A 503 from POST /revoke means the proxy could not write the single-use record, so clearing the local record left a live refresh token nobody could revoke and a hint to retry with nothing left to retry. lite logout now keeps the record, exits 1, and asks to be run again shortly. A refused or unreachable revocation still clears the record and warns as before, and a re-login that replaces a record keeps its existing warning because the new record already stands
revoke_refresh_token discarded the single-use claim result, so a revocation that arrived while Redis was unreachable answered 200 and left the refresh token live. The token endpoint reported the same outage as invalid_grant "already used". The guard now reports first, replayed, or unavailable, and both endpoints answer 503 temporarily_unavailable for an outage (RFC 7009 section 2.2.1, RFC 6749 section 5.2), which the CLI surfaces as a one-line warning while keeping the key it has
The lite group resolves the stored key once for every command and renews a --pkce key on the way in. lite up then asked the token file again, so every start sent a second refresh to the proxy, and once the refresh token was burned the refusal printed twice. _ensure_fresh_login now reuses the key the group resolved when the group read it from the token file, and only re-reads the file after the interactive login it starts itself. Also covers print-token through the group with a renewing session in the tests
The cli group already resolves the stored key for the server it was pointed
at, so print-token re-ran the renewal and, when the refresh token had been
revoked, posted to /token twice and printed the reason twice. print-token now
reuses the group's result whenever the stored record was issued for that
server and no --api-key or LITELLM_PROXY_API_KEY took precedence, and only
resolves the key itself when invoked bare for a different server.
A PKCE credential whose renewal is refused (for example after lite logout ran
on another copy of it) used to fail lite auth print-token with the classic
'Token expired. Run lite login again' hint and no reason, while lite whoami
already named lite login --pkce. fresh_api_key now reports why a renewal
failed through a warn callback whenever no sibling rotation rescued it, the
CLI prints that reason on stderr, and the expiry hint names the command that
produced the credential. Both READMEs document the admin revocation semantics
and the Redis precondition for refresh single use on several workers.
Discovery checks that every endpoint sits on the proxy origin, but the CLI's
requests.Session followed redirects, and requests replays a POST body on
307 and 308, so a token or revocation endpoint answering with one of those
would have sent the code and verifier, or the refresh token, wherever
Location pointed. Every POST now goes out with allow_redirects=False and a
3xx answer fails the command with a message naming where it pointed
The consent page offers the team picker, but a form posted without a team
sealed a teamless grant and the token endpoint minted an unscoped
credential for a team member, escaping the team attribution classic lite
login always applies. The minter now refuses such a grant on redemption
and refresh alike; memberships whose team rows are gone still count as no
team so they cannot lock a user out
The discovery document is accepted only when its issuer is the --base-url the user
typed and every endpoint and the resource share that origin (RFC 8414 section 3.3),
so a tampered or redirected document can no longer point the code, verifier, or
refresh token at another host. After a failed refresh the re-read token record is
used only when it continues the same credential (same proxy, token endpoint, and
resource) and has not expired, so a concurrent login against a different proxy can
never hand this one its key
The proxy's OAuth authorization server (dynamic registration, PKCE S256,
loopback redirects, single-use codes, refresh rotation) gains a proxy-API
audience: /authorize?resource=<proxy origin> renders a consent page with
team selection and /token mints the same per-user credential lite login
mints, so a native CLI can sign a user in through the system browser and
call /v1/* with user and team attribution. Adds GET /.well-known/litellm-cli-auth
as the versioned discovery contract for non-Python clients, POST /revoke
(RFC 7009) for logout, and lite login --pkce, lite logout, and
lite auth print-token on the CLI side. Proxy-API grants only ever redirect
to a loopback address and the server never picks a team on the user's behalf.
Fixes#37332
PR #37579 read `text-X hover:text-X` on a shadcn Button as dead weight and
removed the hover half. On the ghost and outline variants it was not dead: both
carry their own `hover:text-foreground`, and the duplicate in the className was
the thing displacing it through tailwind-merge. Dropping it handed the hover
back to the variant, so the Remove button in a team's logging settings, the
chat storage banner's dismiss control, and the collapsed enterprise-usage rail
all lose their colour the moment you point at them.
Each of the three now carries a distinct hover value, following the alpha-step
idiom the rest of that migration used, which restores the colour and keeps
`local/no-noop-hover-variant` satisfied.
Every other hover utility that PR dropped sits on a plain element or a variant
with no competing `hover:text-`, so those stay as they are.
The UI unit test job narrows a pull request to `vitest related <changed
files>`. `related` maps a file to the tests that import it, so a file no
test imports maps to nothing, and `--passWithNoTests` turns that empty
selection into a green job. package.json, package-lock.json, the Vitest,
Tailwind and TypeScript configs and tests/setupTests.ts are all in that
category even though each of them can change the behaviour of every test
in the suite, so a dashboard dependency bump merged having run no unit
tests at all and only got real coverage later, from the full run on the
push to litellm_internal_staging.
Keep `related` for the common case where a pull request only touches
files under src/, and fall back to the full suite as soon as one changed
file sits outside it. The decision lives in
.github/scripts/select_ui_test_scope.sh so it can be tested on its own,
next to the existing classify_changes.sh gate.
* fix(ui): restore hover feedback and dark-mode variants lost in the token migration
PR #37576 mapped hardcoded Tailwind palette classes onto semantic tokens. Two-tone
hover pairs collapsed onto a single token, so 116 hover utilities across 49 files
became identical to their base class and produced no visible feedback, and in seven
files a dark: variant was dropped while its hardcoded light partner survived, leaving
those elements stuck light in dark mode.
Hover states now follow the alpha-step idiom the shadcn primitives already use
(hover:bg-primary/80, hover:bg-success/20): a duplicated hover:text-X or hover:bg-X
becomes /80, hover:border-border becomes hover:border-ring, and a duplicate is
dropped where another hover utility on the element already carries the change. One
transition-colors that no longer animated anything is removed.
For the dark-mode gaps, indigo maps onto info and amber onto warning. There is no
purple token in globals.css, so the purple sites keep their palette classes and get
their dark: partner back.
* fix(ui): add an eslint rule that fails a hover: utility identical to its base
The token migration collapsed two-tone hover pairs by hand, so nothing catches
the next one. `local/no-noop-hover-variant` reads every string literal and
template chunk and errors when a `hover:X` sits alongside a bare `X`, which is
exactly the shape that renders no hover feedback. It ships at error with no
suppression baseline, so the eleven sites that already carried a dead hover
before the migration are fixed here too.
The rule reads one class string at a time, so a base class supplied by a
different ternary branch than its hover partner is left alone: a selected row
whose resting colour already matches its hover colour is deliberate, not a bug.
* fix(ptu): hand the prune a plain delete filter the query builder can serialise
The bounded sweep built its predicate as a read-only mapping view, which the query
builder refuses to serialise, so the nightly job raised as soon as a config-declared
deployment was priced. The charges were already written by then, which is why the run
looked like it had produced its rows.
The in-memory table these tests run against accepts any mapping, so only a live run
caught it. A predicate builder now returns a plain dict and is asserted as one, and the
catch-up pass has a test covering a config-declared reservation.
* refactor(ptu): build the prune predicate in one shot
Both filter shapes are known upfront, so the bounded one is constructed
directly rather than by mutating a value already declared Final.
The catch-up test took two independent clock reads, which disagree across
UTC midnight; it now derives both the reservation start and the expected
last charged day from a single read, matching the three sibling tests.
* refactor(ui): map hardcoded Tailwind palette classes onto semantic tokens
The dashboard painted itself with literal palette utilities (text-gray-500,
bg-blue-50, border-red-200) that resolve to one fixed color regardless of
theme, so the shadcn token layer and its .dark block could never take effect.
A codemod (scripts/codemod-color-tokens.mjs) rewrites 3,232 of those across
254 files onto the existing token scale: neutrals become foreground /
muted-foreground / muted / border / card, and red, green, amber and blue
collapse into destructive, success, warning and info, with the pale -50 to
-300 tints expressed as opacity modifiers on the same token. Hover and focus
variants map to accent so they lift rather than recess. 210 now-redundant
dark: variants are dropped since the tokens carry both modes.
The .dark palette is retuned to a neutral gray ramp with the sidebar recessed
below the content canvas, replacing the blue-tinted shadcn default where the
sidebar read as a full-height card floating on a near-black page.
Nothing sets the .dark class yet, so light mode is unchanged and dark mode
stays inert until a theme toggle lands.
* chore(ui): drop the one-shot color-token codemod script
Nothing in the dashboard renders antd any more, so the package and the
scaffolding around it can go. This removes `antd` and
`@ant-design/cssinjs` from package.json, deletes the global StyleProvider
the root layout wrapped every page in, drops the `antd` cascade layer and
the z-index override that lifted Base UI popups over an antd Modal, and
retires the lint rules that policed antd imports and antd class selectors
in tests.
Fifteen test files still carried `vi.mock("antd", ...)` factories for
components that stopped importing antd during the migration. They were
inert, and they resolve the real module, so they would have broken the
moment the package left node_modules.
The compatibility shims keep their behaviour and lose the antd name:
`antdRules`/`antdRequired` become `validatorRules`/`requiredRule`,
`isAntdUrl` becomes `isValidUrl`, and `ABOVE_ANTD_MODAL` becomes
`NESTED_DIALOG_LAYER`. Comments that explain why a contract looks the way
it does still name antd, because that history is the reason.
* feat(ptu): accrue flat cost for PTU deployments declared in config.yaml
The flat-cost rollup reads deployments from LiteLLM_ProxyModelTable, and config.yaml
models never reach that table by design, so a PTU deployment declared there accrued no
flat cost at all while still billing its traffic per token. The provider bills the
reservation whichever file declared it.
The rollup now also reads the deployments the router holds that no database row owns,
identified by db_model, skipping the per-request credential clones that carry
original_model_id and reuse their source's PTU config under a fresh id. Registering such
a deployment zeroes its pricing, since reserved capacity already pays for the traffic it
serves, and leaving a rate unset falls back to the public cost map, which makes the double
charge the default rather than an opt-in.
The rules both halves apply now live in one module. The rollup's test for what it will
charge and the router's test for what to zero have to agree, or a deployment one accepts
and the other declines serves its traffic for free. That module also owns the fields the
write endpoints already zero, so the two paths cannot drift: tiered_pricing is emptied
rather than zeroed because its tiers outrank the rates beside them, the search context
table is written zeroed because an absent one means the provider default, and any further
rate the deployment itself declares is zeroed alongside the standing set.
The prune is bounded to the deployments a run scanned, but only for a run that priced a
config-declared deployment. Deciding a row is garbage on staleness alone stays correct
while every run derives its charges from the same table, so a database-only run sweeps
exactly as it did before; once one host's charges come from a file the others cannot read,
a row it never considered is not evidence of anything.
Behaviour change worth calling out: a zeroed deployment sorts ahead of an unpriced sibling
in QualityRouter's cost tiebreak, where an unset rate previously sorted last. Reserved
capacity really is the cheaper choice, but the ordering moves.
* refactor(ptu): drop a Final rebind and two redundant isinstance guards
The basedpyright budget rejected reassigning a Final in the datetime coercion and
two isinstance calls the router entry's own type already guarantees. Filtering the
built records rather than the raw entries removes both guards and leaves
_router_deployment as the single validator.
* fix(ui): render optional array and object MCP tool parameters as JSON inputs
A Python signature like `tags: list[str] | None = None` serialises to
`{"anyOf": [{"type": "array"}, {"type": "null"}]}` with no top-level
`type`, so the tool test panel's control dispatch fell through to the
generic text input. Whatever the user typed was sent verbatim, and the
server rejected it as the wrong type.
Resolve a property to its single non-null union member before choosing a
control, validating, seeding defaults, and coercing the submitted value,
so all four agree and an optional array or object gets the same JSON
textarea a required one already got.
* fix(ui): keep a null-defaulted optional MCP parameter out of the call
A parameter declared `list[str] | None = None` carries `default: null`,
which means the caller should send nothing. Seeding its editor with an
empty container made the field non-blank, so an untouched parameter was
submitted as `[]` or `{}` instead of being omitted.
Treat an explicit null default as "no value" everywhere it is read: the
editor starts blank and shows its placeholder, and the submitted
arguments leave the key out entirely.
Teams.tsx and Teams.test.tsx both conflicted with staging's antd -> shadcn
migration of the team create form.
Teams.tsx: took staging's rewritten import block and dropped `theme` from the
antd import, since this branch replaced `<Content style={{ padding: token... }}>`
with the Tailwind inset. Dropped both `const { Text } = Typography` (staging
removed its last use) and `const { token } = theme.useToken()` (this branch
removed its last use).
Teams.test.tsx: took this branch's PageHeader-shaped assertions over staging's
older tab-bar lookup, and restored the `within` import that staging had dropped.
Removed the `toHaveClass` snapshot of the antd tab-bar Tailwind classes and the
`.closest(".ant-tabs")` lookup: staging added local/no-antd-class-selectors as a
zero-violation error rule, and those assertions are inert in jsdom anyway. Every
behavioural assertion in that test is unchanged.
KeyValueInput and QueryParamInput each seeded a private copy of their rows
from the value prop with a one-time useState initializer. The antd form they
were written for hid that: rc-field-form bumps an internal resetCount key on
resetFields, which remounts a Field's children, so the private copy was thrown
away on every reset. react-hook-form's reset does not remount, and the modal is
hidden rather than unmounted, so after Cancel the rows stayed on screen holding
the old values while the form value went back to empty.
The visible cost was a blocked create flow. A leaked header row made the modal
look configured, but the form value behind it was gone, so submitting a fresh
path and target was refused with "Please configure the headers" and no request
was sent. Typing one character into the leaked row put a value back and the
submit went through, which is not something a user can guess.
Both inputs are now controlled off the value prop, which is an array of pairs
rather than a record. A record cannot represent a row whose name is still empty,
which is the reason the private copy existed: two blank rows collapse into one
and a half-typed row disappears as it is typed. With pairs the field value is
the editable shape, the second source of truth is gone, and a form reset clears
the rows like every other field. add_pass_through converts to a record at submit,
so the request payload is unchanged.
Headers now require at least one row with a non-empty name. Previously that was
enforced by accident, because adding a row did not notify the form at all.
PR #37550 taught the backend unit-test shards to read the pull request's own
file list, but four required jobs were never wired to that gate and ran in full
on every pull request regardless of what it touched. A UI-only pull request
still paid roughly 17 runner-minutes of Python work it could not have affected,
and a backend-only one still installed and built the dashboard.
Lint and the MCP suite now take the existing backend decision. The dashboard
build and unit tests take a new ui decision, which tracks ui/ rather than
reusing client: client deliberately runs whenever the backend changes, because
it gates CircleCI's end-to-end jobs that drive a real proxy, while the build and
the unit tests cannot see the backend at all. CI config counts as ui-relevant
too, so a pull request that rewrites the dashboard workflows still exercises
them instead of shipping unvalidated.
The gate stays inside the job rather than moving to on.paths or to a job-level
condition on the shard callers. A workflow filtered out by on.paths never starts
and never reports, so a required check waits forever, and a skipped caller job
publishes its own name instead of the nested "<shard> / Run tests" the ruleset
requires. Both were measured before settling on this shape.
Three setup steps in the shard base and in the documentation job also leaked
past the gate, so a skipped shard still spent about twelve seconds installing uv
and restoring its cache. They now carry the same condition, and the documentation
job stops cloning litellm-docs when it has nothing to validate.
* refactor(ui): migrate the model and router settings pages off antd
Converts the add model flow, credential panels, model settings and router
settings onto the shadcn primitives, moves the mapping table onto the
shared DataTable, and drops the dead uploadProps prop chain that only
existed to carry antd's UploadProps type.
* fix(ui): split comma-separated custom technical keywords into one term each
* refactor(ui): migrate the MCP servers pages off antd
Converts the MCP server create, edit, connect and permission screens plus
the MCP tools and selector components onto the shadcn primitives, and
rewrites the test helpers that drove antd's select and collapse DOM.
* fix(ui): finish the MCP servers antd migration so the shared field rules have one contract
mcpFieldRules and MCPPermissionManagement were already flipped to the shadcn
prop shape, but CreateMCPServer and UserEnvVarsModal were still rendering antd,
so the create modal spread onValueChange onto an antd Select that ignores it and
passed searchValue props that no longer exist. Convert both off antd, drop the
searchValue plumbing the MultiSelect now owns, and normalise tag values before
the tag list renders so a delimited or empty string cannot crash it.
Rewrite testUtils.selectOption to drive the shadcn listbox instead of
.ant-select, expand the collapsed permission panel before querying its switches,
and assert the dismiss case after a reopen now that Dialog unmounts closed
content.
* fix(ui): split multi-tag entries the MCP tag inputs commit as one value
The tag input hands back whatever the admin typed as a single custom value, so
"read,write" was stored verbatim and reached the backend as one malformed scope.
tagsControl already split delimited values on the way in; run the same
normalisation on the way out and dedupe, so both directions agree.
* fix(ui): stop splitting tag entries that are not scope lists
The previous commit split every tag field on whitespace and commas, but only a
scope list is delimited. A stdio arg or an access description item may contain
both characters as part of the value, so splitting them changed the argv the
process receives. Keep those entries verbatim and move the splitting behind
scopesControl, which the OAuth, token exchange and ID-JAG scope fields use.
* fix(ui): split tag entries on comma only, matching the antd token separator
Every tag field here was an antd Select carrying a comma token separator, so a
comma committed a tag and nothing else did. Splitting on whitespace as well
broke stdio args, and splitting neither left comma-separated extra headers and
access groups stored as one malformed value. Apply the comma rule in both
directions, trim each entry, and drop the scope-specific helper the previous
commit added, since the backend types scopes as a list rather than the
space-delimited string that helper assumed.
* fix(ui): stop rewriting stored tag values that an admin never edited
Stdio args are process argv, so a comma inside one argument and a
deliberately repeated flag both have to survive a round trip through the
edit modal. Two places were rewriting them. tagsControl split and deduped
the value it read back from the server, and MultiSelect re-split every
already-committed chip on each change rather than only the entry just
typed. Both now leave settled values alone, which keeps the antd token
separator applying to typing and nothing else.
* refactor(ui): migrate shared primitives and common components off antd
Adds the success variant to the shared Alert plus success, warning and
info variants to Badge, introduces UtcDateTimeInput to replace antd's
DatePicker, and converts the common components and key/team helpers onto
the shadcn primitives.
* fix(ui): keep MultiSelect and budget input faithful to their antd behaviour
Restore the clear-all control MultiSelect lost, split comma-separated
custom entries into one value per token, and stop rounding the budget
input on every keystroke so a fractional amount survives typing.
* test(ui): drive the access group picker through the migrated MultiSelect
AccessGroupSelector no longer renders an antd Select, so the placeholder
is an input label rather than a text node and the popup inerts the page
until it closes.
The field replaces the standard RPM/TPM checks for batch submissions, so a
key holder or team admin writing it could pick their own batch quota.
Mirrors the output-token-estimate admin gate: change-based, so resending
the stored value stays allowed, and enforced on key generate, update, bulk
team-key update, regenerate, and team new/update.
The dashboard drew its icons from two libraries at once: lucide-react,
which shadcn/ui ships with, and @ant-design/icons, left over from antd.
This moves the last 39 files onto lucide and drops the dependency, so
the icon set matches the component library everywhere.
antd icons sized themselves from the inherited font-size and rendered as
role="img" with an aria-label, neither of which a lucide svg does, so the
swap carries explicit size classes and gives the two icon-only plugin
buttons real accessible names.
`ProxyConfig.get_config()` walked the parsed config and replaced every
`os.environ/<KEY>` string with `get_secret(value)` before anything initialized
the secret manager, so a key held only by the manager resolved to `None` and
that `None` was written back into the config. The later fallback in
`load_config` could not recover it, because the key now existed with a `None`
value.
Hoist the initialization into `get_config()`, ahead of the resolution pass, so
every entrypoint gets it: the CLI already did this itself, but the microservice
entrypoints (`gateway/main.py`, `backend/main.py`) uvicorn the app directly and
bypass the CLI. `load_config`'s own call is now redundant and is dropped, so
startup builds the manager once instead of building one and discarding it.
`get_config()` also runs on management-endpoint request paths, so this returns
early once a manager exists rather than rebuilding the client per request.
Also warn when a reference the manager would have been asked for resolves to
`None`. The reporter had no log line at all to work from. `get_secret` only
reaches the manager when reads are enabled and the name is in `hosted_keys`, so
`secret_manager_would_be_consulted` mirrors that gate and keeps the warning off
env-only references, which are expected rather than an error.
A Redis over-limit verdict now survives a failing rollback DECRBY instead of
escaping into the in-memory fallback and granting tokens the counter already
rejected; the unrolled increments expire with the TTL. pop_reservation now
falls through to the local record when the Redis pop succeeds but finds
nothing, so a reservation saved in memory after a transient Redis save
failure still refunds on cancel or completion.
detect-backend-changes diffed the event payload's base.sha against the
checked-out ref. Those are two different points in time: actions/checkout
resolves refs/pull/N/merge, and GitHub recomputes that ref whenever the base
branch advances, so the diff picked up whatever landed on staging between the
event firing and the job starting. On a recent UI-only pull request three
backend commits from staging were attributed to the branch, and every backend
shard ran in full
Ask the API which files the pull request touches instead. That is the same set
the Files changed tab shows, and it is immune to either endpoint moving. The
shell body moves into .github/scripts/detect_backend_changes.sh so it can be
exercised directly, and the fail-open paths now also cover an API failure, a
file list past the API's 3000-entry listing ceiling, and a classifier that
prints something unexpected
Every dynamic tracer-provider build called Resource.create, which scans the entry
points of every installed distribution, roughly 3ms and 200 file opens. The dynamic
providers reach it from the async logging path, which runs on the event loop serving
requests, so past the provider cache bound every request paid it and delayed the
requests in flight alongside it
The value derives only from the logger's config and process environment, so it is
built once per logger and reused. This logger's own init-time providers share it,
which also removes redundant startup builds. ArizeLogger overrides _init_tracing and
still builds its own, so it keeps one extra build
Refs LIT-5437
