agent-feed vs AIWG
Repository-owned AI coding workflows compared
agent-feed and aiwg both provide reusable instructions and workflow assets for AI coding tools, while leaving the underlying AI providers in place. agent-feed centers on a shared repository protocol with task and verification gates; aiwg offers a broader catalog of specialist workflows and project context for work across sessions and domains.
agent-feed: Govern AI Coding Workflows in Your Repository
Agent Feed installs a repository-owned workflow for AI coding tools, with shared instructions, task and review gates, checks, and tool adapters. It suits teams that want consistent, verifiable AI-assisted changes across Codex, Claude Code, and Cursor.
AIWG: Reusable Workflows for AI Coding Tools
AIWG provides reusable project context, specialist agents, and workflows for tools including Claude Code, Codex, Copilot, and Cursor. Its TypeScript CLI deploys these assets across providers and supports persistent project artifacts, reviews, and optional decision support.
| agent-feed | AIWG | |
|---|---|---|
| Language | Python | TypeScript |
| License | MIT | MIT |
| Stars | 4 | 224 |
| Forks | 0 | 34 |
| Last analyzed | Oct 12, 2026 | Oct 20, 2026 |
Key differences
- agent-feed focuses on consistent, repository-defined coding practices, including scope, review, verification, and handoff; aiwg covers software development and other domains with specialist agents, skills, and templates.
- agent-feed uses `AGENTS.md` and a structured `.agents/` workflow with adapters for Codex, Claude Code, and Cursor; aiwg deploys assets to provider-readable locations and stores project artifacts in `.aiwg/`.
- agent-feed is written in Python and requires Python 3.11 or newer; aiwg is written in TypeScript and requires Node.js 20 or newer.
- agent-feed includes checks for workflow structure, references, scripts, skills, and adapters; aiwg includes optional utilities for indexing, diagnostics, memory, MCP integration, and decision support.
- agent-feed is quiet in the recent measured period and has one contributor; aiwg has regular commits and releases, with contribution activity concentrated overwhelmingly in one person.
- Both projects use the MIT license. agent-feed reports 4 stars and 0 forks, while aiwg reports 224 stars and 34 forks.
Choose agent-feed if you…
- want repository-owned instructions and checks to keep AI coding workflows consistent across supported tools.
- need explicit task boundaries, decision checkpoints, verification gates, and session handoff practices.
- prefer a focused workflow foundation that can incorporate team-specific constraints and skills.
Choose AIWG if you…
- need reusable project context and saved decisions to guide work across sessions.
- want specialist review workflows or domain frameworks beyond software development.
- need a TypeScript toolkit with optional utilities and bounded agent workflows, subject to provider and environment support.
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.