secrets-engine/tests/test_decision_replay.py

185 lines
8 KiB
Python
Raw Normal View History

"""FLEX-DEC-2026-012 against independent producer request/output fixtures."""
import copy
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
import json
from datetime import datetime, timezone
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
from pathlib import Path
from types import SimpleNamespace
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
import pytest
from secrets_engine.authorization import REQUEST_TENANT, build_action_request, request_digest, validate_decision_envelope
from secrets_engine.decision_check import check_decision, require_supported_pdp_address
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
from secrets_engine.errors import DecisionError
FIXTURES = Path(__file__).parent / "fixtures" / "flex-auth-replay"
CASES = {
"rotate": ("check_request_allow_rotate.json", "decision_rotate.json", "sha256:de67324f54187055307a833235f83ced9fcd3a20952a27b3d19493ed39734345"),
"destroy": ("check_request_allow_destroy_dual_control.json", "decision_destroy_dual_control.json", "sha256:c749ee2dc3cdf927a70a3e5b27cff4d97a438d3264153b4b2e3bcacbaf82091a"),
"deny": ("check_request_deny_wrong_tenant.json", "decision_wrong_tenant_deny.json", None),
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
}
def pair(name):
inp, out, _ = CASES[name]
return json.loads((FIXTURES / inp).read_text()), json.loads((FIXTURES / out).read_text())
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
def validate(request, decision, **over):
# Validate at the actual producer decision time; do not rewrite evidence.
now = datetime.fromisoformat(decision["provenance"]["decision_time"].replace("Z", "+00:00"))
return validate_decision_envelope(decision, request, accepted_policy_packages={"secrets-engine.catalog-lane.lifecycle"}, accepted_policy_versions={"v2"}, now=now, **over)
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
@pytest.mark.parametrize("name", ["rotate", "destroy"])
def test_real_producer_request_validates(name):
request, decision = pair(name)
assert request_digest(request) == decision["binding"]["submitted_request_digest"]
assert request_digest(request) != decision["binding"]["request_digest"]
assert decision["binding"]["request_digest"] == CASES[name][2]
assert decision["provenance"]["registry_snapshot_digest"] == "sha256:f5a309bc0b36721fd6d9ad7f53eb21222162bc2eac62a0ab0802a9a1d51340bb"
assert decision["provenance"]["policy_package_digest"] == "sha256:bd11c5fe77ce6439c65fea225ad6b71d2110efc5e7b5bc9b499c59cd0a53b8b4"
assert validate(request, decision).action == name
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
@pytest.mark.parametrize("name", ["rotate", "destroy"])
def test_correlation_fields_are_excluded(name):
request, _ = pair(name)
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
baseline = request_digest(request)
for key, value in (("id", "other"), ("policy_version", "v99"), ("caring_context", {"extra": "context"})):
assert request_digest({**request, key: value}) == baseline
def test_approval_join_is_between_evaluator_origin_values():
request, decision = pair("destroy")
pdp = request["context"]["approval"]["binding"]["pdp_digest"]
assert pdp == decision["binding"]["approval_binding_digest"] == "sha256:fa07becfaa471394d06aee5fa3cd66352bf0cc69ef24900240489684cda8cd56"
without = copy.deepcopy(request)
del without["context"]["approval"]
assert request_digest(without) != pdp
assert validate(request, decision, expected_approval_binding_digest=pdp).action == "destroy"
@pytest.mark.parametrize("change", ["claim_id", "claim_absent", "purpose", "action", "tenant", "field", "target"])
def test_allow_cannot_be_replayed_for_changed_submitted_material(change):
request, decision = pair("destroy")
if change == "claim_id":
request["context"]["approval"]["approval_id"] = "other-approval"
elif change == "claim_absent":
del request["context"]["approval"]
elif change == "purpose":
request["context"]["purpose"] = "other-purpose"
elif change in ("action", "tenant"):
request[change] = "other"
elif change == "field":
request["resource"]["attributes"]["fields"] = ["other"]
else:
request["resource"]["id"] = "catalog:other"
with pytest.raises(DecisionError, match="submitted request digest"):
validate(request, decision)
@pytest.mark.parametrize("value", [None, "", "sha256:" + "f" * 64])
def test_missing_or_wrong_submitted_binding_refused(value):
request, decision = pair("rotate")
if value is None:
del decision["binding"]["submitted_request_digest"]
else:
decision["binding"]["submitted_request_digest"] = value
with pytest.raises(DecisionError, match="submitted request digest"):
validate(request, decision)
@pytest.mark.parametrize("value", [None, "bad", "sha256:" + "f" * 64])
def test_approval_binding_missing_malformed_or_wrong_refused(value):
request, decision = pair("destroy")
decision["binding"]["approval_binding_digest"] = value
with pytest.raises(DecisionError, match="approval[_ ]binding_digest|approval binding digest"):
validate(request, decision)
def test_claim_free_shortcut_refused():
request, decision = pair("rotate")
with pytest.raises(DecisionError, match="carried approval"):
validate(request, decision, expected_approval_binding_digest=decision["binding"]["request_digest"])
def test_wrong_tenant_and_superseded_policy_refused():
request, decision = pair("deny")
assert decision["reason"] == "wrong_tenant"
assert decision.get("lifetime") is None
with pytest.raises(DecisionError, match="effect is not allow"):
validate(request, decision)
request, decision = pair("rotate")
decision["provenance"]["policy_version"] = "v1"
with pytest.raises(DecisionError, match="policy version"):
validate(request, decision)
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
def test_real_expired_allow_refused():
request, decision = pair("rotate")
fix: exclude correlation fields from the flex-auth request digest Verified the digest join against flex-auth's T03 replay fixtures and found request_digest was hashing fields docs/canonical-request-digest.md excludes. The material is tenant, subject, action, resource, context only: id is correlation, policy_version lives in provenance, caring_context is hashed separately. This engine included all three when present. Because the join adopts the served request id, every real production request would have carried one, so the computed digest would have matched no issued decision and failed closed against every correct allow. Same unsatisfiable shape as the removed AUTHORITY constant. The old pinned constant was computed with the id inside the material, so it was wrong and its passing proved nothing. Replaced with fixture-driven tests over two real envelopes (vendored with provenance) plus a structural test that correlation fields do not move the digest. Both fixtures are needed: input_claim_digests.context appears only with a non-empty context. Also stops computing the native claim digest. The claim's binding.action and binding.target speak approval-engine's vocabulary while ours speaks the catalog's, and no mapping is published; flex-auth makes no cross-check and states the correspondence is ours via pdp_digest. A claim recording no pdp_digest now fails closed naming the missing mapping rather than comparing two different languages. That mapping is a prerequisite for destroy. 274 tests pass. Production still fails closed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD Assistant: claude-code Assistant-Model: opus Assistant-Process: 393550@bnt-lap001 Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4
2026-09-06 14:17:38 +02:00
with pytest.raises(DecisionError, match="lifetime has expired"):
validate_decision_envelope(decision, request, accepted_policy_packages={"secrets-engine.catalog-lane.lifecycle"}, accepted_policy_versions={"v2"}, now=datetime(2100, 1, 1, tzinfo=timezone.utc))
feat: bind the destroy gate to approval_binding_digest and pdp_path The vocabulary mapping this path was waiting on is not coming: gate-house rejected it in GH-DEC-2026-008, because a translation can be confidently wrong and fails open by accepting a claim approved for a different action. The stronger option arrived instead, and both halves are enforced here. flex-auth published binding.approval_binding_digest (FLEX-DEC-2026-007) to fix the circularity this repo reported: a pdp_digest recorded at issue time can never equal the request_digest of the request that carries the claim in its hashed context, so with GH-DEC-2026-008 requiring that equality, destroy would have failed closed forever on a check no correct record could pass. - authorization.approval_binding_digest implements the published exclusion rule, including Go's context,omitempty behaviour when stripping empties the context; digest_material drops an empty context for the same reason. - validate_decision_envelope recomputes the field rather than trusting it, refuses a claim-bearing request whose decision records none, and compares the claim's digest from step 1 against it -- never against request_digest, which still covers the claim so it stays a sound replay identity. - validate_approval_claim requires binding.pdp_path true before using pdp_digest at all. Path intent is never inferred from a digest that happens to be present; pre-schema-v3 approvals carry pdp_path false regardless of any digest they hold. Replay fixtures re-vendored from dd3ce4c. The destroy pins moved a second and final time; approval_binding_digest did not, which is the point. The fixture now demonstrates the property instead of asserting it: we rederive fa07becf... from its own request through our canonical implementation, proving we hash the same material flex-auth does rather than pinning a constant we cannot reproduce. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E4tNMAYcSQmZWUE4wqP4ij Assistant: claude-code Assistant-Model: opus Assistant-Process: 715726@bnt-lap001 Assistant-Session: 80a42b32-cba6-4b23-8be0-68819b1a6092
2026-09-06 20:39:59 +02:00
def test_request_builder_tenant_boundary():
fix: send the tenant the policy package scopes us to, and adopt v2 build_action_request emitted no tenant field at all. The deployed secrets-engine.catalog-lane.lifecycle v2 package reads known_tenant := "tenant:platform" request_tenant := object.get(input, "tenant", "") so an absent tenant is not an ignored field, it matches the wrong_tenant first-denial branch. Every gated action this engine sent would have been denied -- and the omission also produced a request_digest that could match no correctly issued decision, since tenant is hashed material. That is the same class of defect as hashing excluded fields, arriving from the other direction, and again only a real artifact exposed it. Found by answering the GLAS-WP-0015 tenant-alignment question instead of assuming the values lined up. - REQUEST_TENANT is pinned against the vendored allow envelopes, so a package retenanting fails a test rather than denying production. - An empty tenant is refused at build time. - Accepted policy version moves v1 -> v2. v1 had no tenant rule and failed open: a rotate under tenant:coulomb returned allow against the deployed package. flex-auth superseded rather than amended it, because a fail-open correction has to be visible as a version change. A test pins that a v1 decision is refused. - Vendored decision_wrong_tenant_deny.json as the denial evidence glas asked for, with tests that we refuse it on effect before anything else and that a deny legally carries no lifetime. docs/tenant-alignment.md states the three tenant values as this repo holds them. It does not resolve the JWT/store mapping: service_auth.TENANT is tenant:coulomb, which is exactly the value the package denies. That is either two layers sharing a namespace format or one wrong constant, and picking between them without an owner ruling is the fail-open shape GH-DEC-2026-008 rejected for action vocabularies. Both constants stay as they are, deliberately not unified. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E4tNMAYcSQmZWUE4wqP4ij Assistant: claude-code Assistant-Model: opus Assistant-Process: 715726@bnt-lap001 Assistant-Session: 80a42b32-cba6-4b23-8be0-68819b1a6092
2026-09-06 22:32:54 +02:00
entry = SimpleNamespace(id="glas-primary", stage="prod")
args = dict(subject_id="secrets-engine", subject_type="service", purpose="rotation")
request = build_action_request(entry, "rotate", **args)
fix: send the tenant the policy package scopes us to, and adopt v2 build_action_request emitted no tenant field at all. The deployed secrets-engine.catalog-lane.lifecycle v2 package reads known_tenant := "tenant:platform" request_tenant := object.get(input, "tenant", "") so an absent tenant is not an ignored field, it matches the wrong_tenant first-denial branch. Every gated action this engine sent would have been denied -- and the omission also produced a request_digest that could match no correctly issued decision, since tenant is hashed material. That is the same class of defect as hashing excluded fields, arriving from the other direction, and again only a real artifact exposed it. Found by answering the GLAS-WP-0015 tenant-alignment question instead of assuming the values lined up. - REQUEST_TENANT is pinned against the vendored allow envelopes, so a package retenanting fails a test rather than denying production. - An empty tenant is refused at build time. - Accepted policy version moves v1 -> v2. v1 had no tenant rule and failed open: a rotate under tenant:coulomb returned allow against the deployed package. flex-auth superseded rather than amended it, because a fail-open correction has to be visible as a version change. A test pins that a v1 decision is refused. - Vendored decision_wrong_tenant_deny.json as the denial evidence glas asked for, with tests that we refuse it on effect before anything else and that a deny legally carries no lifetime. docs/tenant-alignment.md states the three tenant values as this repo holds them. It does not resolve the JWT/store mapping: service_auth.TENANT is tenant:coulomb, which is exactly the value the package denies. That is either two layers sharing a namespace format or one wrong constant, and picking between them without an owner ruling is the fail-open shape GH-DEC-2026-008 rejected for action vocabularies. Both constants stay as they are, deliberately not unified. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E4tNMAYcSQmZWUE4wqP4ij Assistant: claude-code Assistant-Model: opus Assistant-Process: 715726@bnt-lap001 Assistant-Session: 80a42b32-cba6-4b23-8be0-68819b1a6092
2026-09-06 22:32:54 +02:00
assert request["tenant"] == REQUEST_TENANT
with pytest.raises(DecisionError, match="requires a tenant"):
build_action_request(entry, "rotate", tenant="", **args)
feat: adopt the owner-documented PDP access path, and prove it live Adoption asked for by flex-auth (FLEX-WP-0021-T05) and glas-harness (GLAS-WP-0015), plus the first real decision this engine has obtained from the deployed pin -- which found a defect the fixtures could not. ACCESS PATH. require_supported_pdp_address refuses in-cluster Service names and any non-loopback host. This is no longer a unilateral call: the owner path is documented as loopback kubectl port-forward over the authenticated Kubernetes API, which is what authenticates the responder transitively (FLEX-DEC-2026-010). A Service name from a workstation does not fail, it resolves through the DNS search suffix to an unrelated public host, and since decision records carry no signature, a responder knowing the published package and version can return an allow that passes every check we make. Fail-closed protects against a PDP that is absent, not one that lies. The guard runs before the token is read, so a misdirected request cannot leak it; a test pins that ordering. LIVE PROOF. Minted a 10-minute TokenRequest token (audience flex-auth, SA secrets-engine/secrets-engine, mode 0600 outside the worktree, shredded after), forwarded to the named pod, and sent a real CheckRequest for glas-claude-agent-dev-anthropic. Result: allow, catalog_lane_policy_matched, served by v2 (sha256:bd11c5fe...) -- so the redeploy flex-auth flagged as outstanding has landed and the pin no longer serves the tenant-blind v1. Our tenant fix is confirmed against the real service: binding.tenant is tenant:platform. THE DEFECT IT FOUND. The evaluator enriches from its registry before hashing -- subject gains attributes and tenant, resource gains tenant -- so binding.request_digest is over material we never sent and cannot reproduce. validate_decision_envelope rejects every real allow. Every replay test passes because _request_from() rebuilds the request out of the binding, i.e. the enriched form: a self-consistent fake agreeing with itself, which hid this through three rounds of digest work. Third time a real artifact has beaten a fake in this integration. NOT FIXED, DELIBERATELY. Rejecting a valid allow is wrong in the safe direction. Which fields may be enriched is flex-auth's contract to publish; inferring it means accepting a binding that differs from our proposal in a way we decided was benign -- the fail-open shape GH-DEC-2026-008 rejected for vocabularies and FLEX-DEC-2026-007 for digests. Raised with them. Tenant question closed by operator decision 5ed3fb35: tenant:platform exactly, and service_auth.TENANT stays tenant:coulomb because the two identity layers are to remain distinct. Declining to author that mapping was right -- the answer was neither reading offered. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E4tNMAYcSQmZWUE4wqP4ij Assistant: claude-code Assistant-Model: opus Assistant-Process: 715726@bnt-lap001 Assistant-Session: 80a42b32-cba6-4b23-8be0-68819b1a6092
2026-09-07 00:22:12 +02:00
# --- supported PDP address (FLEX-DEC-2026-010) -------------------------------
@pytest.mark.parametrize(
"url,fragment",
[
# The exact address flex-auth originally handed over. It does not fail
# to resolve from a workstation, it resolves to an unrelated public host.
("http://flex-auth-secrets-engine.flex-auth.svc.cluster.local:8080",
"in-cluster Service name"),
("http://flex-auth-secrets-engine.flex-auth.svc.cluster.local.:8080",
"in-cluster Service name"),
("http://flex-auth-secrets-engine.flex-auth.svc:8080",
"in-cluster Service name"),
# A public address is refused even though it would "work": nothing
# authenticates the responder, so reaching something is not reaching
# the pin.
("http://80.158.43.29:8080", "is not loopback"),
("https://flex-auth.example.com", "is not loopback"),
],
)
def test_unsupported_pdp_addresses_are_refused(url, fragment):
with pytest.raises(DecisionError, match=fragment):
require_supported_pdp_address(url)
@pytest.mark.parametrize(
"url",
[
"http://127.0.0.1:18080",
"http://localhost:18080",
"http://[::1]:18080",
],
)
def test_loopback_forward_addresses_are_accepted(url):
require_supported_pdp_address(url)
def test_the_guard_runs_before_the_token_is_read(tmp_path):
"""A bad address must not cause the bearer token to be read, let alone sent.
The token is the thing a misdirected request would leak, so the address
check has to come first rather than alongside.
"""
missing = tmp_path / "never-read.token"
with pytest.raises(DecisionError, match="in-cluster Service name"):
check_decision(
base_url="http://flex-auth-secrets-engine.flex-auth.svc.cluster.local:8080",
token_file=missing,
request={},
)
assert not missing.exists()