@oh-my-pi/pi-catalog 16.1.3 → 16.1.5

package/CHANGELOG.md

@@ -2,6 +2,12 @@
 
 ## [Unreleased]
 
+## [16.1.4] - 2026-06-19
+
+### Fixed
+
+- Fixed Claude 4.6 routing on the `google-antigravity` (and `google-gemini-cli`) Cloud Code Assist providers, whose backend exposes the models asymmetrically: `claude-sonnet-4-6` has no `-thinking` twin and `claude-opus-4-6` has only the `-thinking` twin. The shared `thinkingPair` family was routing thinking efforts on `claude-sonnet-4-6` to a non-existent `claude-sonnet-4-6-thinking` wire id (404 `Requested entity was not found`); replaced both 4.6 entries with bespoke single-wire families that declare the dead ids as `retiredMembers` so `reconcileRetiredRouting` re-points stale bundled-catalog and SQLite-cache rows away from the 404 wire id. Refreshed the bundled `models.json` Sonnet 4.6 entry whose stored `effortRouting` still targeted the dead `-thinking` id. Added `claude-sonnet-4-6` and `claude-opus-4-6-thinking` entries to `ANTIGRAVITY_MODEL_WIRE_PROFILES` capped at the backend's 64000-output-token limit (over-cap requests 400'd with `Request contains an invalid argument`); `modelEnum` is now optional on `AntigravityModelWireProfile` since the Claude wire ids are accepted without a captured `labels.model_enum`. ([#3067](https://github.com/can1357/oh-my-pi/issues/3067))
+
 ## [16.1.3] - 2026-06-19
 
 ### Fixed

package/dist/types/wire/gemini-headers.d.ts

@@ -19,15 +19,18 @@ export declare let getAntigravityUserAgent: () => string;
 /**
  * Per-wire-id Antigravity Cloud Code Assist request constants, captured from the
  * real `antigravity/hub` client against `daily-cloudcode-pa`. `modelEnum` is the
- * opaque `labels.model_enum` token the client tags each request with;
- * `maxOutputTokens` is the fixed `generationConfig.maxOutputTokens` it sends
- * regardless of the thinking budget. Keyed by the routed upstream wire id
- * (post effort-routing), not the collapsed logical id. Checkpoint-only ids
+ * opaque `labels.model_enum` token the client tags each request with — optional
+ * because Anthropic-backed wire ids (e.g. `claude-sonnet-4-6`,
+ * `claude-opus-4-6-thinking`) are accepted without one; the label is purely
+ * telemetry. `maxOutputTokens` is the fixed `generationConfig.maxOutputTokens`
+ * the backend enforces regardless of the thinking budget (Claude caps at
+ * 64000, Gemini accepts the discovered cap). Keyed by the routed upstream wire
+ * id (post effort-routing), not the collapsed logical id. Checkpoint-only ids
  * (e.g. `gemini-3.1-flash-lite`) are intentionally absent — this provider only
  * emits agent requests.
  */
 export interface AntigravityModelWireProfile {
-    modelEnum: string;
+    modelEnum?: string;
     maxOutputTokens: number;
 }
 export declare const ANTIGRAVITY_MODEL_WIRE_PROFILES: Readonly<Record<string, AntigravityModelWireProfile>>;

package/package.json

@@ -1,7 +1,7 @@
 {
 	"type": "module",
 	"name": "@oh-my-pi/pi-catalog",
-	"version": "16.1.3",
+	"version": "16.1.5",
 	"description": "Model catalog for omp: bundled model database, provider discovery descriptors, model identity, classification, and equivalence",
 	"homepage": "https://omp.sh",
 	"author": "Can Boluk",
@@ -34,12 +34,12 @@
 	},
 	"dependencies": {
 		"@bufbuild/protobuf": "^2.12.0",
-		"@oh-my-pi/pi-utils": "16.1.3",
+		"@oh-my-pi/pi-utils": "16.1.5",
 		"arktype": "^2.2.0",
 		"zod": "^4"
 	},
 	"devDependencies": {
-		"@oh-my-pi/pi-ai": "16.1.3",
+		"@oh-my-pi/pi-ai": "16.1.5",
 		"@types/bun": "^1.3.14"
 	},
 	"engines": {

package/src/models.json

@@ -17862,10 +17862,10 @@
 				],
 				"effortRouting": {
 					"off": "claude-sonnet-4-6",
-					"minimal": "claude-sonnet-4-6-thinking",
-					"low": "claude-sonnet-4-6-thinking",
-					"medium": "claude-sonnet-4-6-thinking",
-					"high": "claude-sonnet-4-6-thinking"
+					"minimal": "claude-sonnet-4-6",
+					"low": "claude-sonnet-4-6",
+					"medium": "claude-sonnet-4-6",
+					"high": "claude-sonnet-4-6"
 				}
 			}
 		},

package/src/variant-collapse.ts

@@ -220,8 +220,31 @@ const SHARED_CCA_FAMILIES: readonly EffortVariantFamily[] = [
 		routing: {},
 		thinking: { mode: "budget", efforts: [Effort.Minimal, Effort.Low, Effort.Medium, Effort.High] },
 	},
-	thinkingPair("claude-sonnet-4-6", "Claude Sonnet 4.6"),
-	thinkingPair("claude-opus-4-6", "Claude Opus 4.6"),
+	// Antigravity Cloud Code Assist exposes Claude 4.6 asymmetrically: only the
+	// bare `claude-sonnet-4-6` wire id (no `-thinking` twin) and only the
+	// `claude-opus-4-6-thinking` wire id (no bare twin). Per-effort thinking is
+	// carried in the request body via `thinkingBudget`, so both ids accept on/off
+	// requests. Listing both candidates in `members` (priority order) keeps the
+	// collapse correct if the backend mix ever rebalances; `retiredMembers`
+	// re-points stale collapsed snapshots (bundled catalog rows, cache rows
+	// written by prior generations) away from the dead wire id via
+	// `reconcileRetiredRouting`.
+	{
+		id: "claude-sonnet-4-6",
+		name: "Claude Sonnet 4.6",
+		members: ["claude-sonnet-4-6", "claude-sonnet-4-6-thinking"],
+		retiredMembers: ["claude-sonnet-4-6-thinking"],
+		routing: {},
+		thinking: { mode: "budget", efforts: [Effort.Minimal, Effort.Low, Effort.Medium, Effort.High] },
+	},
+	{
+		id: "claude-opus-4-6",
+		name: "Claude Opus 4.6",
+		members: ["claude-opus-4-6-thinking", "claude-opus-4-6"],
+		retiredMembers: ["claude-opus-4-6"],
+		routing: {},
+		thinking: { mode: "budget", efforts: [Effort.Minimal, Effort.Low, Effort.Medium, Effort.High] },
+	},
 	thinkingPair("claude-sonnet-4-5", "Claude Sonnet 4.5"),
 	thinkingPair("claude-opus-4-5", "Claude Opus 4.5"),
 	thinkingPair("gemini-2.5-flash", "Gemini 2.5 Flash"),

package/src/wire/gemini-headers.ts

@@ -43,15 +43,18 @@ export let getAntigravityUserAgent = () => {
 /**
  * Per-wire-id Antigravity Cloud Code Assist request constants, captured from the
  * real `antigravity/hub` client against `daily-cloudcode-pa`. `modelEnum` is the
- * opaque `labels.model_enum` token the client tags each request with;
- * `maxOutputTokens` is the fixed `generationConfig.maxOutputTokens` it sends
- * regardless of the thinking budget. Keyed by the routed upstream wire id
- * (post effort-routing), not the collapsed logical id. Checkpoint-only ids
+ * opaque `labels.model_enum` token the client tags each request with — optional
+ * because Anthropic-backed wire ids (e.g. `claude-sonnet-4-6`,
+ * `claude-opus-4-6-thinking`) are accepted without one; the label is purely
+ * telemetry. `maxOutputTokens` is the fixed `generationConfig.maxOutputTokens`
+ * the backend enforces regardless of the thinking budget (Claude caps at
+ * 64000, Gemini accepts the discovered cap). Keyed by the routed upstream wire
+ * id (post effort-routing), not the collapsed logical id. Checkpoint-only ids
  * (e.g. `gemini-3.1-flash-lite`) are intentionally absent — this provider only
  * emits agent requests.
  */
 export interface AntigravityModelWireProfile {
-	modelEnum: string;
+	modelEnum?: string;
 	maxOutputTokens: number;
 }
 export const ANTIGRAVITY_MODEL_WIRE_PROFILES: Readonly<Record<string, AntigravityModelWireProfile>> = {
@@ -60,6 +63,11 @@ export const ANTIGRAVITY_MODEL_WIRE_PROFILES: Readonly<Record<string, Antigravit
 	"gemini-3-flash-agent": { modelEnum: "MODEL_PLACEHOLDER_M132", maxOutputTokens: 65536 },
 	"gemini-3.1-pro-low": { modelEnum: "MODEL_PLACEHOLDER_M36", maxOutputTokens: 65535 },
 	"gemini-pro-agent": { modelEnum: "MODEL_PLACEHOLDER_M16", maxOutputTokens: 65535 },
+	// Claude on `daily-cloudcode-pa` rejects `maxOutputTokens > 64000` with a
+	// 400 (`Request contains an invalid argument`). The model_enum label is
+	// untracked for these ids; the backend does not require it.
+	"claude-sonnet-4-6": { maxOutputTokens: 64000 },
+	"claude-opus-4-6-thinking": { maxOutputTokens: 64000 },
 };
 export function getAntigravityModelWireProfile(wireModelId: string): AntigravityModelWireProfile | undefined {
 	return ANTIGRAVITY_MODEL_WIRE_PROFILES[wireModelId];