feadad850b
Code-review MAJOR MA2. The previous `store_blob(root, data: bytes)` signature forced the entire payload into RAM before the 10 MB cap was checked — a hostile-large output blob could OOM the worker before the limit even fired. New signature: `store_blob(root, stream, *, max_bytes=10_485_760)`. The implementation: - reads from `stream` in 64 KB chunks; - updates the sha256 + writes to `<root>/.tmp-<pid>-<rand>.gz` incrementally; - raises `BlobTooLarge(max_bytes)` as soon as the running total crosses the cap, then unlinks the partial temp file via `contextlib.suppress`; - atomic-renames the temp file to the CAS path `<aa>/<bb>/<sha256>.gz` once the stream finishes; - sets `0o750` on the directory and `0o640` on the file with explicit `os.chmod` (does not rely on the process umask). Updated unit tests cover: BlobTooLarge enforcement (with temp-file cleanup), multi-chunk happy path (1.5 MB payload exercising the 64 KB loop), and `max_bytes <= 0` validation.
Mimic — backend
Sprint 0 skeleton. Python 3.12+ / Flask / SQLAlchemy 2 / Alembic / Pydantic 2.
Layout
backend/
├── src/mimic/
│ ├── app.py # Flask app factory + SocketIO init
│ ├── config.py # Pydantic Settings
│ ├── extensions.py # db, migrate, socketio, login_manager
│ ├── db/
│ │ ├── models/ # SQLAlchemy 2 typed models
│ │ ├── repositories/ # data access per aggregate
│ │ └── migrations/ # Alembic
│ ├── schemas/ # Pydantic 2 DTOs
│ ├── api/ # Flask blueprints (REST)
│ ├── ws/ # Flask-SocketIO namespaces
│ ├── connectors/ # C2Connector ABC + payload mapping
│ ├── orchestrator/ # run state machine (stub in sprint 0)
│ ├── templating/ # Jinja2 sandbox + regex_extract
│ ├── audit/ # append-only writer + rotation
│ ├── reporting/ # WeasyPrint builder (stub in sprint 0)
│ ├── rbac/ # group-based permission matrix (F11)
│ ├── importers/ # ATR + C2 journal (stub in sprint 0)
│ └── cli/ # mimic-cli (click)
└── tests/
├── unit/ # SQLite, pure logic
└── integration/ # testcontainers Postgres
Local dev
make install # uv venv + pip install -e .[dev]
make db-up # docker compose up -d postgres
make db-bootstrap # one-time: create the mimic_audit_writer role (see below)
make db-migrate # alembic upgrade head
make run # flask run (debug)
make test # pytest unit
make test-int # pytest integration (testcontainers)
make lint # ruff + mypy strict
Audit writer role (dev)
mimic_audit_writer is provisioned by the Ansible playbook in production
(decision D-010). For local development, create it manually after make db-up:
docker exec -it mimic-postgres psql -U mimic_app -d mimic \
-c "CREATE ROLE mimic_audit_writer LOGIN PASSWORD 'pick-a-dev-secret';"
Then expose the same secret in MIMIC_DATABASE_AUDIT_URL in your .env. The
Alembic migration grants the INSERT-only permission on audit_log against
this role; if it does not exist, the grant block is a no-op (idempotent).
What sprint 0 ships
- Full §8 data model + Alembic initial migration (Postgres-specific constraints: audit_log write-only role, soc_session hash, c2_credential Fernet column).
C2ConnectorABC + dataclasses +payload_typeenum + factory. No real Mythic/Home implementation (blocked on PR1/PR2).- Jinja2 SandboxedEnvironment +
regex_extractfilter (re2). - Local auth (bcrypt + Flask session) + group-based RBAC matching the F11 permission matrix.
- Flat CRUD on engagements / hosts / TTPs / scenarios.
- pytest baseline + testcontainers Postgres scaffolding.
Out of sprint 0
Orchestrator, WebSocket cockpit, real connectors, report generation, audit rotation.