BED-9446: collect enterprise SCIM by default

README.md

@@ -42,7 +42,7 @@ identifier must be supplied together with `key_path` and `enterprise_name`.
 
 ### Enterprise SCIM and hybrid correlations
 
-When `SOURCES__GITHUB__COLLECT_ENTERPRISE_SCIM=true`, a token with enterprise SCIM access is used to collect both `/scim/v2/enterprises/{enterprise}/Users` and `/scim/v2/enterprises/{enterprise}/Groups`. The collector emits normalized `SCIM_Organization`, `SCIM_User`, and `SCIM_Group` nodes plus `SCIM_Contains`, `SCIM_MemberOf`, and `SCIM_Provisioned` relationships. Install the BloodHound SCIM extension alongside this extension to register the shared SCIM kinds.
+A token with enterprise SCIM access is used to collect both `/scim/v2/enterprises/{enterprise}/Users` and `/scim/v2/enterprises/{enterprise}/Groups`. The collector emits normalized `SCIM_Organization`, `SCIM_User`, and `SCIM_Group` nodes plus `SCIM_Contains`, `SCIM_MemberOf`, and `SCIM_Provisioned` relationships. Install the BloodHound SCIM extension alongside this extension to register the shared SCIM kinds.
 
 `SOURCES__GITHUB__EMIT_LEGACY_SCIM_CORRELATIONS=true` temporarily reproduces GitHound-style Okta-to-SCIM correlation relationships. It defaults to false because a dedicated hybrid correlator should own IdP-to-SCIM matching; GitHub remains authoritative for GitHub's SCIM resources and target-system provisioning relationships.
 

src/openhound_github/helpers.py

@@ -20,6 +20,19 @@ class GraphQLPaginationError(RuntimeError):
     pass
 
 
+def scim_skip_reason(exception: BaseException) -> str | None:
+    """Return a user-facing reason for expected SCIM API unavailability."""
+    if not isinstance(exception, requests.HTTPError) or exception.response is None:
+        return None
+
+    status_code = exception.response.status_code
+    if status_code in (401, 403):
+        return "the configured credentials do not have SCIM access"
+    if status_code == 404:
+        return "the GitHub scope does not expose SCIM endpoints"
+    return None
+
+
 class GraphQLCursorPaginator(JSONResponseCursorPaginator):
     def __init__(
         self,

src/openhound_github/lookup.py

@@ -1,4 +1,5 @@
 import json
+import re
 from functools import lru_cache
 
 import duckdb
@@ -8,10 +9,20 @@
 from openhound_github.runner_ids import runner_group_node_id, runner_node_id
 
 
+_SCHEMA_IDENTIFIER_RE = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
+
+
+def _validate_schema_identifier(schema: str) -> str:
+    if not _SCHEMA_IDENTIFIER_RE.fullmatch(schema):
+        raise ValueError(f"Invalid DuckDB schema identifier: {schema!r}")
+    return schema
+
+
 class GithubLookup(LookupManager):
     def __init__(self, client: DuckDBPyConnection, schema: str = "github"):
-        super().__init__(client, schema)
-        self.schema = schema
+        validated_schema = _validate_schema_identifier(schema)
+        super().__init__(client, validated_schema)
+        self.schema = validated_schema
         self.client = client
 
     def _find_single_row(self, *args):
@@ -188,12 +199,21 @@ def org_login_for_id(self, org_node_id: str) -> str | None:
         )
 
     @lru_cache
-    def projected_enterprise_team_exists(self, org_login: str, slug: str):
+    def projected_enterprise_team_id(self, org_login: str, slug: str) -> str | None:
         return self._find_single_object(
-            f"""SELECT slug FROM {self.schema}.projected_enterprise_teams WHERE org_login = ? AND slug = ?""",
+            f"""SELECT node_id FROM {self.schema}.projected_enterprise_teams WHERE org_login = ? AND slug = ?""",
             [org_login, slug],
         )
 
+    @lru_cache
+    def external_identity_id_for_guid(
+        self, guid: str, environment_slug: str
+    ) -> str | None:
+        return self._find_single_object(
+            f"""SELECT id FROM {self.schema}.external_identities WHERE guid = ? AND environment_slug = ?""",
+            [guid, environment_slug],
+        )
+
     @lru_cache
     def repository_node_ids(self):
         return self._find_all_objects(

src/openhound_github/main.py

@@ -63,6 +63,7 @@ def preproc(ctx: PreProcContext):
         "teams": "teams",
         "team_members": "team_members",
         "saml_provider": "saml_provider",
+        "external_identities": "external_identities",
         "applications": "applications",
         "enterprise": "enterprise",
         "enterprise_organizations": "enterprise_organizations",

src/openhound_github/models/__init__.py

@@ -53,7 +53,14 @@
 from .saml_provider import SamlProvider
 from .saml_service_provider import SamlServiceProvider
 from .saml_issuer import SamlIssuer
-from .scim_user import ScimGroup, ScimOrganization, ScimResource, ScimUser
+from .scim_user import (
+    EnterpriseScimOrganization,
+    EnterpriseScimUser,
+    ScimGroup,
+    ScimOrganization,
+    ScimResource,
+    ScimUser,
+)
 from .secret_scanning_alert import SecretScanningAlert
 from .team import Team
 from .team_member import TeamMember
@@ -117,6 +124,8 @@
     "ScimUser",
     "ScimGroup",
     "ScimOrganization",
+    "EnterpriseScimUser",
+    "EnterpriseScimOrganization",
     "RepoRoleAssignment",
     "Environment",
     "EnvironmentSecret",

src/openhound_github/models/enterprise_helpers.py

@@ -2,5 +2,9 @@ def enterprise_team_node_id(enterprise_id: str, team_id: str | int) -> str:
     return f"GH_EnterpriseTeam_{enterprise_id}_{team_id}"
 
 
+def projected_enterprise_team_node_id(org_id: str | None, team_node_id: str) -> str:
+    return f"GH_Team_{org_id}_{team_node_id}"
+
+
 def enterprise_role_node_id(enterprise_id: str, role_id: str | int) -> str:
     return f"GH_EnterpriseRole_{enterprise_id}_{role_id}"

src/openhound_github/models/enterprise_team_organization.py

@@ -1,16 +1,17 @@
 from openhound.core.asset import BaseAsset, EdgeDef
 from openhound.core.models.entries_dataclass import (
-    ConditionalEdgePath,
     Edge,
     EdgePath,
     EdgeProperties,
-    PropertyMatch,
 )
 
 from openhound_github.kinds import edges as ek
 from openhound_github.kinds import nodes as nk
 from openhound_github.main import app
-from openhound_github.models.enterprise_helpers import enterprise_team_node_id
+from openhound_github.models.enterprise_helpers import (
+    enterprise_team_node_id,
+    projected_enterprise_team_node_id,
+)
 
 
 @app.asset(
@@ -59,18 +60,20 @@ def _assigned_to_edge(self):
     @property
     def member_of_team_edges(self):
         org_login = self.login or self._lookup.org_login_for_id(self.node_id)
-        if org_login and self._lookup.projected_enterprise_team_exists(
-            org_login, self.projected_slug
-        ):
+        projected_team_id = (
+            self._lookup.projected_enterprise_team_id(org_login, self.projected_slug)
+            if org_login
+            else None
+        )
+        if projected_team_id:
             yield Edge(
                 kind=ek.MEMBER_OF,
                 start=EdgePath(value=self.enterprise_team_node_id, match_by="id"),
-                end=ConditionalEdgePath(
-                    kind=nk.TEAM,
-                    property_matchers=[
-                        PropertyMatch(key="environmentid", value=self.node_id),
-                        PropertyMatch(key="slug", value=self.projected_slug),
-                    ],
+                end=EdgePath(
+                    value=projected_enterprise_team_node_id(
+                        self.node_id, projected_team_id
+                    ),
+                    match_by="id",
                 ),
                 properties=EdgeProperties(traversable=True),
             )

src/openhound_github/models/projected_enterprise_team.py

@@ -7,6 +7,7 @@
 from openhound_github.graph import GHNode
 from openhound_github.kinds import nodes as nk
 from openhound_github.main import app
+from openhound_github.models.enterprise_helpers import projected_enterprise_team_node_id
 from openhound_github.models.team import GHTeamProperties
 
 
@@ -43,7 +44,7 @@ def as_node(self) -> GHNode:
             properties=GHTeamProperties(
                 name=self.name,
                 displayname=self.name,
-                node_id=f"GH_Team_{self.org_node_id}_{self.node_id}",
+                node_id=projected_enterprise_team_node_id(self.org_node_id, self.node_id),
                 github_team_id=self.node_id,
                 collected=False,
                 slug=self.slug,

src/openhound_github/models/scim_user.py

@@ -137,6 +137,15 @@ def edges(self):
         return []
 
 
+@app.asset()
+class EnterpriseScimOrganization(ScimOrganization):
+    """Enterprise-scoped SCIM organization input model.
+
+    This remains a distinct asset class so the converter can map enterprise and
+    organization SCIM tables independently while emitting the same graph kind.
+    """
+
+
 @app.asset(
     node=NodeDef(
         kind=nk.SCIM_USER,
@@ -189,7 +198,7 @@ def as_node(self) -> ScimNode:
             id=self.id,
             kinds=[nk.SCIM_USER],
             properties=ScimNodeProperties(
-                name=self.id,
+                name=self.user_name or self.id,
                 displayname=display_name,
                 environmentid=self.scope_node_id,
                 external_id=self.external_id,
@@ -213,15 +222,16 @@ def edges(self):
             end=EdgePath(value=self.id, match_by="id"),
             properties=EdgeProperties(traversable=True),
         )
-        yield Edge(
-            kind=ek.SCIM_PROVISIONED,
-            start=EdgePath(value=self.id, match_by="id"),
-            end=ConditionalEdgePath(
-                kind=nk.EXTERNAL_IDENTITY,
-                property_matchers=[PropertyMatch(key="guid", value=self.id)],
-            ),
-            properties=EdgeProperties(traversable=True),
+        external_identity_id = self._lookup.external_identity_id_for_guid(
+            self.id, self.scope_name
         )
+        if external_identity_id:
+            yield Edge(
+                kind=ek.SCIM_PROVISIONED,
+                start=EdgePath(value=self.id, match_by="id"),
+                end=EdgePath(value=external_identity_id, match_by="id"),
+                properties=EdgeProperties(traversable=True),
+            )
         if self.emit_legacy_correlation and self.external_id:
             yield Edge(
                 kind=ek.SCIM_PROVISIONED,
@@ -231,6 +241,11 @@ def edges(self):
             )
 
 
+@app.asset()
+class EnterpriseScimUser(ScimUser):
+    """Enterprise-scoped SCIM user input model."""
+
+
 @app.asset(
     node=NodeDef(
         kind=nk.SCIM_GROUP,