Release v3.8.40

v3.8.40 cycle integration → main. All test gates green (Unit/Integration/Coverage/Node-compat/Quality-Ratchet). The only red check, 'PR Test Policy', is the test-masking heuristic firing on the cumulative ~57-commit release diff (legitimate assert consolidations already reviewed per-PR — Gemini CLI removal #5246, retired GPT models #5280, provider catalog refreshes); overridden with --admin per the documented release-PR convention. CodeQL/SonarQube advisory scans non-blocking; #5278's code already passed CodeQL on main. Homologated on VPS 192.168.0.15 (v3.8.40 healthy).

Author: diegosouzapw

.env.example

@@ -413,7 +413,6 @@ NEXT_PUBLIC_CLOUD_URL=
 # debugging or when routing through a corporate mirror. Used by:
 # open-sse/services/usage.ts.
 #OMNIROUTE_CROF_USAGE_URL=https://crof.ai/usage_api/
-#OMNIROUTE_GEMINI_CLI_USAGE_URL=https://cloudcode-pa.googleapis.com/v1internal:loadCodeAssist
 #OMNIROUTE_CODEWHISPERER_BASE_URL=https://codewhisperer.us-east-1.amazonaws.com
 #OMNIROUTE_OPENCODE_QUOTA_URL=https://opencode.ai/zen/go/v1/quota
 #OMNIROUTE_OPENCODE_GO_QUOTA_URL=https://api.z.ai/api/monitor/usage/quota/limit
@@ -721,7 +720,7 @@ CODEX_OAUTH_CLIENT_ID=app_EMoamEEZ73f0CkXaXp7hrann
 # Used by: open-sse/executors/theoldllm.ts. Default: 30000 (30s).
 # THEOLDLLM_NAV_TIMEOUT_MS=30000
 
-# ── Gemini / Gemini CLI / Antigravity / Windsurf (all Google-based) ──
+# ── Gemini / Antigravity / Windsurf (all Google-based) ──
 # These providers ship public OAuth client_id/secret values (or Firebase Web
 # keys) embedded in their public CLIs/binaries. Defaults are baked into the
 # code via open-sse/utils/publicCreds.ts — leave the env vars unset to use
@@ -730,8 +729,6 @@ CODEX_OAUTH_CLIENT_ID=app_EMoamEEZ73f0CkXaXp7hrann
 #
 # GEMINI_OAUTH_CLIENT_ID=
 # GEMINI_OAUTH_CLIENT_SECRET=
-# GEMINI_CLI_OAUTH_CLIENT_ID=
-# GEMINI_CLI_OAUTH_CLIENT_SECRET=
 # ANTIGRAVITY_OAUTH_CLIENT_ID=
 # ANTIGRAVITY_OAUTH_CLIENT_SECRET=
 # WINDSURF_FIREBASE_API_KEY=
@@ -819,7 +816,7 @@ GITHUB_OAUTH_CLIENT_ID=Iv1.b507a08c87ecfe98
 # VISION_BRIDGE_API_KEY=
 
 # ─────────────────────────────────────────────────────────────────────────────
-# ⚠️  GOOGLE OAUTH (Antigravity, Gemini CLI) & OTHER PROVIDERS — REMOTE SERVERS
+# ⚠️  GOOGLE OAUTH (Antigravity) & OTHER PROVIDERS — REMOTE SERVERS
 # ─────────────────────────────────────────────────────────────────────────────
 # The default Client IDs above ONLY work when OmniRoute runs on localhost.
 # For remote/VPS hosting (including Docker containers on remote servers):
@@ -847,7 +844,7 @@ GITHUB_OAUTH_CLIENT_ID=Iv1.b507a08c87ecfe98
 # Used by: open-sse/executors/base.ts — buildHeaders() dynamic lookup.
 # Update these when providers release new CLI versions to avoid blocks.
 
-CLAUDE_USER_AGENT="claude-cli/2.1.187 (external, cli)"
+CLAUDE_USER_AGENT="claude-cli/2.1.195 (external, cli)"
 
 # Disable the deterministic tool-name cloak applied on both Anthropic-bound paths
 # (executors/base.ts native OAuth + executors/cliproxyapi.ts CLIProxyAPI) —
@@ -857,7 +854,7 @@ CLAUDE_USER_AGENT="claude-cli/2.1.187 (external, cli)"
 # forward the original names verbatim (debugging only).
 # CLAUDE_DISABLE_TOOL_NAME_CLOAK=false
 CODEX_USER_AGENT="codex-cli/0.142.0 (Windows 10.0.26200; x64)"
-GITHUB_USER_AGENT="GitHubCopilotChat/0.45.1"
+GITHUB_USER_AGENT="GitHubCopilotChat/0.54.0"
 ANTIGRAVITY_USER_AGENT="antigravity/2.0.1 linux/arm64 google-api-nodejs-client/10.3.0"
 KIRO_USER_AGENT="AWS-SDK-JS/3.0.0 kiro-ide/1.0.0"
 # KIRO_VERIFY_FULL_CRC=false           # opt-in: full per-frame message CRC validation on the Kiro event stream (debug corrupted streams; prelude CRC + TLS already protect framing)
@@ -872,9 +869,8 @@ KIRO_USER_AGENT="AWS-SDK-JS/3.0.0 kiro-ide/1.0.0"
 # Used by: open-sse/executors/kiro.ts
 # KIRO_VERIFY_FULL_CRC=false
 QODER_USER_AGENT="Qoder-Cli"
-QWEN_USER_AGENT="QwenCode/0.15.11 (linux; x64)"
+QWEN_USER_AGENT="QwenCode/0.19.3 (linux; x64)"
 CURSOR_USER_AGENT="Cursor/3.4"
-GEMINI_CLI_USER_AGENT="google-api-nodejs-client/10.3.0"
 
 # Override Codex client version sent in headers independently of the
 # CODEX_USER_AGENT string. Used by: open-sse/config/codexClient.ts.
@@ -1362,6 +1358,11 @@ APP_LOG_TO_FILE=true
 # Health check result cache TTL (ms).  Default: 30000 (30s)
 # PROXY_HEALTH_CACHE_TTL_MS=30000
 
+# Unhealthy health check result cache TTL (ms). Default: 2000 (2s)
+# Keeps transient fast-fail timeouts from poisoning a proxy for the full
+# healthy-result cache window under high concurrency.
+# PROXY_HEALTH_UNHEALTHY_CACHE_TTL_MS=2000
+
 # Allow OAuth and provider validation flows to bypass a pinned proxy and connect
 # directly when proxy reachability pre-checks fail. Default: false.
 # Also configurable from Dashboard > Settings > Feature Flags.

.github/workflows/docker-publish.yml

@@ -98,11 +98,13 @@ jobs:
             PROMOTE="${PROMOTE_INPUT:-false}"
           else
             git fetch --tags --quiet || true
-            HIGHEST=$(git tag -l 'v[0-9]*' | sed 's/^v//' | grep -vE -- '-(rc|alpha|beta|pre|next)' | sort -V | tail -1 || echo "")
-            if [ -n "$HIGHEST" ] && [ "$VERSION" = "$HIGHEST" ]; then
-              PROMOTE="true"
-            else
-              echo "Version $VERSION is not the highest semver tag (highest=${HIGHEST:-<none>}). Not promoting :latest."
+            # Decide via the extracted helper, which folds VERSION into the
+            # candidate set so the result is independent of git-tag sync timing
+            # on `release` events (#5301). Without that, the freshly-created tag
+            # is often not yet visible here and :latest stays a release behind.
+            PROMOTE=$(git tag -l 'v[0-9]*' | bash scripts/ci/should-promote-latest.sh "$VERSION")
+            if [ "$PROMOTE" != "true" ]; then
+              echo "Version $VERSION is not the highest stable semver. Not promoting :latest."
             fi
           fi
           echo "promote_latest=$PROMOTE" >> "$GITHUB_OUTPUT"

.github/workflows/wiki-sync.yml

@@ -9,7 +9,7 @@ name: Wiki Sync
 # It does NOT overwrite existing wiki pages by default: several docs sources still carry
 # stale counts (e.g. ARCHITECTURE.md says "177 providers" while the wiki cover is 226),
 # so blind overwrite would regress the wiki. Full content parity (--update-existing) is
-# gated on regenerating those sources — see docs/ops/DOCUMENTATION_AUDIT_REPORT.md.
+# gated on regenerating those sources first.
 
 on:
   push:

.mcp.json.example

@@ -196,7 +196,7 @@ Every field is optional. Defaults mirror v0.1.0 behaviour so existing `opencode.
 | `combos`              | `boolean` | `true`  | Discover `/api/combos` and surface them as pseudo-models with LCD capabilities. Combos are keyed under the `combo/<slug>` namespace and labelled `Combo: <name>` in the model picker so they're distinguishable from raw provider/model pairs.                                                                                                                                                                                                                                                                                                                               |
 | `enrichment`          | `boolean` | `true`  | Pull display names from `/api/pricing/models` AND per-million-token pricing (`input`, `output`, `cached` → `cacheRead`, `cache_creation` → `cacheWrite`) from `/api/pricing`, then overlay both onto the live catalog (so the UI shows `Claude 4.7 Opus` with `cost.input: 5`, `cost.output: 25` instead of raw IDs and zeroed cost).                                                                                                                                                                                                                                        |
 | `compressionMetadata` | `boolean` | `false` | Pull `/api/context/combos` so combo names get tagged with their compression pipeline, e.g. `Combo: claude-primary [rtk🟡 → caveman🟠]`. Intensity tokens render as traffic-light emoji (🟢 lite/minimal · 🟡 standard · 🟠 aggressive/full · 🔴 ultra) so the picker advertises "how compressed" each combo is at a glance.                                                                                                                                                                                                                                                  |
-| `providerTag`         | `boolean` | `true`  | Prepend a short upstream-provider label to the enriched display name with `" - "` separator, so `cc/claude-opus-4-7 → Claude - Claude Opus 4.7` differs visibly from `kr/claude-opus-4-7 → Kiro - Claude Opus 4.7` in the OC TUI model picker. Label resolution: use `/api/pricing/models[<alias>].name` verbatim when ≤8 chars (e.g. `Claude`, `Kiro`, `Codex`, `Qwen`), otherwise fall back to `UPPER(alias)` (e.g. `GitHub Models` → `GHM`, `Gemini-cli` → `GEMINI-CLI`). Idempotent. Combos intentionally skipped (the `Combo: ` prefix already conveys multi-upstream). |
+| `providerTag`         | `boolean` | `true`  | Prepend a short upstream-provider label to the enriched display name with `" - "` separator, so `cc/claude-opus-4-7 → Claude - Claude Opus 4.7` differs visibly from `kr/claude-opus-4-7 → Kiro - Claude Opus 4.7` in the OC TUI model picker. Label resolution: use `/api/pricing/models[<alias>].name` verbatim when ≤8 chars (e.g. `Claude`, `Kiro`, `Codex`, `Qwen`), otherwise fall back to `UPPER(alias)` (e.g. `GitHub Models` → `GHM`, `Gemini` → `GEMINI`). Idempotent. Combos intentionally skipped (the `Combo: ` prefix already conveys multi-upstream). |
 | `usableOnly`          | `boolean` | `false` | Read `/api/providers` and filter the catalog to providers that have at least one connection with `isActive: true` AND `testStatus: 'active'`. Subtract-filter semantics: providers unknown to BOTH the pricing-models catalog AND the connection table pass through (so synthetic prefixes like `agentrouter/*` survive). On fetch failure the filter is disabled for the refresh — never hides the whole catalog.                                                                                                                                                           |
 | `diskCache`           | `boolean` | `true`  | Persist the last successful `/v1/models` + `/api/combos` + enrichment + connections + compression snapshot to `${OPENCODE_DATA_DIR ?? ~/.local/share/opencode}/plugins/omniroute-<providerId>.json`. On a subsequent cold start where `/v1/models` throws (network down / IP whitelist drop / 5xx) the static block hydrates from the snapshot so OC's model picker survives offline. Soft-fail on read/write — never blocks publishing.                                                                                                                                     |
 | `geminiSanitization`  | `boolean` | `true`  | Strip `$schema`/`$ref`/`additionalProperties` from tool params when the model id matches `gemini`                                                                                                                                                                                                                                                                                                                                                                                                                                                                            |

@omniroute/opencode-plugin/README.md

@@ -1225,15 +1225,15 @@ export interface OmniRouteEnrichmentEntry {
     cacheWrite?: number;
   };
   /**
-   * Provider alias prefix seen in `/v1/models` ids (e.g. `cc`, `gemini-cli`).
+   * Provider alias prefix seen in `/v1/models` ids (e.g. `cc`, `gemini`).
    * Populated by `defaultOmniRouteEnrichmentFetcher` from
    * `/api/pricing/models` keys. Drives the `usableOnly` alias↔canonical
    * resolution.
    */
   providerAlias?: string;
   /**
    * Canonical provider id used by `/api/providers` connections (e.g.
-   * `claude`, `gemini-cli`, `kiro`). Populated from the per-provider
+   * `claude`, `gemini`, `kiro`). Populated from the per-provider
    * `entry.id` field inside `/api/pricing/models`.
    */
   providerCanonical?: string;
@@ -2046,7 +2046,7 @@ export function formatCompressionPipeline(pipeline: OmniRouteCompressionStep[]):
 export interface OmniRouteProviderConnection {
   /** Connection UUID. */
   id: string;
-  /** Canonical provider id, e.g. `claude`, `gemini-cli`, `kiro`. Matches `entry.id` in `/api/pricing/models`. */
+  /** Canonical provider id, e.g. `claude`, `gemini`, `kiro`. Matches `entry.id` in `/api/pricing/models`. */
   provider: string;
   /** Connection auth flavor, e.g. `apikey`, `oauth`, `cookie`. */
   authType?: string;
@@ -2125,7 +2125,7 @@ export const defaultOmniRouteProvidersFetcher: OmniRouteProvidersFetcher = async
  * walk only the namespaced keys to derive the alias↔canonical mapping).
  *
  * Returns:
- *   - `aliases`: set of alias prefixes safe to keep (e.g. `cc`, `gemini-cli`).
+ *   - `aliases`: set of alias prefixes safe to keep (e.g. `cc`, `gemini`).
  *   - `canonicals`: set of canonical provider ids (e.g. `claude`, `kiro`).
  *
  * Callers should treat membership in EITHER set as "usable" — raw model
@@ -2174,7 +2174,7 @@ export function usableProviderAliasSet(
   }
   // Always include every usable canonical as an alias too — handles the
   // common case where `/v1/models` ids use the canonical id directly
-  // (e.g. `gemini-cli/gemini-1.5-pro`).
+  // (e.g. `gemini/gemini-1.5-pro`).
   for (const canonical of usableCanonicals) aliases.add(canonical);
   return { aliases, canonicals: usableCanonicals, knownAliases };
 }
@@ -3174,7 +3174,6 @@ export function sanitizeGeminiToolSchemas(payload: unknown): unknown {
  *     `gemini-2.5-flash`, etc.)
  *   - `models/gemini-…` (Google Generative AI canonical id form)
  *   - `google-vertex/gemini-…` (OpenCode + AI-SDK Vertex routing prefix)
- *   - `gemini-cli/…` (real OmniRoute alias surfaced on b35 prod `/v1/models`)
  *
  * Liberal by design: a false positive (cleaning a payload that didn't
  * need cleaning) costs only a structuredClone + one walk; a false negative

@omniroute/opencode-plugin/src/index.ts

@@ -1311,9 +1311,9 @@ test("config: providerTag (default-on) prepends '<provider> - ' to enriched raw-
         "gemini-3-flash",
         {
           name: "Gemini 3 Flash",
-          providerAlias: "gemini-cli",
-          providerCanonical: "gemini-cli",
-          providerDisplayName: "Gemini-cli",
+          providerAlias: "gemini",
+          providerCanonical: "gemini",
+          providerDisplayName: "Gemini",
         },
       ],
     ])
@@ -1335,10 +1335,7 @@ test("config: providerTag (default-on) prepends '<provider> - ' to enriched raw-
     entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "Claude - Claude Sonnet 4.6"
   );
-  assert.equal(
-    entry.models["opencode-omniroute/gemini-3-flash"].name,
-    "Gemini-cli - Gemini 3 Flash"
-  );
+  assert.equal(entry.models["opencode-omniroute/gemini-3-flash"].name, "Gemini - Gemini 3 Flash");
   // Combos stay untouched — `Combo: ` prefix already conveys multi-upstream.
   assert.equal(entry.models["opencode-omniroute/claude-tier"].name, "Claude Tier");
 });

@omniroute/opencode-plugin/tests/config-shim.test.ts

@@ -212,8 +212,8 @@ test("shouldSanitizeForGemini: google-vertex/gemini-1.5-flash → true", () => {
   assert.equal(shouldSanitizeForGemini({ model: "google-vertex/gemini-1.5-flash" }), true);
 });
 
-test("shouldSanitizeForGemini: gemini-cli/gemini-2.5-pro → true (real OmniRoute alias)", () => {
-  assert.equal(shouldSanitizeForGemini({ model: "gemini-cli/gemini-2.5-pro" }), true);
+test("shouldSanitizeForGemini: gemini/gemini-2.5-pro → true", () => {
+  assert.equal(shouldSanitizeForGemini({ model: "gemini/gemini-2.5-pro" }), true);
 });
 
 test("shouldSanitizeForGemini: claude-sonnet-4 → false", () => {