feat(clustering): combined edge weight + medium-tier wiring

The clusterer now drops a single high-tier function call in favor of
a tier-weighted sum. Tier multipliers (high=1.0, medium=0.6, low=0.2,
very_low=0.05) are tuned so the threshold (1.0) admits high-tier
agreement alone while leaving every weaker tier — and every
combination of weaker tiers — under threshold.

Per-tier discipline tested:
- high alone clusters
- medium alone does NOT cluster (supporting signal only)
- low alone does NOT cluster (fixture 1's failure mode)
- very-low alone does NOT cluster (fixture 2's failure mode)
- all three weak tiers stacked still don't reach threshold
- high + medium clusters (high already saturates)

The combination is forward-compatible: low + very-low contributions
are computed today but always project to 0.0 because the production
adapter doesn't populate credentials / ASN-edge inputs into the
fixture path yet. Their contribution becomes load-bearing in commit 7
when the low-tier landing tightens the F1 / F2 bounds.

Fixture 4 (paused_campaign) ratchet added: high-tier signal carries
the multi-day-silence campaign into one identity. Time-agnostic
invariant — silence is irrelevant to the edge weight.
This commit is contained in:
2026-04-26 08:22:10 -04:00
parent de2f4c3a62
commit f7da33726c
4 changed files with 159 additions and 9 deletions

View File

@@ -287,6 +287,36 @@ def test_shared_wordlist_passes_with_production_clusterer():
)
def test_paused_campaign_passes_with_production_clusterer():
"""Fixture 4: one campaign split across two operational windows by
a multi-day silence. Both halves share JA3 + HASSH + payload + C2;
the production clusterer must fold them into one identity. Time-
agnostic invariant: the silence window is irrelevant to clustering."""
from tests.clustering.fixture_harness import assert_fixture_bounds
from tests.factories.campaign_factory import generate, load_yaml
corpus = generate(load_yaml(FIXTURE_DIR / "paused_campaign.yaml"), seed=0)
assert_fixture_bounds(
corpus, _production_clusterer_predict,
FIXTURE_DIR / "paused_campaign.expected.yaml",
)
def test_cluster_observations_medium_alone_does_not_fuse():
"""Two observations sharing only command-sequence (medium-tier)
must stay in distinct clusters — medium is a supporting signal."""
a = Observation(
observation_id="a",
commands_by_phase={"discovery": ("ls", "id", "uname")},
)
b = Observation(
observation_id="b",
commands_by_phase={"discovery": ("ls", "id", "uname")},
)
labels = cluster_observations([a, b])
assert labels["a"] != labels["b"]
def test_vpn_hopping_passes_at_identity_level_with_production_clusterer():
"""Fixture 2: one rotating actor with stable JA3 + HASSH across
5 ASNs. The production clusterer must fold all 5 observations into

View File

@@ -10,7 +10,9 @@ from __future__ import annotations
import pytest
from decnet.clustering.impl.similarity import (
EDGE_THRESHOLD,
Observation,
combined_edge_weight,
from_synthetic,
high_weight_edge,
low_weight_edge,
@@ -179,6 +181,70 @@ def test_observations_carry_no_timestamps():
# ─── from_synthetic adapter ────────────────────────────────────────────────
# ─── combined_edge_weight tier discipline ─────────────────────────────────
def test_combined_high_alone_crosses_threshold():
a = _obs(ja3="ja3-shared")
b = _obs(ja3="ja3-shared")
assert combined_edge_weight(a, b) >= EDGE_THRESHOLD
def test_combined_medium_alone_below_threshold():
"""Single medium-tier match must NOT cluster — medium is a
supporting signal, never a clustering driver on its own."""
a = _obs(commands_by_phase={"discovery": ("ls", "id", "uname")})
b = _obs(commands_by_phase={"discovery": ("ls", "id", "uname")})
weight = combined_edge_weight(a, b)
assert 0 < weight < EDGE_THRESHOLD
def test_combined_low_alone_below_threshold():
"""Credential-only overlap must NOT cluster — fixture 1's failure mode."""
a = _obs(credentials=frozenset({("root", "toor"), ("admin", "admin")}))
b = _obs(credentials=frozenset({("root", "toor"), ("admin", "admin")}))
weight = combined_edge_weight(a, b)
assert 0 < weight < EDGE_THRESHOLD
def test_combined_very_low_alone_below_threshold():
"""ASN-only overlap must NOT cluster — fixture 2's failure mode."""
a = _obs(asn=64500)
b = _obs(asn=64500)
weight = combined_edge_weight(a, b)
assert 0 < weight < EDGE_THRESHOLD
def test_combined_all_weak_tiers_still_below_threshold():
"""Even all three weaker tiers stacked don't reach threshold —
only a high-tier signal does."""
a = _obs(
asn=64500,
credentials=frozenset({("root", "toor")}),
commands_by_phase={"discovery": ("ls",)},
)
b = _obs(
asn=64500,
credentials=frozenset({("root", "toor")}),
commands_by_phase={"discovery": ("ls",)},
)
# 0.6*1.0 (medium) + 0.2*1.0 (low) + 0.05*1.0 (very_low) = 0.85
weight = combined_edge_weight(a, b)
assert weight < EDGE_THRESHOLD
def test_combined_high_plus_medium_clusters():
a = _obs(ja3="ja3-x", commands_by_phase={"discovery": ("ls",)})
b = _obs(ja3="ja3-x", commands_by_phase={"discovery": ("ls",)})
assert combined_edge_weight(a, b) >= EDGE_THRESHOLD
def test_combined_no_signal_returns_zero():
a = _obs()
b = _obs()
assert combined_edge_weight(a, b) == 0.0
def test_from_synthetic_round_trip():
"""The adapter projects a SyntheticAttacker into an Observation
that the edge functions can score over."""