Agent version control: a four-layer model for sharing agents across teams
yangyi · x · 2026-10-21
Yangyi shares engineering notes from building NewMax Agent-im: when teams share agents, prompts, Skills, memory, secrets, and permissions tangle together and fork/merge quickly breaks down — an infra-level opportunity for an "Agent GitHub."
- Four-layer split: definitions (system prompt/soul.md, Git-style diffing), dependencies (Skills/CLIs/connectors, package-manager style link + lockfile), state (shared/specialized/session memory, semantic merging via data versioning), environment (secrets/permissions, never forked, declared and granted per deployment)
- Layered overlay like Docker images: base layers linked read-only, upper layers forked, resolving sync and merge in multi-team collaboration
- Memory merging via atomic memories, promotion pipelines, and eval sets rather than text diff
- Sketches a seven-layer enterprise "Agent Resource" infrastructure, arguing it should be an open protocol combining Docker Registry, branchable databases, enterprise identity, and CI.
More from coding & agent
- AWS Transform's .NET agent now converts Azure SDK code to AWS equivalents — DigitalColmer · 2026-10-21
- Claude Code renders lyric videos frame-by-frame in Chromium, no AI video model needed — aitrendz_xyz · 2026-10-21
- Rule-based Self-Layout system renders a full Mos Eisley 3D map with 5,367 sensibly placed objects — jasonkneen · 2026-10-21
- Redditors ask: what does it really take to keep AI agents reliable in production? — djgoel123 · 2026-10-21
- 'How Do People Code Without Claude?' — 'Easy, We Use Codex' — alexmacgregor__ · 2026-10-21
- Google releases free 1-hour AI engineering course, from first agent to autonomous systems — msharmas · 2026-10-21