Replace browser login with edge-asserted identity#51
Open
czpython wants to merge 6 commits into
Open
Conversation
czpython
force-pushed
the
commonzenpython/eng-729-edge-identity-extract-request-actor-context-pr-1
branch
from
July 21, 2026 20:58
f1971b5 to
0c7a693
Compare
Base automatically changed from
commonzenpython/eng-729-edge-identity-extract-request-actor-context-pr-1
to
main
July 21, 2026 20:59
Close the stream only when identity actually fails to resolve; drop the count helper the identity test can derive; pin that the session-era 404s and /me set no cookie.
A stale unbound completion attaches to the operator the request resolved instead of minting a rival none-mode account; the credential transaction commits before the provider await so flushed row locks never span network I/O; an identity recheck that resolves a different account broadcasts the invalidation so account-scoped UI remounts.
Nothing past the credential commit may turn a durable connect into a client-visible failure — the single-use flow is already spent. Refresh trouble logs and rolls back its own transaction; the connect returns 200.
Materialize the response DTO before the commit expires the row and keep rollback failures from escaping — after the point of durability the reply depends on nothing but data already in hand.
czpython
force-pushed
the
commonzenpython/eng-730-edge-identity-headernone-door-cutover-onboarding-deploy
branch
from
July 21, 2026 21:02
771d6aa to
8e9cad3
Compare
login_id params, a user-facing error, a refresh docstring, a log line, and two signed-in comments still spoke the deleted door's language.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
druks stops authenticating browsers. The edge asserts identity; druks verifies-and-maps it per request:
DRUKS_AUTH_MODE=none(loopback, exactly one operator account) andheader(trustedDRUKS_AUTH_HEADERassertion, open enrollment; the edge gates who reaches druks).jwttrails in a later PR./api/auth/session,/login/*, and/logoutare gone (they fall to the JSON 404 catch-all); nothing sets cookies. Redis loss no longer signs anyone out.POST/DELETE /api/harnesses/{name}/connection...connects a subscription to the current account; in none/zero the first completed connection creates the sole operator (onboarding = "account has no HarnessConnection", no new state). PAT management admits the operator identity only, never a bearer.AuthProvider/Landing/HarnessLogin→IdentityBootstrap/Onboarding/HarnessConnectFlowoverGET /api/auth/me; a 401 renders an edge/config error, never onboarding; the query cache remounts per account id.Part of ENG-730 (edge-identity arc, PR 2 of 3; stacked on #50)
Open naming question flagged for review:
current_operator_accountvscurrent_browser_accountfor the edge/none-only resolver.🤖 Generated with Claude Code