goon vs DeclarAgent
Go-based AI workflow projects compared
goon is a self-hosted daemon and CLI for running AI workflows, including software tasks and scheduled automations. DeclarAgent runs reusable YAML runbooks through a CLI or MCP server, with validation and controls for side effects.
goon: Run Self-Hosted AI Workflows as a Daemon
goon is a Go-based, self-hosted daemon for running AI workflows, from software tasks and pull requests to scheduled reports and monitoring. It supports human approval through a web UI, terminal, or Telegram, with configurable workflow stages and persistent memory.
DeclarAgent: Run Auditable Workflows for AI Agents
DeclarAgent runs YAML-defined workflows through shell commands, built-in actions, and HTTP requests. Its validation, dry-run, approval gates, and MCP integration suit teams that want agents to invoke controlled CLI workflows.
| goon | DeclarAgent | |
|---|---|---|
| Language | Go | Go |
| License | MIT | MIT |
| Stars | 3 | 10 |
| Forks | 0 | 0 |
| Last analyzed | Oct 17, 2026 | Oct 14, 2026 |
Key differences
- goon focuses on continuous task execution, issue tracking, project memory, and approval across web, Telegram, and terminal interfaces; DeclarAgent focuses on defining and invoking explicit YAML plans.
- goon includes a software workflow for coding and preparing pull requests, while DeclarAgent provides reusable steps for shell commands, file and JSON actions, and HTTP requests.
- goon connects to Jira or GitHub Issues and supports several code-hosting services for pull requests; DeclarAgent exposes plans as MCP tools or can generate a Claude Code Skill.
- Both are written in Go and use the MIT license. goon is listed with 3 stars and DeclarAgent with 10; both have 0 forks.
- DeclarAgent documents macOS testing, with Linux expected and Windows requiring WSL or a similar POSIX-compatible shell. goon is described as a self-hosted daemon with locally stored project state and memory.
Choose goon if you…
- need a continuously running worker for software tasks or scheduled workflows.
- want issue integrations, local project memory, and approval through a web UI, Telegram, or terminal.
- prefer per-role model routing across listed providers, including Ollama.
Choose DeclarAgent if you…
- want to define reusable, version-controllable workflows as YAML plans.
- need validation, previews, structured results, or explicit approval for destructive steps.
- want plans callable through MCP or a generated Claude Code Skill.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.