Open-source Swamp compiles repeated agent skills into workflows: 8x fewer tokens, 2x faster
udmrzn · x · 2026-11-16
Developers open-sourced Swamp (745 GitHub stars) to fix a common waste: coding agents re-read skills and re-decide every step on every run. With Swamp, you describe a task once and the agent breaks it into reusable behaviors with typed inputs/outputs, wired together by plain code.
Key features:
- Model calls only on steps that need judgment
- Secrets pulled from vaults and redacted from outputs
- Versioned, queryable data from every run plus a full execution record
- Plugs into coding agents you already use (e.g., Claude)
Paired with the referenced article "Claude Subagents vs. Agent Teams," it's a useful reference for agent orchestration and token-cost engineering.
More from coding & agent
- A zero-AI chatbot: 1,000 DeepSeek agents wrote 300,000 if-else rules — victormustar · 2026-11-16
- 67k-star diagram-design skill makes AI coding tools draw editorial-grade diagrams — adnan_hashmi · 2026-11-16
- Dev's take on AI workflows: feed the right data and build verification loops — iamsahaj_xyz · 2026-11-16
- Every tool is becoming an API for AI agents: a field guide to the agent stack — DEVIL_QI · 2026-11-16
- Alibaba open-sources OpenCodeReview, its internal AI code reviewer using ~1/9 the tokens of Claude Code — udmrzn · 2026-11-16
- Open-source CalDAV MCP Server lets AI assistants manage your iCloud Calendar — Weekly_Mountain2783 · 2026-11-16