feat(ttp): E.4.a extract decnet/cli/ttp.py with worker run + backfill CLI

The TTP worker entry moved out of decnet/cli/workers.py into its own
module so the TTP CLI surface (worker + admin verbs) is colocated,
mirroring decnet/cli/canary.py / webhook.py / swarm.py.

- New `decnet/cli/ttp.py` with `decnet ttp` (worker, ExecStart-stable
  for decnet-ttp.service) and `decnet ttp-backfill --since-days N`.
- `decnet ttp-backfill` walks Attacker.commands and CanaryTrigger
  history, dispatches each row through the live CompositeTagger,
  persists tags via repo.insert_tags (idempotent INSERT OR IGNORE).
  --dry-run / --source command|canary|all / --batch-size supported.
- Backfill deliberately bypasses bus publish — historical replay
  must not re-trigger SIEM/webhook fan-out per TTP_TAGGING.md
  §"Bus topics" loop-prevention invariant.
- Added `iter_attacker_commands_since` / `iter_canary_triggers_since`
  read-only iterators on TTPMixin + abstract bindings on
  BaseRepository.
- Master-only via gating; both `ttp` and `ttp-backfill` listed in
  MASTER_ONLY_COMMANDS.
This commit is contained in:
2026-05-02 01:35:17 -04:00
parent e84b522fd3
commit 301d3feee9
7 changed files with 673 additions and 55 deletions

View File

@@ -1,4 +1,6 @@
from abc import ABC, abstractmethod
from collections.abc import AsyncIterator
from datetime import datetime
from typing import Any, Optional
from decnet.web.db.models.topology import DeckyRow, EdgeRow, LANRow, TopologySummary
@@ -1320,6 +1322,24 @@ class BaseRepository(ABC):
"""
raise NotImplementedError
@abstractmethod
def iter_attacker_commands_since(
self, since: "datetime",
) -> "AsyncIterator[tuple[Any, list[dict[str, Any]]]]":
"""Yield (Attacker, decoded_commands) pairs since *since*.
Used by ``decnet ttp backfill`` (E.4) to replay shell-command
history through the live tagger. Read-only.
"""
raise NotImplementedError
@abstractmethod
def iter_canary_triggers_since(
self, since: "datetime",
) -> "AsyncIterator[Any]":
"""Yield ``CanaryTrigger`` rows since *since*. Used by backfill."""
raise NotImplementedError
@abstractmethod
async def list_techniques_by_identity(
self, uuid: str,

View File

@@ -12,6 +12,9 @@ per-dialect ``SQLiteRepository`` / ``MySQLRepository`` subclasses
"""
from __future__ import annotations
import json
from collections.abc import AsyncIterator
from datetime import datetime
from typing import Any
from sqlalchemy import func, select
@@ -25,6 +28,7 @@ from decnet.web.db.models import (
TechniqueRollupRow,
TTPTag,
)
from decnet.web.db.models.canary import CanaryTrigger
from decnet.web.db.sqlmodel_repo._helpers import _MixinBase
@@ -275,6 +279,55 @@ class TTPMixin(_MixinBase):
for r in res.all()
]
# ── Backfill iterators (E.4) ────────────────────────────────────
#
# Read-only iterators consumed by ``decnet ttp backfill`` to replay
# historical events through the live :class:`CompositeTagger`. The
# CLI builds :class:`TaggerEvent` objects from these and persists
# results via :meth:`insert_tags` — same idempotent path the bus
# worker uses, no bus publish.
#
# Per TTP_TAGGING.md §"Order of work" / §"Bus topics" the historical
# replay deliberately bypasses bus publish so SIEM/webhook fan-out
# does not re-fire on already-attributed events.
async def iter_attacker_commands_since(
self, since: datetime,
) -> AsyncIterator[tuple[Attacker, list[dict[str, Any]]]]:
"""Yield ``(Attacker, decoded_commands)`` pairs since *since*.
Walks every :class:`Attacker` whose ``last_seen >= since`` and
decodes the JSON ``commands`` blob; non-list / malformed
payloads are skipped silently (the JSON column is best-effort
per the model docstring).
"""
async with self._session() as session:
stmt: Any = (
select(Attacker).where(col(Attacker.last_seen) >= since)
)
res = await session.execute(stmt)
for row in res.scalars().all():
try:
decoded = json.loads(row.commands or "[]")
except (ValueError, TypeError):
continue
if not isinstance(decoded, list):
continue
yield row, [c for c in decoded if isinstance(c, dict)]
async def iter_canary_triggers_since(
self, since: datetime,
) -> AsyncIterator[CanaryTrigger]:
"""Yield :class:`CanaryTrigger` rows fired since *since*."""
async with self._session() as session:
stmt: Any = (
select(CanaryTrigger)
.where(col(CanaryTrigger.occurred_at) >= since)
)
res = await session.execute(stmt)
for row in res.scalars().all():
yield row
async def list_distinct_techniques(self) -> list[TechniqueRollupRow]:
"""Fleet-wide distinct-technique rollup with counts +
most-recent-seen timestamps.