A registered gateway DCR client can now POST /token with
grant_type=urn:ietf:params:oauth:grant-type:token-exchange and an IdP
JWT as subject_token. The gateway proves the JWT the way its JWT auth
does, resolves the user and team, and answers with the proxy-API
credential and a refresh token, so a fresh laptop with only an IdP login
gets a gateway key without a browser round trip.
"/token" joins mcp_inference_routes so the default JWT team allowlist
reaches the exchange, and the JWT auth builder accepts any header
mapping so the request headers pass through unchanged.
The ssh parser rebuilt the clone url it was given, stripping a trailing .git and
appending one back. Git treats that suffix as optional, and the hosts whose clone
paths are not org/repo break when it is forced on, so an Azure DevOps v3 or a
CodeCommit v1/repos url registered through the form would no longer clone.
It also carried its own host pattern, which demanded an alphabetic final label and
so rejected internal hosts like gitlab.internal.k8s2 that the https path accepts.
Rewrite the scp form into an ssh:// url purely to validate it, reuse the https host
and credential checks through a shared isSafeHost, and store exactly what the user
typed. Only a url that survives the round trip unchanged is accepted, which is what
keeps traversal segments out of the feed, so the two ssh regexes, the dots-only
guard and the clone-url builders all collapse into one function.
Adds a persistent total_spend column to LiteLLM_VerificationToken and LiteLLM_DeletedVerificationToken, incremented in the same write as spend and left alone by budget resets. Surfaces it on /key/info, /key/list and the Admin UI Virtual Keys table and key detail view
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The four .parentElement reads in the new lifecycle tests pushed
testing-library/no-node-access to 712 against a 707 budget, failing
frontend-lint. The rows now carry data-testid="lifecycle-row" and the test
picks a row with within(), which keeps the assertion tied to the specific row
rather than the whole panel and takes the count back to 707.
#39050 changed RequestLifecycle from sorting every entry with
(a.start_time ?? 0) to filtering on isTimed, which drops any entry whose
start_time/end_time are null. That was the right call for the not_run entries
the PR introduced, but it also drops entries that DID run and simply carry no
timing, and those are pre-existing: add_standard_logging_guardrail_information_to_request_data
defaults start_time, end_time and duration to None, and the conduct guardrail
passes none of them. One such entry used to draw the whole four-row lifecycle
and now draws nothing, so an admin opening that log sees an empty
Request Lifecycle panel.
An entry now stays on the lifecycle when it is timed OR when it ran, so not_run
keeps the exclusion #39050 wanted and every other shape comes back. Offsets are
number | null and render as an em dash rather than a fabricated T+0ms, which is
what a null minus a null used to produce on the base. Entries without timing
sort after the timed ones and the base time comes from the timed entries, so
real offsets are unchanged.
The two new tests fail on the base component and pass here; #39050's own
not_run tests keep passing untouched, which is what makes this additive rather
than a revert.
Overlapping routing_groups persisted from the Admin UI raised inside
Router._init_routing_groups during the DB config reconcile, which skipped
loading SSO, guardrails and the other DB-backed settings while leaving the
proxy healthy. /config/update now returns 400 for overlapping models,
duplicate names, the reserved default name and unknown strategies before
writing, the Router builds every group selector before replacing its state
so a rejected update keeps the previous groups routing, and the proxy applies
routing_groups separately from the other router settings so an already
persisted invalid value is logged and skipped instead of aborting the
reconcile. The Admin UI modal blocks picking a model another group owns.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
A team can now carry a per-model budget map that every key on the team
inherits. A key's own model_max_budget entry for the same model takes
precedence, so it is gated on and billed to the key alone.
Backend: NewTeamRequest/UpdateTeamRequest accept model_max_budget (validated
like the key-level field, enterprise gated); the value is hydrated onto
UserAPIKeyAuth via the token view, TeamGrants and the carried budget state;
_check_team_model_budget enforces it in the centralized common checks; the
limiter meters spend under team_model_spend:<team>:<model>:<duration> and
skips the team counter when the key overrides; /team/update lets only a
proxy admin raise, re-window or drop a cap; /team/info exposes usage.
The Anthropic context-management compaction summary subrequest runs the
same team gate. Both fallback token-view SQL definitions project the column.
UI: team create and edit forms reuse the key-level ModelMaxBudgetEditor,
premium gated, sending {} to clear and omitting unchanged fields.
A key entry overrides the team cap only when it spend-gates the model
(non-negative max_budget); a row that only carries tpm/rpm limits or a
negative cap leaves the team cap in force.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Replace the username-global lockout with counters keyed by source address and by
source/username pair. Each has a fixed counting window (60s) and a separate
block TTL (300s). Blocks are soft: a correct password still signs in, wrong
passwords from a blocked key take one of 5 held slots per worker and are held
30s before a 429. Once a pair is blocked its failures stop counting against the
source. The source scope runs only when trusted_proxy_ranges is set, IPv6 is
grouped by /64, and per-source limits accept IP and CIDR overrides with
longest-prefix matching. Redis is authoritative through one Lua script per
failure, with bounded per-worker fallback when Redis raises.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
A loading flag only flips once the fetch effect runs, so the render right after a
date or filter change still reported the previous range as loaded and let an export
read its rows. Stamp the completed range on the hook and compare it during render
instead, the way the tiles already do.
Also stop the failure banner claiming a page loaded when the first request is what
failed, which left it reading 1/1.
Roll request_duration_ms of successful, non-internal requests into the
daily spend tables as total_response_time_ms plus timed_requests, expose
both through the daily activity endpoints, and derive the average in the
Usage -> Model Activity view of the Admin UI
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
