tomtan/EverOS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
79b3df4de22dabf2ea53cd9994ac8f3740475466
EverOS/CHANGELOG.md
Elliot Chen 518b8eca85 chore: initialize EverOS 1.0.0 
md-first memory extraction framework for AI agents.

Markdown is the single source of truth; SQLite holds state and LanceDB
provides the rebuildable vector + BM25 + scalar index. The codebase follows
a single-direction DDD layering (entrypoints -> service -> memory -> infra,
with component / core / config cross-cutting) enforced by import-linter.

Engineering surface:
- Coding conventions in .claude/rules/ (path-scoped) and workflows in
  .claude/skills/ (/commit, /new-branch, /pr).
- GitHub Actions CI runs make lint + test + integration; pre-commit mirrors
  the gates locally (ruff, hygiene hooks, gitlint commit-msg).
- Commit messages follow Conventional Commits, enforced by gitlint.
- make lint also enforces datetime two-zone discipline and OpenAPI drift.
2026-06-06 07:33:17 +08:00

1.6 KiB
Raw Blame History

Changelog

All notable changes to EverOS are documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Unreleased

Unreleased changes on dev will be listed here.

1.0.0 - 2026-06-03

First public release of EverOS — a Markdown-first memory extraction framework for AI agents.

Added

  • Markdown as source of truth — all memory persists as plain .md files you can open, edit, grep, and version with Git.
  • Lightweight three-piece storage — Markdown (truth) + SQLite (state / queue / audit) + LanceDB (vector + BM25 + scalar index). No external services required.
  • Hybrid retrieval — BM25, vector, and scalar filtering in a single LanceDB query.
  • Cascade index sync — editing a .md file triggers a file watcher → entry-level diff → sub-second LanceDB sync.
  • Dual-track memory — user-track (Episodes / Profiles) and agent-track (Cases / Skills).
  • Multi-source extraction — conversations, workflows, agent traces, and file knowledge.
  • CLI + HTTP API — the everos command-line tool and a FastAPI server, async-first throughout.
  • Pluggable providers — LLM / embedding / rerank via the OpenAI-compatible protocol (works with OpenAI, OpenRouter, vLLM, Ollama, …).
  • Decoupled algorithms — memory extraction algorithms live in the standalone everalgo-* libraries published on PyPI.