Skip to content

Latest commit

 

History

History
1281 lines (960 loc) · 85.3 KB

CHANGELOG.md

File metadata and controls

1281 lines (960 loc) · 85.3 KB

Changelog

All notable changes to this project are documented in this file. Dates use ISO format (YYYY-MM-DD).

This repository's current stable release line is 2.x. Full release notes live in docs/releases/ — this file is the short version. Pre-0.1.0 iteration history is archived in docs/releases/legacy-pre-0.1-history.md. Conventions: entry headings are ## [x.y.z] - YYYY-MM-DD; section headings come from Added, Changed, Fixed, Removed, Security, Internal, Notes; issue and pull-request references link as [#N](https://github.com/ndycode/codex-multi-auth/pull/N).

[2.26.0] - 2026-11-04

The runtime proxy and local bridge route native Decisions API requests to API and ZDR credentials, with failover inside the selected pool. A successful resets redeem now reaches a running router, including a hard pin, without a restart. Reset-credit state moves to version 3, so upgrade the CLI and router together. Full notes.

[2.25.0] - 2026-10-31

Codex voice and web search work with runtime rotation on, the packaged-app router can chain a second local proxy, macOS app binds are managed through launchd, GPT-6.1 Terra is a first-class model, and API routes gain US endpoints and a PSP pool. Saved re-logins take effect in a running router, and refresh coordination fails closed. Full notes.

[2.24.0] - 2026-10-20

Regular accounts are used up before emergency backups, last-resort reset recovery spends the soonest-expiring credit first, resets list --refresh shows each credit's expiry, and cached rate limits are rechecked after an hour so pinned accounts recover promptly. Full notes.

[2.23.0] - 2026-10-16

Proxy environment variables reach every outbound request, status defaults to a compact view with a live codex-ma web dashboard, resume --last keeps the right provider, and native WebSocket forwarding recovers safely from upstream drops. Full notes.

[2.22.0] - 2026-10-14

Native Codex dictation rides the managed OAuth pool, sequential scheduling honors saved account order within priority tiers, and API/ZDR routes get daily capability checks plus WebSocket disconnect diagnostics. Full notes.

[2.21.0] - 2026-10-11

Tracking release: merges the community-submitted client_version fix for OAuth model discovery. Runtime behavior is identical to 2.20.0. Full notes.

Notes

  • #734 independently diagnosed and fixed the same upstream client_version requirement as #735 (which shipped in 2.20.0). Both PRs resolved to the same shipped implementation, so the merge records the contributor's work without changing behavior.

[2.20.0] - 2026-10-11

Luna Reserve arrives as a first-class routing fallback, direct model discovery self-heals its now-required client version, and usage rows stop losing token counts on headerless SSE streams. Full notes.

Added

  • First-class Luna Reserve (gpt-reserve) support: registered as a Luna-compatible model with the Luna reasoning ladder and tool capabilities, parsed from the official multi-bucket rateLimitsByLimitId response by limitName: "gpt-reserve", and reported through an additive lunaReserve object in limits --json --refresh (offered/available, used/remaining %, window duration, reset timestamp, limit ID, normal Luna slug). An absent bucket reports unknown rather than 0%, and a genuine quota 429 on a Luna model falls back to gpt-reserve without spending reset credits or feeding ordinary quota bookkeeping (#732)

Fixed

  • check OAuth model discovery no longer fails with HTTP 400 on fresh installs: the now-required client_version query parameter falls back to the installed codex --version when the inventory has no stored version, and the detected value is persisted for the next check (#733, #735)
  • Responses SSE bodies without a reliable Content-Type are sniffed via a bounded 4 KiB prefix instead of being parsed as one JSON document, so usage-ledger rows carry real token counts and stream completion checks still run (#731)

[2.19.1] - 2026-10-06

Light-terminal readability plus a full dependency and toolchain refresh: the login dashboard gains a terminal background mode (auto/dark/light), all five open Dependabot advisories are resolved, and the runtime now requires Node.js 22.19.0 or later. Full notes.

Changed

  • Node.js floor raised to >=22.19.0 (was >=18.17.0): undici 8 and Vitest 5 both require Node 22, and Node 20 is end-of-life. CI now tests on Node 22.x and 24.x (#730)
  • Runtime dependencies: hono 4.13.9 → 4.13.13, ws 8.21.3 → 8.22.0, zod 4.4.3 → 4.6.5, undici 6.29.0 → 8.11.2 (#730)
  • Dev toolchain: Vitest 4 → 5.0.2, ESLint 10.0 → 10.12, typescript-eslint 8.56 → 8.71.1, lint-staged 16 → 17.6, fast-check 4.5 → 4.10, @types/node 20 → 26. TypeScript is split deliberately: typescript 6.0.3 keeps the JS compiler API (eslint, tsserver, tests) while tsgo (typescript 7.0.2, native) runs build and typecheck via an explicit path (#730)
  • GitHub Actions pins moved to current SHAs: checkout v7, setup-node v7, upload/download-artifact v7/v8, codeql-action v4.38.2 (#730)
  • ProxyAgent pins allowH2: false so undici 8's H2-by-default ALPN negotiation keeps the upstream wire protocol at HTTP/1.1 (#730)

Fixed

  • Dashboard title, account labels, section labels, and help text are readable on light terminal themes: a persisted uiColorMode dashboard setting (auto/dark/light, default auto), a CODEX_TUI_COLOR_MODE process override, and COLORFGBG auto-detection pick light-mode foreground tokens across truecolor, ansi256, and ansi16 profiles (#729)
  • Muted prompt text no longer stacks the dim attribute on the muted foreground in light mode, which previously made secondary text faint again (#729)
  • Light-mode ansi16 muted uses normal black: bright black (ansi16(90)) maps to mid-gray in common palettes, which was the original contrast failure (#729)

Security

  • All five open Dependabot advisories resolved, npm audit clean: source-map-js <1.2.2 (high, indexed source-map event-loop DoS), brace-expansion 4.x <5.0.12 (quadratic-expansion ReDoS), vitest/@vitest/mocker 2.1.0–4.1.11 (redirect-mock path traversal / arbitrary file read), and @humanfs/node <0.16.8 (recursive copy follows symlinks outside the source tree) — all dev-graph only (#730)
  • hono stays pinned at 4.13.13, brace-expansion override at 5.0.12, and undici is now the exact-pinned 8.11.2 (see SECURITY.md for rationale) (#730)

[2.19.0] - 2026-10-02

Subscription visibility and launch performance: status, list, and check accounts report the selected workspace, observed plan, and first-use priming state per account, with opt-in priming for unused Pro Max subscriptions — and the shadow Codex home no longer reads multi-gigabyte linked SQLite files into memory on every launch and cleanup. Full notes.

Added

  • Per-account subscription status on status (text and --json), list, and check accounts: selected workspace, observed plan label, autoPrime setting, first-use priming state, and observation freshness (#726)
  • check accounts --prime opt-in first-use priming for unused personal subscriptions; automatic priming for autoPrime policy accounts now recognizes promax/Pro 500 (#726)
  • byWorkspace quota-cache namespace keyed by hashed record+workspace identity so organization members sharing an accountId cannot collide; token-only accounts derive workspace identity from the access token (#726)

Fixed

  • Linked shadow-home SQLite files (main database plus -wal/-shm sidecars) are no longer read into memory for launch snapshots or sync-back — the shadow path is the original file itself (#727)
  • Forked rollouts index and display by their first session_meta; the parent's appended metadata no longer hijacks the rollout id (#727)
  • Concurrent quota-cache writes no longer lose priming outcomes: primingObservedAt merges evidence independently of the quota timestamp, an overlapping timeout cannot erase a completion, and a genuinely newer observation still retires stale evidence (#726)
  • Strict policy reads reject malformed autoPrime/paused/drained rows; load failures surface as unknown instead of off (#726)
  • Disabled or invalid workspace selections never fall back to a sibling workspace or the stored binding, including token-only accounts (#726)

Security

  • undici 6.28.0 → 6.29.0: high-severity fixes for WebSocket permessage-deflate decompression DoS, response splitting via the retry interceptor, and unrequested WebSocket subprotocol DoS

[2.18.0] - 2026-09-30

Model-catalog update tracking upstream openai/codex@b1e72963: GPT-6.1 Sol becomes the default model, the retiring gpt-5.5 family migrates to its named replacements ahead of the 2026-10-14 OAuth shutdown, and the leaked gpt-6-astra-aeon slug retires onto the flagship. Full notes.

Added

  • gpt-6.1-sol support across the model map, resolver, reasoning-effort ladder (low–ultra, default low), capability profile, pricing card, config templates, and the standalone scripts/codex.js wrapper. gpt-6.1-sol is now DEFAULT_MODEL, mirroring upstream commit b1e72963 which made it the catalog's priority-1 model.

Changed

  • Retired gpt-5.5 and gpt-5.5-pro (including dated snapshot ids): they now resolve to gpt-6-sol and gpt-6-astra respectively, matching OpenAI's migration targets ahead of the 2026-10-14 OAuth shutdown.
  • Retired gpt-6-astra-aeon (never a durable catalog id): resolves to gpt-6-astra.
  • Unsupported-model fallback chain re-floored away from gpt-5.5: gpt-6.1-sol → gpt-6-sol → gpt-5.6-sol → gpt-6-luna, with the pro tier stepping gpt-6-astra → gpt-6-sol first.
  • gpt-5 and unknown GPT-5-ish requests now land on gpt-5.6-sol; gpt-5.4 routes to gpt-6-sol, gpt-5.4-mini/nano to gpt-6-luna, and gpt-5.4-pro to gpt-6-astra.
  • Pricing for retired ids is resolved through the replacement model's rate card.
  • Context-window estimates table emptied deliberately: models with uncertain Codex-vs-API limits (gpt-6.1-sol, gpt-6-*, gpt-5.6-*, Daybreak) return null unless a user override is supplied, rather than inventing a window.

Fixed

  • Usage pricing resolves bare aliases (gpt-6.1, gpt-6, gpt-5.6, astra) and effort-suffixed ids through the exact alias map, so raw ledger rows price at the canonical card instead of failing a maxCostUsd budget closed.
  • Wrapper forward bound raised to 5 spawns so the deepest supported-model staircase (gpt-6.1-sol → … → gpt-5.6-luna) completes on the last allowed forward, matching the plugin host's outbound-attempt budget.

[2.17.3] - 2026-09-28

Eighteen-PR review-and-hardening stack: OAuth redaction and boundary-trust fixes, cross-process CAS for governance stores, clock-skew-safe merges, credential-signature backup rotation, V1 migration preservation, reset-backend discovery for the relocated ChatGPT layout, plus perf, packaging, property and chaos test coverage. Full notes.

Security

  • Close OAuth log-redaction gaps in body and URL scrubbers — bearer ordering, fragments, nested query values, encoded separators, token-like hostnames, JSON key names, chunked entropy scrubbing (#719)
  • Enforce trust at library boundaries: audit-log 0600 re-assertion, malformed token-digest rejection, Windows root-relative override refusal, AbortSignal.any listener-leak fix (#710)
  • Close regex DoS on overlong email labels and timer-overflow near-zero failover clamp (#721)

Fixed

  • Cross-process CAS for JSON governance stores under the directory lock (#713)
  • Hybrid updatedAt floor so clock skew cannot lose writes (#723)
  • Backup rotation keyed by credential signature; chained pending-auth healing (#712)
  • V1→V3 migration preserves hybrid-era fields and per-family indexes (#717)
  • constructor/__proto__ keys safe in budget and routing merges (#714)
  • Wrapper signal relay and dead-owner shadow-home sweep safety (#720)
  • Active-account sync fingerprints canonical plus mirror files (#711)
  • Relocated ChatGPT backend discovery; ambiguous reset identities rejected (#724)

Performance

  • Linear account dedup via incremental identity indexes (#708)
  • Cached runtime policy stores with fingerprint validation (#709)
  • Fewer transform allocations; non-throwing breaker probe (#716)

Packaging

  • Ship skills/; live brace-expansion pin via self-ref override (#722)
  • Smaller tarball, faster typecheck, slimmer forward-path imports (#711)

Tests

  • Property coverage for storage/transformer/SSE/budget/redaction (#714)
  • Real fs/net fault injection and live concurrency chaos suites (#715)

Docs

  • Living documentation rewritten from audited architecture (#718)
  • Release hygiene: v2.17.1 notes restored, links re-aligned (#707)

[2.17.2] - 2026-09-28

Version resync over 2.17.1 — no functional changes. 2.17.1 was prepared but never published; the fixes below shipped in the published 2.17.2 artifact. This release re-aligns the package version, changelog, and release-notes filenames. Full notes.

[2.17.1] - 2026-09-28

Automatic subscription priming observes each selected workspace on its own record, shared-organization accounts no longer share one quota entry, and concurrent quota writes merge instead of clobbering. App-bind cleanup now works on non-English POSIX systems. Full notes.

Fixed

  • Automatic first-use checks prime each account's selected workspace independently; quota observations are keyed by credential record and exact workspace, so shared-organization accounts no longer collide and a Personal-workspace check no longer deletes a valid organization entry (#706)
  • Quota cache saves merge against a pre-edit baseline under the file lock, so an older session's save can no longer discard another process's workspace observation (#706)
  • Pending rotated credentials carry recordId, keeping the account's identity key stable through a refresh that only persisted through the journal (#706)
  • unbind-app and helper cleanup now stop owned processes on non-English POSIX locales; the ps -o lstart= identity probe pins LC_ALL=C, so it parses on every system instead of failing and leaking live routers

[2.17.0] - 2026-09-26

Codex CLI 0.156+ no longer refuses org-sourced accounts, and an opt-in native desktop binding adds capability-aware routing, API/ZDR pools, reset credits and Responses WebSockets. Full notes.

Added

  • With runtime rotation on, interactive mcodex resume lists sessions across providers, including desktop sessions saved under openai that Codex's own picker hides. Explicit ids, --last, remote and non-interactive invocations keep native behaviour, and a failed discovery falls back to the native picker. Needs a Codex version with app-server thread/list (#699)
  • rotation bind-app --native keeps the desktop app on its real OpenAI login (Remote Control pairing keeps working) while inference routes through the account pool. Requires cli_auth_credentials_store = "file"; --custom-provider and unbind-app roll it back (#702)
  • Native mode routes each request only to an account and workspace whose catalog advertises the requested model, effort and speed, serves Responses over WebSockets, and schedules subscriptions by a 5% reserve, priority tier and earliest reset. Catalog outages never block inference (#705)
  • A switch pin stays strict in native mode: tiers, the reserve and automatic reset redemption apply only when nothing is pinned, and a model the pinned account doesn't advertise is refused with 403 model_not_available_in_account_catalog (#705)
  • login --api adds API and ZDR credentials as explicit api/<model> and zdr/<model> pools that fail closed and are never a fallback for subscription requests. api-routes.json stores the keys in plain text (mode 0600 on macOS/Linux; Windows relies on your user profile's ACLs) (#705)
  • account priority <index> <0..9> (default 1) and resets list|redeem|auto for earned subscription reset credits; resets auto last-resort redeems only when every eligible subscription is blocked, manual (the default) never does (#705)
  • Subscription priming, opt-in only: check --prime, or per-account account auto-prime <index> on|off (default off, every 15 minutes while a router runs). Priming uses subscription quota (#705)
  • check accounts|resets|capabilities runs one part of check (#705)
  • status --json adds apiAccounts, totalAccountCount, selectionMode, modelInventory, and per-account priority, autoPrime, lastInferenceRequestAt, reset-credit and forecast fields (#705)
  • fix --live rebinds an org-sourced account the backend no longer authorizes, reports rebound-unauthorized-workspace, and resyncs ~/.codex/auth.json when the active account changed (#701)

Changed

  • New logins pick a unique Personal workspace over an organization default. With several workspaces and no unique Personal one, an interactive login asks and a non-interactive login keeps the automatic choice and warns with --org (#705)
  • login checks the chosen workspace against wham/accounts/check (10-second timeout, fails open). An unauthorized automatic choice falls back to the backend's default; an explicit --org is saved as chosen but auth.json gets the authorized default (#701)
  • Plain check also refreshes reset credits and model discovery. Billable API capability probes reuse results younger than 15 minutes; only check capabilities forces them (#705)
  • check rejects unknown arguments with exit 1 (2.16.0 ignored them), so scripts passing extra flags to check need updating (#705)
  • ws (8.21.3) is a new runtime dependency. Downgrading below 2.17.0 drops account priority tiers on the next policy write; see docs/upgrade.md (#705)

Fixed

  • ~/.codex/auth.json no longer gets an org-... id as tokens.account_id, which Codex CLI 0.156+ refuses on every request ("selected workspace missing from routing discovery"). The token's chatgpt_account_id is written instead, a stale org id from an older release is rewritten, and a switch no longer carries the previous account's workspace id onto new tokens (#703, for #700)
  • best --live, forecast --live and report --live keep an explicit or org-sourced workspace binding across a token refresh instead of overwriting it with the token's default (#701)
  • A refreshed token whose accounts.json write keeps failing (Windows EBUSY/EPERM) is kept live and journaled to <accounts-file>.pending-auth.json instead of being discarded after the old refresh token was already spent (#705)
  • On macOS and Linux, the codex app app-server shim in ~/.codex/multi-auth/app-server-shims/ runs the original Node in place instead of a hard link or copy, which current Homebrew Node could not start from, so the desktop app's app-server launch failed (#705)
  • A quota reset header too large to represent (e.g. a 308-digit reset-after-seconds) no longer reads as a window that never resets: reset times are capped at 7 days ahead, and formatWaitTime prints unknown instead of NaNs / Infinitym NaNs (45b7dd20)
  • Windows: a codex.cmd npm shim on PATH is launchable when package and npm root -g lookup both miss (ad59405c)
  • Windows: CODEX_MULTI_AUTH_REAL_CODEX_BIN pointing at codex.cmd or its sh shim resolves to the package's codex.js instead of failing every launch (933c1ff7)

Security

  • hono pinned to 4.13.9 (2.16.0 shipped 4.12.33, inside the <=4.13.4 range of seven advisories including GHSA-8j4g-w8fx-2239) and the dev-only nanoid to 3.3.18 for GHSA-2v37-7h3g-55p8 (#704)

[2.16.0] - 2026-09-23

GPT-6 Sol and Luna are supported, and every retired model now runs on its named replacement. Full notes.

Added

  • gpt-6-sol and gpt-6-luna with the upstream catalog's reasoning ladders, published standard and Fast pricing, and config template entries. Before this, both silently ran GPT-6 Astra (#696)

Changed

  • Retired models (every codex model, gpt-5.1, gpt-5.2, the gpt-5.4 family, gpt-5-mini/nano, the *-chat-latest snapshots) now resolve to the replacement OpenAI names, are removed from the templates, and are priced at the replacement's rate. The fallback floor is gpt-5.5 and the quota probe no longer tries retired models. Codex ids now use the gpt-5.2 prompt family and rotation bucket (#697, #698)

Fixed

  • GPT-6 requests with more than 272K input tokens are priced at the published long-context rate instead of the cheaper short-context rate (#696)
  • status -m <codex model> reads the account family the runtime proxy actually keys, not the unused codex slot (#698)

[2.15.0] - 2026-09-17

Image generation and editing now run through the authenticated runtime rotation proxy, and built-in image_gen is exposed to Codex 0.154.0 sessions without native-auth fallback. Full notes.

Added

  • The runtime proxy accepts POST /images/generations and POST /images/edits (with /v1 aliases) and forwards request bodies unchanged to the upstream image endpoints, reusing bearer authentication, account rotation, managed OAuth refresh, and usage accounting. Image requests get a minimum five-minute upstream-header timeout and are not replayed on ambiguous transport failures (#694)
  • Provider configuration adds a non-secret x-openai-actor-authorization=codex-multi-auth-local compatibility marker, stripped case-insensitively at the upstream boundary, so Codex 0.154.0 exposes its built-in image_gen tool for runtime-rotation sessions (#695)

[2.14.0] - 2026-09-09

A long-running task no longer dies when the backend reports that the selected model is at capacity: the runtime proxy waits it out and re-sends instead of burning through the account pool. Full notes.

Added

  • The runtime proxy waits out a "selected model is at capacity" response instead of rotating. There was no handling for this error before: it fell into the generic status >= 500 branch, which cools the account down and rotates. That is the wrong shape, because capacity is a property of the MODEL rather than of an account, so every account in the pool fails identically, the rotation leaves server-error cooldowns on healthy accounts on the way past, the transient-attempt budget runs out within seconds and the request ends as a pool-exhausted 503. The wait backs off 2s, 5s, 15s, 30s then 60s, an upstream retry-after wins when one is sent, and the whole thing stops at a wall-clock deadline that covers the upstream calls too, not just the time asleep. The responding account is left completely unpenalized: its pool token is refunded and it is never marked rate limited or cooled down. A client disconnect during a wait abandons the request rather than sending another upstream call for a response nobody is reading (#692, closing #689)
  • CODEX_MULTI_AUTH_MODEL_CAPACITY_RETRY_MS sets that deadline: 10 minutes by default, 1 hour maximum, 0 to restore the previous rotate-and-fail behaviour. An unparseable value falls back to the default rather than disabling, so a typo cannot silently switch the feature off (#692)

[2.13.0] - 2026-09-08

A machine-readable quota snapshot, an explicit loopback upstream for the runtime proxy, and a manual switch that actually revalidates a rate-limited account. Full notes.

Added

  • codex-multi-auth limits --json [--refresh] prints a stable quota snapshot for local integrations, so nothing has to scrape terminal output or read the internal cache file. Cached mode makes no network requests; --refresh reuses the dashboard's sequential refresh and its five-minute freshness floor. The payload carries schemaVersion: 1, a selection block naming the pin, the per-family active index and the resulting routedIndex, and one row per account. Absent provider values are explicit JSON null, access tokens and probe-model names are never emitted, and account emails are masked in label exactly as forecast --json masks them (#688, closing #687)
  • CODEX_MULTI_AUTH_RUNTIME_PROXY_UPSTREAM_BASE_URL routes the runtime rotation proxy through an explicit local upstream for operators chaining it into a local inspection proxy. The value must be HTTP with a numeric loopback host (127.0.0.0/8 or [::1]) and an explicit port; a name such as localhost is rejected because the OS resolves it at connect time and the request carries the managed OAuth bearer token. It fails closed: an invalid value, a proxy that cannot start, or runtime rotation being disabled all exit non-zero on a request-bearing invocation rather than falling back to the direct backend (#690)

Fixed

  • switch <n> clears the selected account's rate-limit windows in the same atomic write that sets the pin. After an out-of-band quota reset the markers survived, so the runtime proxy kept treating the account as blocked and deferring or returning 503 until the recorded window expired on its own: the one command meant to say "use this account now" could not unblock it. A 429 that raced the switch is still preserved; one recorded earlier is not (#691)
  • The runtime proxy drops the switched account's cached quota observations when it observes a new selection generation. It previously cleared session affinity and nothing else, so a pre-switch reading kept driving preemptive deferrals against the account just selected (#691)
  • A failed debounced save re-arms with backoff and a bounded retry count. The debounce timer clears itself before the save runs, so one transient failure, such as a Windows antivirus scanner holding accounts.json open, silently discarded every rate-limit window, cooldown and rotated refresh token recorded since the last successful write (#691)
  • flushPendingSave() logs a failure and resolves instead of rejecting. RuntimeRotationProxy.close awaits it with no catch, so a save failure during teardown aborted the rest of proxy shutdown (#691)
  • Every reader of affinityGeneration, and the increment itself, now require a safe integer. The counter is how a running proxy notices a selection change, and past 2^53 Math.max(counter, disk) + 1 equals the counter, so one hand edit or corrupted write to a 16-digit value froze it: each later switch wrote the same number and no proxy ever observed another selection change. A poisoned counter is dropped, so the next write repairs it to 1

[2.12.0] - 2026-09-04

Effort suffixes resolve to one key per model, usage is priced by service tier instead of assuming standard rates, and a class of lookup defect where a prototype member name was mistaken for a model id is closed. Full notes.

Added

  • Service-tier aware pricing. Every rate in MODEL_PRICING is a STANDARD-tier rate and the estimator had no tier input, so a session billed at OpenAI's Fast tier was priced at half its cost and a maxCostUsd cap could overrun without tripping. The tier is read off the Responses payload (both the bare object and the stream-event shape), carried on UsageTokenCounts so it reaches the ledger through every existing path, and applied at Astra's published Fast rate of $20 / $100 per 1M. A tier with no published rate is reported as unknown cost rather than approximated, so a budget fails closed exactly as it does for an unpriced model (#686)
  • gpt-6-astra carries a cached-input rate of $1 standard and $2 on Fast. It shipped without one in 2.11.0, which made Astra the only row in the table billing cached tokens at the FULL input rate: a tenfold over-statement on a cache-heavy session that trips a cost cap far too early. The 90% cached discount is a uniform platform rate every other row already encodes at input/10 (#686)

Fixed

  • max and ultra effort suffixes are stripped from a model id, so gpt-6-astra-ultra no longer keys separately from the model it routes to. Four call sites stripped only none|minimal|low|medium|high|xhigh, a set that predates GPT-5.6: the entitlement cache blocked the same account and model pair twice under two keys, the capability policy recorded outcomes under a key the matrix never reads, and the fallback chain found no row for an effort-suffixed model. The shared stripModelEffortSuffix derives its set from REASONING_EFFORTS and names the one real exception, codex-max and gpt-5.1-codex-max, whose final segment is part of the model NAME (#685)
  • The service tier survives the runtime recorder and a ledger round-trip. createRuntimeUsageRecorder rebuilds the ledger input field by field and normalizeParsedUsageRow rebuilds only the numeric token fields, so a Fast response through the runtime proxy, the default path, priced at the standard rate, and a persisted Fast row read back as standard (#686)
  • A response reporting its tier as fast is recognised. The upstream catalog names the tier priority and displays it as "Fast", so either spelling can arrive; unmapped it became unknown and failed a cost budget closed on a tier that does have a rate (#686)
  • normalizeUsageLedgerRow validates the service tier instead of accepting any truthy value. Every other field in that normaliser is checked against its allowed set, and the ledger read path already rejected an unknown tier, which left the write path as the only unvalidated way onto a row

[2.11.0] - 2026-09-04

GPT-6 Astra and the Daybreak cyber models are first-class, and five defects found while wiring them up are fixed. Full notes.

Added

  • GPT-6 Astra is a first-class model family. gpt-6-astra and the long-horizon gpt-6-astra-aeon each resolve to their own canonical id with the frontier reasoning ladder (low through max, plus ultra, which is rewritten to max on the wire exactly as upstream Codex does). Bare gpt-6 and astra resolve to the flagship, and none/minimal are coerced up to low because Astra does not accept them. A dedicated GPT-6 resolver claims every id the alias table does not name, including the gpt-6-astra-pro plan tier, dated snapshots, and any tier OpenAI adds later, so an unrecognised GPT-6 id can never fall through to GPT-5.5. That silent downgrade is the exact failure v2.5.0 had to fix for GPT-5.6, one major version down. aeon deliberately keeps its own id rather than collapsing into the flagship: it is a behaviourally different model, not a rename. Both shipped config templates carry the two models, and the codex-multi-auth-codex wrapper mirrors the same map, with a model-by-effort parity suite pinning the two together
  • The Daybreak cyber models from the upstream Codex catalog (gpt-daybreak-blue-latest, gpt-daybreak-red-latest, plus daybreak-blue and daybreak-red shorthands) resolve to their own ids. Their slugs carry neither a codex nor a gpt 5 token, so every resolver declined them and asking for the cyber-permissive model silently ran GPT-5.5. An unrecognised Daybreak id resolves to blue, the defensive variant, so a typo cannot upgrade a caller into the permissive one. They stay out of the picker templates, matching their visibility: hide upstream
  • gpt-6-astra is priced at its published launch rate, $10 per 1M input and $50 per 1M output on the standard service tier. No cached-input rate was published, so cached tokens bill at the full input rate, which over-states cost and makes a maxCostUsd budget trip early rather than late. OpenAI's Fast service tier is 2x standard, and no row in this table has ever carried a service-tier dimension, so a session billed at that tier is under-counted by half for Astra exactly as it already is for gpt-5.6-sol and every other priced model. Nothing in this repo selects that tier: fastSession is a local latency setting that lowers reasoning effort, not OpenAI's billed Fast mode. gpt-6-astra-aeon and both Daybreak models have no published rate and are listed in UNPRICED_ROUTABLE_MODELS rather than guessed at, so a cost budget fails closed while they are in the window
  • An unsupported-model fallback chain for Astra: gpt-6-astra steps to gpt-5.6-sol, and gpt-6-astra-aeon steps to the flagship first, since that is still GPT-6 Astra without the long-horizon behaviour. Astra rolls out org by org, so an account that is not entitled yet gets a real unsupported-model response for it. The chain is opt-in (fallbackOnUnsupportedCodexModel defaults to false) and fires only on that response, so it never swaps the model out from under a request the account could have served
  • gpt-5.6-sol gains its own chain entry to gpt-5.5. The chain is walked one hop at a time (index.ts reassigns the model and passes that as the next requestedModel), so a model with no entry of its own ends the walk. GPT-5.6 shipped without one, which would have stranded the Astra path a rung short of a model every account has. Terra and Luna stay without entries: nothing steps into them
  • All four new models are listed in UNESTIMATED_ROUTABLE_MODELS. OpenAI published two different context windows for Astra on launch day, 1.05M for the API surface and 272K for Codex, and this wrapper talks to the Codex backend. Set contextBudgetGuardModelWindowOverrides once you know your real ceiling

Fixed

  • The codex-multi-auth-codex wrapper resolves codex model ids that its explicit list does not name, matching lib. Its final clause tested model === "codex" where resolveCodexCatalogModel tests for the codex substring, so ids such as gpt-6-codex resolved to nothing in the wrapper while lib resolved them to the current codex model

  • The wrapper buckets Daybreak ids into the gpt-5.2 prompt family for status instead of returning no family at all. Their slug matched none of resolveModelFamilyForStatus's branches

  • resolveModelFamilyForStatus strips the provider prefix before classifying. Every branch is a startsWith, and resolveStatusModel forwards the raw --model value, so openai/gpt-6 matched none of them and status routing fell back to activeIndex instead of the family index. This affected openai/gpt-5.6-sol and openai/gpt-5.1 the same way before GPT-6 existed

  • gpt6, written with no separator, resolves to Astra rather than GPT-5.5. tokenizeModelId splits on non-alphanumeric runs, so it produced a single gpt6 token, the gpt + 6 pair never formed, and the id fell to DEFAULT_MODEL while still passing capability-policy's catalog gate. Gate and resolver now agree on it

  • The codex-multi-auth-codex wrapper retries the GPT-6 rows too. It keeps its own WRAPPER_UNSUPPORTED_MODEL_FALLBACK_CHAIN, which shipped without them, so codex-multi-auth-codex --model gpt-6-astra on an unentitled account exited with the failure code while the plugin-host path retried, and the docs advertised the retry for both. A parity test now pins the GPT-6 rows across the two tables. Note the wrapper's retry has always been unconditional rather than gated on fallbackOnUnsupportedCodexModel; it prints the swap to stderr, and the docs now say which path gates it and which does not

  • The bare-astra branch of the GPT-6 resolver is anchored to a GPT-6 version token or to an id with no GPT version at all. It existed for picker labels such as Astra Pro, but it also claimed any id merely containing astra, so a gpt-4-astra-x would have run the frontier model for a caller who named GPT-4

  • capability-policy's catalog gate matches the tokenizer's separator rule. It tested a hand-picked [-_\s] where tokenizeModelId splits on any run of non-alphanumerics, so gpt.6-turbo and gpt---5 kept their raw string as the policy key while routing resolved them to a canonical model, leaving the store tracking a key no request reads. It is still narrow enough that gpt-4o, gpt-4.5-preview and gpt.4-turbo do not reach the resolver

  • Prototype member names are no longer mistaken for model ids. Model strings reach several lookup tables raw from the caller, so constructor, __proto__ and toString indexed them and came back truthy. Three consequences, all fixed: estimateUsageCostUsd returned NaN rather than null, which silently made a maxCostUsd cap unenforceable because NaN >= limit is false; the unsupported-model fallback chain threw TypeError: targets is not iterable inside the request path; and models --model <name> emitted a row whose normalizedModel, promptFamily and capabilities were all undefined. The fallback chains are now null-prototype objects, which also stops an unsupportedCodexFallbackChain key of __proto__ reassigning the object's prototype

[2.10.0] - 2026-08-31

A pinned account gets a real bounded retry instead of a single attempt, and a new opt-in guard can pause a session before it overflows the model's context window. Full notes.

Added

  • Context Budget Guard, experimental and shipping disabled. Enable it from Settings → Experimental or with contextBudgetGuardEnabled. It tracks how full each session's context window is from the tokens each turn actually carries (input_tokens plus the reasoning-stripped output, never total_tokens) and answers the next request locally with a /compact notice once usage crosses contextBudgetGuardHardPercent, before that request costs an upstream round-trip. A soft threshold attaches a non-blocking x-codex-context-budget-percent header instead. Window sizes are best-effort estimates, not published facts, so contextBudgetGuardModelWindowOverrides always wins and a model with no estimate is skipped rather than guessed at. The pause is one-shot per measurement: it drops what it recorded as it fires, so the /compact turn it asks for is never blocked by it (#681)

Fixed

  • A pinned request retries a transient upstream failure instead of returning codex_pinned_account_unavailable on the first one. Every transient branch cools the account down before the next selection pass, and selection refuses a cooling-down pin, so the retry budget could never be spent. A pinned retry now waives that account's own cooldown, and only from the second pass of the same request, with 250ms/500ms/1s backoff and a cap of min(retryAllAccountsMaxRetries + 1, 4) upstream attempts. Rate limits, open circuits, disabled accounts and policy blocks all still stop it, the cooldown still applies to every other request and still sets retry_after_ms, and a pin already cooling down on arrival is still refused without an upstream call (#683)
  • A pinned request is no longer refused by the local token bucket. That bucket spreads load across the selectable pool, and a pin has no alternative account, so a drained bucket could only reject a request the pinned account could serve. Circuit-breaker admission still applies, and the bucket is neither consumed nor refunded for a pin (#682)
  • account_skip_reasons names the admission gate that actually rejected a request. The reason was re-derived from account state afterwards, which reports the first blocker it finds — so a drained bucket on a cooling-down account was reported as the cooldown, and a rejection from a half-open circuit whose probe another request had just claimed was reported as an exhausted bucket (#682)
  • A context-window override below 1 token is dropped instead of being stored as a window of zero. contextBudgetGuardModelWindowOverrides validated positivity on the raw value and floored afterwards, so any value in (0, 1) was accepted by the schema and then silently ignored (#681)

[2.9.2] - 2026-08-30

A maintenance release with no runtime, CLI, or configuration changes: npm run typecheck:scripts failed on every clean checkout. Full notes.

Fixed

  • npm run typecheck:scripts builds generated output before it runs the compiler. scripts/codex-multi-auth.js imports ../dist/lib/codex-manager.js and tsconfig.scripts.json typechecks that file with checkJs, so on a fresh clone the script stopped with TS2307: Cannot find module '../dist/lib/codex-manager.js'. The build now lives in the script itself, matching pack:check, coverage, bench:runtime-path and generate:schema, which covers all three CI call sites and a clean local tree in one place (#680, thanks @smpark00)

Changed

  • CONTRIBUTING.md lists npm run typecheck:scripts in the documented local gate, now that it works from a clean checkout (#680)
  • The validate CI job drops a build step made redundant by the script change, leaving per-job compile counts unchanged, and test/ci-workflows.test.ts derives its build-ordering guard over every job in ci.yml instead of a hardcoded pair, reading steps by YAML key position so a commented-out or if:-guarded build cannot satisfy it (#680)

[2.9.1] - 2026-08-28

A --cost budget can no longer be silently defeated by the models most worth capping, and refreshing one account from the login dashboard stops disturbing the others. Full notes.

Added

  • codex-multi-auth login --account <index|email|account_id> re-authenticates exactly one saved account, leaving the active selection, every per-family rotation position, and a manual switch pin untouched. The write is refused if the OAuth identity is not the account you named, so the previous credentials stay intact. Implies --preserve-selection, and cannot be combined with --org (#679, thanks @fnmendez)

Fixed

  • A --cost budget is enforced for models with no published price instead of counting them as free. 2.9.0 fixed the guard comparing 0 >= limit, but eight of the fifteen routable model ids (gpt-5.4-pro, gpt-5.5-pro, gpt-5.2-pro, gpt-5.4-mini, gpt-5.4-nano, gpt-5-mini, gpt-5-nano, gpt-5.1 — every pro tier among them) had no MODEL_PRICING entry, so their rows priced to null, aggregated as $0.00, and a cost cap never fired: two million tokens on gpt-5.5-pro passed a --cost 1 cap. Usage summaries now carry unpricedRequests and a --cost limit fails closed with cost limit cannot be evaluated. The missing rates are deliberately not guessed — a wrong figure would move the trip point rather than fix it (fe9b0d1)
  • A plain login no longer resets every model family's rotation position onto the account just signed in. Families track which account they are on independently, so one that had rotated away from a rate-limited account kept its own position; adding an account discarded that. Families that were following the global selection still follow it (fe9b0d1)
  • Refreshing an account from the login dashboard no longer rewrites the native ~/.codex/auth.json unconditionally. Refreshing account 3 while account 1 was active handed plain codex account 3's credentials, silently switching accounts; the sync now runs only when the refreshed account is the one Codex is using — which also means refreshing the active account does update it, where a selection-preserving refresh previously left it on expired tokens (#679)
  • Refreshing a disabled account no longer returns it to rotation. The merge set enabled: true unconditionally, so topping up a credential silently undid a deliberate disable; a targeted refresh now preserves the flag and says the account is still out of rotation (#679)

[2.9.0] - 2026-08-25

budget set --cost and --tokens were accepted but never enforced, a dead network path counted against account health, and account removal left live sessions routed to the wrong account. Full notes.

Added

  • The usage ledger records real token counts and cost. Both the runtime rotation proxy and the plugin host read the upstream usage object out of the response as it streams past, so codex-multi-auth usage reports actual consumption instead of zeros (#678)
  • Transport failures emit a trace-correlated proxyLog.error line. That path never throws, so it previously produced no structured log entry at all (#677)

Fixed

  • maxTokens and maxCostUsd budget limits are enforced. Every ledger row recorded zero tokens and zero cost, so the guard compared 0 >= limit and never fired — budget set --cost 50 allowed unlimited spend, with only --requests doing anything (#678)
  • Reasoning tokens are subtracted out of the output bucket before recording. The upstream reports output_tokens inclusive of output_tokens_details.reasoning_tokens while the cost estimator prices the two as disjoint, so recording both raw would bill reasoning twice and trip a cost cap early (#678)
  • A streaming request's ledger row is written when the stream ends rather than when the handler returns, which is the only point at which its token counts exist. A client that disconnects mid-stream still records a row, without counts (#678)
  • An upstream transport failure before response headers no longer feeds the account's circuit breaker or health tracker. A socket reset, TLS drop, or fetch timeout is a property of the network path, not of the account, and three of them used to open that account's breaker for ~30s during an outage that affected every account equally (#677)
  • Deleting an account, restoring a backup, and the automatic removal of revoked-token accounts bump affinityGeneration. Session affinity is keyed by account index, so without the bump a running proxy kept routing in-flight conversations to whatever account slid into the vacated slot (#678)
  • Two accounts with neither an accountId nor an email no longer share one account-policy entry. Both hashed the literal "unknown", so pausing, draining, or tagging one applied to all of them; the key now falls back to the refresh token and remains independent of account position (#678)
  • withStreamingFailover respects consumer backpressure. It enqueued every chunk as fast as the upstream delivered it, so a slow client buffered the whole response in memory — a 50-chunk stream was fully drained for a consumer that had read one chunk (#678)
  • A transport failure records a runtime skip reason for the account it happened on, instead of clearing the skip-reason overlay that forecast and report read (#677)

[2.8.7] - 2026-08-21

codex "your prompt" took the slow shadow-home startup path, prompt text after -- was parsed as configuration, and the pinned-account 503 called every recovery deadline a limit reset. Full notes.

Fixed

  • A root TUI launch carrying the optional initial prompt (codex "your prompt") keeps the canonical CODEX_HOME. It was classified as a subcommand launch and took the shadow home, whose omitted SQLite state made Codex rebuild its whole thread index from rollouts before submitting the prompt you had already typed (#674)
  • A prompt forced with -- is treated as a prompt, so codex -- exec opens a session whose prompt is the word "exec" rather than dispatching the exec subcommand — matching how Codex itself parses it (#674)
  • The launcher no longer reads flags out of prompt text: --account, -m/--model, and --config= written after a -- are left alone instead of being stripped, treated as settings, or rewritten in place on a retry (#674)
  • Wrapper-injected options are placed on the option side of -- instead of appended. Appending put them in the subcommand's positional list, so codex exec -- ... failed with unexpected argument '-c' found, codex sandbox -- echo hi passed the pair to the sandboxed command, and codex mcp add t -- echo hi wrote it into that server's stored args in config.toml (#674)
  • The pinned-account 503 words its recovery deadline by the blocker actually holding the account — circuit breaker, cooldown, or rate limit — instead of calling every deadline a recorded limit reset, which made provider outages read as blown subscription quotas (#676)
  • That wording is taken from the pin's live runtime state rather than the retry loop's selection verdict, so the common case where the pinned account 429s mid-request no longer reports the internal token already-attempted with a deadline attributed to nothing (#676)
  • The quota phrasing is used only when the rate limit is what bounds recovery; when a cooldown or breaker ends later, the deadline is worded neutrally (#676)

Changed

  • docs/development/ARCHITECTURE.md and docs/development/CONFIG_FLOW.md describe the interactive-TUI branch as covering prompt-bearing launches, matching the routing the wrapper now performs (#674)

[2.8.6] - 2026-08-16

forecast --model reported an account ready while every request to it failed, and the pinned-account 503 told forced pins to run a command that clears nothing. Full notes.

Fixed

  • forecast, best, and report with --model check that model's prompt family instead of a hardwired codex family. An account held down by a live limit for the family you asked about no longer reports ready with no reasons (#670)
  • Those commands no longer report a delay caused by a rate limit on a different model in the same family — availability now uses exactly the two keys account selection consults (#670)
  • When a family-wide and a model-specific limit are both active, the reported wait is the later one rather than the earlier, so the stated time is actually usable (#670)
  • A recorded rate-limited state backed by a live limit for the requested family is no longer discarded as stale (#670)
  • The pinned-account 503 no longer advises unpin for a pin set by --account / CODEX_MULTI_AUTH_FORCE_ACCOUNT, which unpin cannot clear. Forced pins are told to relaunch (#671)
  • That 503 now reports when the account recovers, taking the latest of its rate limit, cooldown, and circuit-breaker deadline (#671)
  • A permanently blocked pin — disabled, no enabled workspace, invalidated token, policy block — reports no recovery time instead of a deadline that expires into another 503 (#671)
  • That suppression now also holds when the request itself is what disabled the account, which previously surfaced as already-attempted and hid the block (#671)
  • A corrupt timestamp in stored account state no longer turns the pinned 503 into a generic 500 carrying no account index, reason, or skip map (#671)

Added

  • pin_source, reset_at, and retry_after_ms on codex_pinned_account_unavailable responses (#671)

Changed

  • docs/reference/error-contracts.md documents the new fields, both pin kinds, and how retry_after_ms differs between the pinned and pool-exhausted codes (#671)

[2.8.5] - 2026-08-13

Background rotation helpers never shut down — 183 of them at 5.58 GB on one machine, the oldest 33 hours past a 12-hour timeout. Full notes.

Fixed

  • Helpers no longer treat a reused process ID as a live launcher. Owner identity is now process ID plus start time (#663, #664)
  • Each helper writes its own status file instead of all of them overwriting one, so rotation status no longer reports a random helper (#663, #664)
  • Helper metadata is cleaned up on exit and swept on the next launch, instead of accumulating forever — 701 leftover files on one machine (#663, #664)
  • Helpers left behind by a short-lived command are reaped in 15 minutes rather than 12 hours (#665)
  • That reaper no longer kills a live codex app session that is simply idle between messages (#669)
  • rotation unbind-app no longer strands owner files it can't pair with a status record (#666, #669)
  • rotation status can no longer name one helper on its status line while marking a different helper's account as current (#667, #669)
  • A stale status file with a reused process ID no longer counts as a running helper (#667, #669)
  • Malformed process IDs in status files are rejected instead of being probed (#667, #669)

Added

  • CODEX_MULTI_AUTH_APP_ROTATION_MAX_LIFETIME_MS — hard cap on helper lifetime, default 24h, 0 disables (#664)
  • CODEX_MULTI_AUTH_APP_ROTATION_DETACHED_IDLE_MS — shorter timeout for a helper whose launcher died and that never served a request, default 15m, 0 keeps the full idle timeout (#665, #669)

Changed

  • Helper status files are now per-process: runtime-rotation-app-helper.<pid>.json. The old shared path is still read but no longer written (#664)
  • The stale-metadata sweep runs after the helper spawns, so it no longer delays codex app or TUI startup (#669)
  • rotation unbind-app stops helpers in parallel instead of one at a time (#669)

Security

  • Removed the deprecated @openauthjs/openauth dependency. PKCE is now generated directly from node:crypto per RFC 7636 (#661)
  • Test-only failure hooks in the published wrapper now require an explicit opt-in, so a stray environment variable can't arm one (#668, #669)

[2.8.4] - 2026-08-12

codex app-server didn't work at all with rotation on, which is the default. Full notes.

Fixed

  • app-server refused to start on any machine that had run one before, and served a frozen thread list when it did start. It now uses your real Codex home rather than a mirrored copy (#659, #662)
  • The CLI shim app-server didn't need was setting CODEX_MULTI_AUTH_RUNTIME_ROTATION_PROXY=0 on everything the server spawned, so nested commands ran unrotated and billed the wrong account (#662)
  • A short-lived app-server left a helper process running for 12 hours after exit (#662)
  • A helper that couldn't start surfaced as an unhandled-rejection stack trace, leaking a temporary directory. It's now a one-line error and exit 1 (#662)

Changed

  • A rotation proxy that can't start now fails app-server outright rather than silently running it unrotated — a wrong-account bill is invisible, a failed start isn't (#662)

[2.8.3] - 2026-08-09

Fixed

  • Quota windows with no label, duration, or reset time are hidden instead of rendering as a permanent 100% (#657)

[2.8.2] - 2026-08-09

Five fixes across login, uninstall, diagnostics, and quota. Full notes.

Fixed

  • Manual and incognito login now show the complete authorization URL, so pasting the callback back no longer fails with a state-mismatch error (#652)
  • Uninstall instructions pointed at the old scoped package name, which can't remove what's installed (#653)
  • why-selected showed a winning account the rotation proxy would never have picked. It now runs the same gates as the real thing (#654)
  • Accounts with a revoked token are marked token-invalid — re-login needed and stay out of rotation instead of rejoining after a cooldown (#655)
  • Exhausted weekly and monthly quotas wait for the real reset instead of retrying after a flat two hours and hitting an immediate 429 (#656)
  • Rotation helpers verify they own a process before shutting it down, so an unrelated process is never mistaken for one of ours (#658)

[2.8.1] - 2026-08-02

Fixed

  • mcodex resume and fork hung on a blank screen. They were using a temporary copy of your Codex home that deliberately omits the session database, so the thread you asked for genuinely wasn't there (#647, #648)
  • The shell prompt didn't always come back after an interrupted Codex exit (#648)
  • --help no longer starts a rotation proxy and leaves a background helper behind (#648)

Security

  • hono 4.12.21 → 4.12.33: JSX context leaking between requests, XSS via the cx() escaping bypass (#650)
  • undici 6.25.0 → 6.28.0: Set-Cookie injection, WebSocket fragment DoS, keep-alive queue poisoning, SameSite downgrade. Staying on 6.x deliberately — 7.x would raise the Node floor to 20 (#650)

[2.8.0] - 2026-07-28

Fixed

  • macOS stopped asking for the login keychain on every launch. The official CLI reads cli_auth_credentials_store from config.toml, and we only wrote it during a login or switch — so a third-party front-end running the official binary kept hitting the keychain (#641, #642)
  • Interactive sessions no longer reindex your session history on every launch. They run against your real Codex home instead of a temporary copy (#639, #643)
  • doctor --fix can repair the credential store directly, and reports correctly on a machine with no accounts yet (#642)

Changed

  • cli_auth_credentials_store is written on first run and re-checked on startup, not just during a login or switch. Opt out with CODEX_MULTI_AUTH_ENFORCE_CLI_FILE_AUTH_STORE=0 (#642)
  • Two interactive sessions can now run at once against the same state, the same as running the official CLI twice (#639)

[2.7.1] - 2026-07-24

A bug hunt. Two of these could route to the wrong account or stall refreshes machine-wide. Full notes.

Fixed

  • A manual pin could route to a different account than the one you pinned. Pins are held by position, and account removal only remapped the active index — so removing a revoked account silently shifted your pin (#638)
  • One failed refresh blocked refreshes for every process for 20 seconds, because failures were cached alongside successes (#637)
  • A refresh waiting on the shared lease could be evicted and start a second refresh for the same token, which fails with invalid_grant on a healthy account (#637)
  • Monthly (Codex Business) windows were labelled 5h (#635, #636)
  • An account at 100% used with no reset time could still be recommended as your best option (#637)
  • Exhaustion was decided on a rounded percentage, so 99.6% used benched an account that still had quota (#637)
  • import and legacy migration dropped your pin and reset the affinity counter (#637)
  • Session trimming cut the leading instructions it had just decided to keep (#637)
  • A timed-out request never got its token back, gradually starving the bucket (#637)
  • Project- and profile-scoped budgets never applied — stored under a normalised key, looked up with the raw one (#637)

[2.7.0] - 2026-07-23

Added

  • check now shows when each quota window resets, not just how much is left (#633)

Fixed

  • Quota percentages kept their red/yellow/green colouring once a reset time was appended (#633)

[2.6.1] - 2026-07-14

Fixed

  • OAuth login on WSL. No browser ever opened, because WSL reports itself as Linux and xdg-open isn't installed there. It now opens the Windows browser (#630)
  • The manual-paste clipboard copied to the distro's clipboard rather than the Windows one you paste from (#630)
  • A lost OAuth callback now explains the Windows/WSL port conflict instead of just timing out (#630)

[2.6.0] - 2026-07-11

Added

  • Diagnostics probe with GPT-5.6, falling back for accounts without access (#627)

Fixed

  • GPT-5.6 now appears in older Codex builds' model pickers (#626)
  • Upgrades no longer miss newly shipped models — the config installer merged the provider block shallowly, so an upgraded config never gained anything new (#626)
  • The wrapper understands GPT-5.6. It re-implements the model map and never got the v2.5.0 work, so gpt-5.6-* requests silently resolved to gpt-5.5 (#626)

Changed

  • pidOffsetEnabled defaults on, so parallel agents spread across accounts instead of all picking the same one and cascading into 429s (#628)

[2.5.0] - 2026-07-10

Added

  • GPT-5.6 (Sol, Terra, Luna) as first-class models, plus the max and ultra reasoning tiers above xhigh. ultra is rewritten to max before the request is sent, matching upstream

Fixed

  • gpt-5.1-codex-max is a model id, not Codex at max effort — it was nearly rerouted onto plain gpt-5.1-codex
  • Unrecognised gpt-5.6-* ids silently ran GPT-5.5

[2.4.0] - 2026-07-09

Added

  • codex-multi-auth-codex --account <index|email|id> forces one invocation onto a single account. Ephemeral (never touches your saved pin) and fail-hard (never silently uses a different account) (#623, #624)

[2.3.3] - 2026-06-19

Nine fixes across rate limiting, refresh, streaming, and storage. Full notes.

Fixed

  • A single bad retry-after could bench an account for 31 years. Windows are now clamped to 7 days (#617)
  • A slow refresh could delete a lock it no longer owned, leaving two processes refreshing at once — and since refresh tokens are single-use, logging the account out (#617)
  • A routine save could revert a freshly rotated token, permanently breaking that account's next refresh (#617)
  • A transient 429 benched an account for hours by folding the healthy weekly window into the retry delay (#617)
  • forecast recommended a strictly worse account, because it took the longest of both quota windows rather than the binding one (#617)
  • Upstream streaming failures were reported to the client as success, with the account recorded as healthy and retry suppressed (#617, #618)
  • The SSE parser required a space after data:, so a spec-valid line parsed as zero events (#617)
  • The V1→V3 storage migration discarded migrated account bodies, so a rate-limited account read as available and could burst 429s (#619)
  • OAuth expires_in accepted zero and negative values, minting an already-expired token and driving a tight refresh loop (#619)

[2.3.2] - 2026-06-16

Fixed

  • An orphaned app-bind left config.toml pointing at a dead proxy with no CLI way out. rotation unbind-app now self-heals without a backup, and rotation status reports "bound but unmanaged" instead of "not configured" (#614)

[2.3.1] - 2026-06-16

Added

  • codex-multi-auth history lists every local Codex session regardless of provider. codex resume filters by the provider recorded in each session, so rotation hides sessions created under the native provider — the files were always there (#612)

[2.3.0] - 2026-06-15

Fixed

  • A permanent 503 that wedged rotation even with healthy accounts. Recovery reloaded from disk, restoring the same state that had wedged the pool, and the guard refused to run in exactly the situation it existed for (#606, #607)
  • Two cooldown paths changed account state without scheduling the write, so a restart dropped it (#608, #609)

[2.3.0-beta.3] - 2026-06-11

Fixed

  • Streaming stalled indefinitely for slow clients, buffering without bound
  • Account deduplication needed more than one pass; it now loops until stable

Security

  • Atomic writes use crypto.randomBytes for temporary filenames instead of Math.random() (#517)

[2.3.0-beta.2] - 2026-06-11

Added

  • Opt-in sequential scheduling drains one account before moving to the next, so quota windows stagger instead of resetting together (#509)

Fixed

  • Drain-first mode advanced its pointer on a transient fallback, breaking the invariant it exists for
  • An expired token could be forwarded after a successful refresh, producing a 401 and a wrong invalidation cooldown
  • Flagged-account restore dropped workspaces, so a multi-workspace account lost its list permanently
  • A transient disk failure during quota-cache reload wiped every other account's data
  • Login reported Added account when it had updated or rebound an existing one (#512)
  • login --manual reported every failure as Cancelled, including a malformed URL and a state mismatch (#512)

Security

  • CI workflow steps pinned to exact commit SHAs (#519)
  • Response headers under our own namespace are blocked by prefix, so a header added later is blocked by default (#546)

[2.2.2] - 2026-06-03

Fixed

  • forecast reported working accounts as unavailable. token-exhausted has no natural expiry and nothing cleared it, so it lingered indefinitely — a successful request now clears it

[2.1.3] - 2026-05-01

Fixed

  • Wrapper-launched sessions repair session_index.jsonl after known thread-store write damage, serialised so concurrent sessions can't collide
  • App-bind status warnings resolve the active status path before printing guidance

[2.1.2] - 2026-04-30

Removed

  • The global codex bin, which collided with the official Codex npm, native, and Homebrew installs. codex-multi-auth and codex-multi-auth-codex are unchanged

[2.1.1] - 2026-04-29

Fixed

  • usage, account, budget, bridge, integrations, models, and monitor were falling through to the official Codex CLI instead of being handled locally
  • bridge token list --json returned invalid JSON on an empty store, and included token hashes

[2.1.0] - 2026-04-29

Added

  • Local-only usage ledger, account policy metadata, routing profiles, and budget guards
  • Model and account capability matrix, with policy evaluated before account selection
  • Optional loopback-only bridge with /health, /v1/models, and /v1/responses
  • rotation reset-rate-limits to clear stale runtime timers

2.0.1 - 2026-04-25

Changed

  • Runtime rotation is enabled by default. Opt out with rotation disable, codexRuntimeRotationProxy=false, or CODEX_MULTI_AUTH_RUNTIME_ROTATION_PROXY=0

Fixed

  • Install and update self-heal the packaged app bind and launcher routing

2.0.0 - 2026-04-25

Added

  • The runtime rotation proxy: failover across accounts mid-session, without restarting Codex. Loopback-only with an authenticated local client key
  • rotation enable|disable|status|bind-app|unbind-app, with app binding fully reversible

Fixed

  • The proxy left a stale content-encoding header on responses it had decoded

1.3.2 - 2026-04-24

Fixed

  • Interactive forwarded sessions were capturing output, which broke TTY use. TTY sessions now inherit terminal stdio; non-TTY runs still capture so model-fallback can inspect errors (#437)

1.3.1 - 2026-04-24

Added

  • GPT-5.5 and GPT-5.5 Pro as first-attempt models, with deterministic fallback for older runtimes and accounts without access (#435)

Fixed

  • Model fallback now only triggers on a real upstream unsupported-model or no-access response (#435)

1.3.0 - 2026-04-17

Added

  • why-selected explains which account rotation picked and why (#410)
  • verify --paths for path-safety inspection (#410)
  • A feature-flagged routing mutex for safer concurrent selection, off by default (#412)

Fixed

  • Post-audit hardening across OAuth URL redaction, redirect-URI handling, selector null handling, short-429 retry ordering, recovery writes, and malformed SSE warnings

[0.1.8] - 2026-03-11

Added

  • Codex CLI sync: target detection, import adapters, named backup wrappers, and sync orchestration. Codex CLI state stays mirror-only (#72)

Fixed

  • Cleared accounts could revive themselves when the initial delete partially failed (#71)

0.1.7 - 2026-03-03

Added

  • The first consolidated codex-multi-auth release, bringing runtime, TUI, account management, and docs into one package (#4, #14)
  • Rotating backup fallback, and automatic promotion of a real backup when primary storage holds fixture data (#29)
  • Hardened Windows command routing across codex.bat, codex.cmd, and codex.ps1 (#27)

Fixed

  • Codex CLI account switching stabilised, and storage identity fixed across worktree branch changes (#27, #28)

0.1.6 - 2026-03-03

Fixed

  • Updates could reset your accounts when storage was only reachable through recovery artifacts
  • Codex CLI sync wrote auth to a different profile directory than CODEX_HOME
  • Account switches now fail fast when the required Codex auth write doesn't complete

0.1.5 - 2026-03-03

Fixed

  • A Windows crash (UV_HANDLE_CLOSING) on wrapper shutdown, after the command had already completed

0.1.4 - 2026-03-03

Fixed

  • Stuck refresh lanes and duplicate refresh churn, via token normalisation and stale/timeout recovery
  • switch <index> keeps local selection deterministic and reports sync failures clearly

0.1.3 - 2026-03-03

Fixed

  • switch <index> reported failure when only the optional Codex host-state sync was unavailable — the local switch had succeeded

0.1.2 - 2026-03-03

Added

  • Rotating backup recovery across .bak, .bak.1, and .bak.2, staged then committed so a failure can't leave a partial history chain
  • Startup cleanup for orphaned staging files from interrupted writes, which could otherwise sit around holding token material

Fixed

  • Windows home resolution now tries USERPROFILE → HOME → HOMEDRIVE+HOMEPATH → homedir()

0.1.1 - 2026-03-01

Fixed

  • The OAuth callback is pinned to 127.0.0.1:1455 rather than localhost, which never arrived on machines resolving IPv6 first
  • A clean global install could fail at startup; the codex bin now lazy-loads the auth runtime

Changed

  • Account emails dedupe case-insensitively
  • Per-project storage is shared across linked Git worktrees
  • The settings hub was split into five focused modules; Q now cancels without saving

0.1.0 - 2026-02-27

Added

  • The first stable release: multi-account OAuth for Codex
  • The codex auth ... command family for login, switching, diagnostics, and reporting
  • Dashboard settings hub and backend reliability controls
  • Rotation and resilience modules for refresh, quota deferral, and failover

Legacy History

Pre-0.1.0 iteration history is archived in docs/releases/legacy-pre-0.1-history.md.


[2.26.1] - 2026-11-13

Native catalog discovery, login synchronization, and runtime rotation now handle recent Codex CLI changes safely. The release includes fixes from #765, #766, and #767. Full notes.

Fixed

  • Rotation provider arguments are inserted before the command separator, so forwarded native CLI arguments remain intact. (#765)
  • Native login changes are adopted by running state without losing account synchronization. (#766)
  • Catalog probes use the installed Codex version, recover from rejected probes, and retry shared requests without leaking stale version state. (#767)

[2.26.2] - 2026-11-18

GPT-6.2 model support and runtime-helper lifetime hardening are now available in the latest release. The release includes fixes from #769 and #770. Full notes.

Added

  • GPT-6.2 Sol and Astra model families are recognized across model resolution, request transformation, context limits, pricing, usage redaction, catalog fixtures, and the runtime wire path. (#769)

Fixed

  • The runtime helper keeps a verified owner alive for a bounded lease instead of allowing stale verification state to bypass the lifetime cap. (#770)
  • Owner verification waits for an in-flight probe after wake-up, keeping helper shutdown decisions deterministic across timing boundaries. (#770)