:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
"""Activity-log hooks for SignalScoutConfig and SignalTeamConfig changes.
`ModelActivityMixin.save()` only *emits* `model_activity_signal` — it does not persist
anything. These receivers are the consumers that turn the emitted signal into a stored
`ActivityLog` row. Without them, nothing about either model would ever appear in the
activity log. Imported from `SignalsConfig.ready()` so they register at startup.
"""
import dataclasses
from typing import Any
import structlog
from posthog.models.activity_logging.activity_log import (
ActivityContextBase,
ActivityScope,
Change,
Detail,
changes_between,
field_name_overrides,
log_activity,
)
from posthog.models.activity_logging.model_activity import get_current_trigger
from posthog.models.signals import model_activity_signal, mutable_receiver
from posthog.models.user import User
from .models import SignalScoutConfig, SignalTeamConfig
logger = structlog.get_logger(__name__)
@dataclasses.dataclass(frozen=True)
class SignalScoutConfigContext(ActivityContextBase):
skill_name: str
@mutable_receiver(model_activity_signal, sender=SignalScoutConfig)
def handle_signal_scout_config_change(
sender: type[SignalScoutConfig],
scope: ActivityScope,
before_update: SignalScoutConfig | None,
after_update: SignalScoutConfig | None,
activity: str,
user: User | None,
was_impersonated: bool = False,
**kwargs: Any,
) -> None:
instance: SignalScoutConfig | None = after_update or before_update
if instance is None:
return
changes = changes_between(scope, previous=before_update, current=after_update)
if (
before_update is not None
and after_update is not None
and before_update.allowed_mcp_tools != after_update.allowed_mcp_tools
and not any(
change.field == field_name_overrides["SignalScoutConfig"]["allowed_mcp_tools"] for change in changes
)
):
# The generic diff treats null and [] as empty, but these select different permissions.
changes.append(
Change(
type=scope,
field=field_name_overrides["SignalScoutConfig"]["allowed_mcp_tools"],
action="changed",
before=before_update.allowed_mcp_tools,
after=after_update.allowed_mcp_tools,
)
)
if before_update is None and after_update is not None and after_update.write_scopes:
# `changes_between` records no fields for a creation, so a scout created with a grant
# would leave the log unable to say what it held before the first edit. Write access
# decides what an unattended agent can change, so its first value is part of the trail.
changes.append(
Change(
type=scope,
field=field_name_overrides["SignalScoutConfig"]["write_scopes"],
action="created",
after=after_update.write_scopes,
)
)
if before_update is None and after_update is not None and after_update.allowed_mcp_tools is not None:
for field in ("allowed_mcp_tools", "tool_preset"):
changes.append(
Change(
type=scope,
field=field_name_overrides["SignalScoutConfig"][field],
action="created",
after=getattr(after_update, field),
)
)
log_activity(
organization_id=None,
team_id=instance.team_id,
user=user,
was_impersonated=was_impersonated,
item_id=instance.id,
scope=scope,
activity=activity,
detail=Detail(
changes=changes,
name=instance.skill_name,
# Set by system-driven saves (the inactivity sweep) so an entry with no user reads as
# "this job did it" rather than as an unattributed edit.
trigger=get_current_trigger(),
context=SignalScoutConfigContext(skill_name=instance.skill_name),
),
)
@mutable_receiver(model_activity_signal, sender=SignalTeamConfig)
def handle_signal_team_config_change(
sender: type[SignalTeamConfig],
scope: ActivityScope,
before_update: SignalTeamConfig | None,
after_update: SignalTeamConfig | None,
activity: str,
user: User | None,
was_impersonated: bool = False,
**kwargs: Any,
) -> None:
instance: SignalTeamConfig | None = after_update or before_update
if instance is None:
return
# On create, diff against a fresh default row rather than None. `changes_between` yields
# nothing when `previous` is None, which would drop the create branch of update_or_create() —
# the path Slack onboarding takes to set a channel on a team whose row was never materialized.
# Diffing against defaults keeps an all-default create (team creation, lazy read) silent while
# still auditing a create that carries a non-default value.
previous = before_update if before_update is not None else SignalTeamConfig()
changes = changes_between(scope, previous=previous, current=after_update)
# Only persist saves that moved a field, which also keeps idempotent re-saves out of the log.
if not changes:
return
log_activity(
organization_id=None,
team_id=instance.team_id,
user=user,
was_impersonated=was_impersonated,
item_id=instance.id,
scope=scope,
activity=activity,
detail=Detail(
changes=changes,
# Singleton per team, so there's no per-row name to show; label it where a reader
# would go to find the setting instead of exposing the model name.
name="Inbox settings",
trigger=get_current_trigger(),
),
)