Infrastructure●●Solid
Kelos – Define your AI coding agent workflow as YAML on Kubernetes
Kubernetes-native AI agent orchestration via YAML; early but self-hosting Kelos development proves the loop.
WizardryBig BrainNiche Gem
gjkim042
454mo ago
Back to browse
GitHub Repository
Build contained, event-driven AI agents. Define your agent in a single config file, run it in a container and deploy it anywhere.
4 starsTypeScript
I built an agent framework where the agent is just one file
by RatulMaharaj·Jul 13, 2026·1 point·1 comment
AI Analysis
●●●BangerBig BrainCozy
Single-file YAML agent definitions with SQLite audit trails beat complex orchestration frameworks.
Strengths
- •Defines entire agent logic, triggers, and permissions in a single YAML config file.
- •Deny-by-default security model with SQLite audit trails for every permission decision.
- •Docker-native deployment with hardened base images and built-in budget caps.
Weaknesses
- •Alpha software with changing interfaces and potential breaking changes.
- •Limited to specific triggers like Discord, Slack, Telegram, and webhooks.
Category
Target Audience
Backend engineers deploying long-running AI agents
Similar To
LangGraph · AutoGen · CrewAI
Similar Projects
Developer Tools●●Solid
Imara – policy enforcement layer for MCP agents (npx imara)
Transport-layer policy enforcement beats forking your agent code.
Solve My ProblemBig Brain
dnakitare
2018d ago
Developer Tools●●Solid
Agent-pulse – local gateway that fans out AI agent events to clients
Replaces scattered per-project agent hooks with one global config and SSE fan-out.
Solve My ProblemNiche Gem
SantiagoBobrik
104mo ago
AI/ML●Mid
Local-first multi-agent harness
Another local agent framework competing with established orchestration tools.
Ship It
undersky
2219d ago
Security●●Solid
Drydock – VM Sandboxes for macOS Autonomous Coding Agents
Host-side API key gateway keeps credentials out of agent sandboxes entirely.
Big BrainNiche Gem
sricola
1024d ago
Security●●●Banger
GatewayStack – Deny-by-default security for OpenClaw tool calls
OpenClaw agents could read your SSH keys; this blocks it at the process level, not advisory skills.
Solve My ProblemSlick
davidcrowe
104mo ago