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.
This commit is contained in:
Elliot Chen
2026-06-05 22:35:51 +08:00
commit 518b8eca85
636 changed files with 160553 additions and 0 deletions

39
tests/fixtures/README.md vendored Normal file
View File

@ -0,0 +1,39 @@
# Test Fixtures
The fixtures in this directory are checked-in artifacts. The two largest
of them were originally produced by converter scripts that have been
removed from the repository — the fixtures themselves are the source of
truth and should rarely need to be regenerated.
## `long_conversation_locomo_caroline_melanie.json`
Derived from the LoCoMo dataset (`snap-research/locomo`, file
`data/locomo10.json`).
**License: CC BY-NC 4.0** — non-commercial use only. This file is
carved out of the project-wide Apache-2.0 license; see `NOTICE` at the
repository root for the explicit declaration.
If you need to regenerate it: fetch `locomo10.json` from the upstream
LoCoMo project, then adapt it to the fields the tests consume
(`everos_session_id`, `everos_user_id_for_test`, `speakers`,
`dialog`, ...). The original conversion script is not part of the
public repository.
## `agent_trajectories/*.json`
Hand-curated subset of agent tool-call traces. The selection that the
e2e tests load is enumerated in
`tests/e2e/test_add_flush_agent_pipeline_e2e.py` — that file is the
source of truth.
If you need to add or replace a trajectory, author the fixture by hand
(or with your own one-off script). The previous converter pulled from
internal paths and is not maintained as part of this repository.
## `search_seed/`
LanceDB / SQLite seed bundle for `/search` e2e. Produced by
`_dump_search_seed.py` (kept in-tree); regenerate by running that
script after a successful `add → flush` pipeline against the LoCoMo
fixture above.