project-afs
Health Uyari
- No license — Repository has no license file
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 5 GitHub stars
Code Gecti
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
- Permissions — No dangerous permissions requested
Bu listing icin henuz AI raporu yok.
agentic memory orchestration
AFS
AFS is a CLI-first, local-first workspace memory kernel for agents.
It gives a repository an inspectable .afs/ workspace for memory records, search, sessions, attachments, snapshots, and proof-oriented workflows without requiring a daemon or hosted control plane.
What AFS supports today
- workspace bootstrap and inspection (
afs init,afs status) - memory CRUD with append-only history
- text search with the current DSL (
afs search query ... --dsl) and explain output - explicit sessions and attachments
- export/import, event journaling, and adapter loading
- optional local surfaces for serve, sync, workflow, trace/eval, checkpoints, snapshots, and ops reporting
AFS stays inspect-first: filesystem state is the source of truth, indexes are rebuildable, and optional surfaces remain explicit commands rather than background automation.
Install
AFS is currently source-first. Install it from a local checkout:
python -m pip install -e .
python -m afs --help
After the editable install, the afs console command should also be available:
afs --help
Quickstart
Initialize a workspace, store one memory, search it back, and inspect the stored state:
afs init
afs status
afs memory create \
--kind fact \
--body "JWT auth is enabled" \
--tag auth \
--meta project=afs \
--json
afs search query "jwt kind:fact tag:auth meta.project=afs" --dsl --explain --json
afs memory show <memory-id> --json
afs memory history <memory-id> --json
afs session create --json
afs session add <session-id> --role user --content "Summarize the auth note"
afs session show <session-id> --json
Use python -m afs instead of afs if you prefer the module entrypoint during local development.
Documentation
- public docs overview:
website/docs/overview.md - install guide:
website/docs/getting-started/install.md - quickstart:
website/docs/getting-started/quickstart.md - architecture reference:
docs/architecture.md - historical planning docs:
docs/archive/
Development
Run the test suite:
pytest tests -v
Recommended local development entrypoint:
python -m pip install -e .[dev]
python -m afs --help
Public documentation source lives under website/docs/.
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi