian vault. Manages tasks, knowledge, and finances through natural language.
-
agentforge-runtime
AgentForge Protocol + Cloud: create, run, monetize, evaluate, and operate AI agents across APIs, marketplaces, workflows, channels, and enterprise operations.
-
codemium
Persistent coding intelligence for Codex — project-aware context, adaptive reasoning, scoped changes, and risk-aware verification.
-
godot-agent
Godot AI agent CLI, Skill, and MCP server with structured JSON/schema output — headless scene/script/export automation plus live runtime control.
-
aileron
Tamper-evident flight recorder for AI agents - hash-chained, signed audit log of every tool call, with MCP proxy policy enforcement and anomaly detection.
-
vegaloom
One writes, one reviews — worker 与 reviewer 上下文隔离的 AI 编码工作流 Harness，以项目验证命令、风险门禁和运行证据判断任务能否完成。
-
outpost
A small personal kit that installs prompts for coding agents.
-
wgpl
The DevOps-first WireGuard control plane. Automate zero-downtime deployments and manage peers across multiple servers effortlessly.
-
roadkeep
Keeps a project's roadmap, changelog, improvements and strategy files to one format - a schema at the point of insertion, not a linter after the fact. Python CLI + Claude Code plugin.
-
renforge-mcp
Free, open-source MCP server for Ren'Py visual-novel development — give AI agents full control of a Ren'Py project (analyze, write, run, see, test).
-
grounded-reasoning
Zero-token, precision-guaranteed relation-algebra verifier for LLM/agent multi-hop reasoning
-
AlfredOS
An open-source, self-hostable, multi-user, multi-persona, security-hardened agentic OS.
-
Velonus
AI-native security copilot for Python developers. Scans for secrets, vulnerabilities, and dependency CVEs — then tells you how to fix them.
-
claude-code-from-scratch
Build Claude Code from scratch with a step-by-step tutorial in TypeScript and Python for learning agent coding tools
-
AladdinAI
AladdinAI: open-source BYOI (Bring Your Own Infrastructure) AI Agent platform. Connect your own VMs, NVIDIA NIM, OpenAI, Anthropic, Ollama — build and manage AI agents visually.
-
mymeal
Self-hosted recipes, AI meal planning, pantry & smart shopping lists — Home Assistant native (add-on + HACS integration, MCP, Assist voice)
-
unbluff
Self-verification hooks + a reasoning skill for Claude Code - keep the agent honest across your prompts, its claims, your tests, parked work, and memory. Fail-silent, stdlib-only, no network.
-
agentweb
Turn websites into fast CLI commands that any coding agent can use
-
gpt_export_distillation
Distill ChatGPT exports into structured Markdown archives for cleanup, review, and reuse
-
edgeverdict
A code review gate that verifies changes by executing tests, not by judging diffs. LLM proposes; a deterministic gate decides.
-
olympus
Self-hosted, partitioned agent-memory fleet: bare-metal Debian setup + mem0/Qdrant/Ollama stack + Hermes/Claude Code/OpenCode/Pi integrations, run from one olympus CLI/TUI.
-
dotagents
My main list of vendor-agnostic agent skills and harness configs
-
runbookproof
Continuously verify commands in documentation, runbooks, and AI-generated instructions.
-
odoo-forge
Modular Odoo platform: layered projects, declarative manifests, pluggable execution backends
-
aegis
The meta-harness that brings real agentic collaboration as a first-class citizen
-
argus-agent
A harness that makes a small (or modest) LLM reliably run multi-step agentic tasks — tools, skills, memory, scheduling.
-
wetgeving-nl
Community-driven repository van alle Nederlandse wetgeving in schone Markdown. Elke wet = 1 bestand, elke wijziging = 1 nette commit. Met proposals-sectie zodat iedereen echte wetsvoorstellen kan indienen via pull requests.
-
CountryWeather
Resilient, data-driven API automation framework. Architected for multi-env testing, network stability, performance SLAs, and AI-governed maintainability and pytest and allure reporting
-
avatar-mix
Generate professional videos from scripts or URLs using HeyGen avatars, dynamic layouts, automated captions, and custom audio effects.
-
ai-badger
Source-of-truth marketplace + project scaffolder for custom Claude/agent skills, personas, and instructions. Organized as {stack|common}/{feature}; welcome-ai-badger scaffolds a repo, feed-badger contributes back.
-
numclassify
python module for all number programs
-
revops-skills
Automate Salesforce RevOps tasks with Claude agent skills designed for pipeline hygiene, forecast reconciliation, and deal pressure testing.
-
chatbot-saas
RAG-powered chatbot-as-a-service — embed an AI chatbot trained on your own docs into any website
-
partyline
One wire, every agent — a shared telephone line for Claude Code agents. (It's a JSONL file.)
-
exomem
Self-hosted MCP server that makes your Obsidian/markdown vault searchable — text, PDFs, Office docs, images, audio — from any MCP client. Hybrid retrieval over a typed, governed corpus, sub-second at 50k notes; your files stay plain markdown.
-
Stepik-Python-Grader
Локальный грейдер для курсов «Поколение Python» на Stepik. Скачивает тесты к задаче с сайта и позволяет не только проверить решение локально, но и сравнить несколько решений более честно: сначала по корректности, потом по benchmark-метрикам.
-
work-in-germany
Automated job board for Germany — tech, Werkstudent, Ausbildung, and entry-level roles, updated every 6 hours
-
PromptQueue
Dependency-free prompt scheduler for AI rate-limit resets. Queue prompts for Claude, Codex, ChatGPT, Gemini, Cursor, Antigravity, and more.
-
attestral
Open-source security scanner for MCP servers and AI agents: finds prompt injection, tool poisoning, and excessive agency, and models the cloud your agents can reach. pip install attestral.
-
AI-Employee-Silver
Automate Gmail and LinkedIn tasks with AI drafts, human approval, and scheduled posting
-
codex-coding-os
Spec-first workflow pack for OpenAI Codex: skills, AGENTS.md templates, workflow gates, fresh-context review, validation, and handoff.
-
AynOps
An Open Sourced Model Context Protocol (MCP) Local server that gives AI Clients real-time cybersecurity reconnaissance capabilities
-
icode-skill
A 6-step coding workflow for Claude Code — Plan → Review → Finalize → Code → Deep Check → Audit. Run all at once or step-by-step, switch models between steps. Claude Code 六步编码工作流：计划 → 审查 → 定稿 → 编码 → 复检 → 终审，一步到位或分步执行，步骤间可切换模型。
-
epicurus
Self-hosted, modular, local-first personal-assistant platform — an AI agent plus pluggable integration modules.
-
invinoveritas
The verification layer for autonomous agents — an independent, capital-aware verdict before an irreversible action (/review), a cryptographically signed proof after, and a public on-chain track record, losses included. Free EdgeProof backtest validation. Underneath: memory, sandboxed execution, marketplace. Pay in sats, USDC (x402), or card.
-
DevRadar
Discover Hackathons, AI Tools, and Developer Opportunities
-
akashic-aurora
Memory for AI agents that learns what actually helped: lessons injected at the moment of action, credited by outcome, challenged by counter-evidence. Local-first, multi-agent, test-gated.
-
VecMem
Manage LLM agent memory with a resource-aware dual-tier system that uses HNSW indexing to balance short-term context and long-term knowledge retention.
-
QuantPilot
A local market terminal: Finviz-class screening over 7,000 US stocks, live finance news channels, and a Bloomberg-style keyboard UI. No API key, no account, no monthly bill.
-
ai-jury
Stdlib-only Python CLI that runs native coding-agent CLIs (Claude Code, Codex, Antigravity) + a free local model as one review jury — review a diff/PR/issue, debate, verify, synthesize one verdict (chair or panel vote). Local-first, secure by default.
-
agent-skills
Agent Skills — skill collection for AI agents (Sisyphus, opencode): github-repo-hygiene, test-graphics, reddit-karma, presentation-maker, docs-system. Autonomous instructions with skill.json manifests, CI validation, full documentation.
-
rss-telegram-bot
Automated Crypto RSS Feed Telegram Bot powered by Python & GitHub Actions (100% Free Hosting)