nexgenr 1.0.0 → 1.0.1

package/README.md

@@ -1,6 +1,6 @@
 # nexgenr
 
-NexgenR CLI is a terminal AI coding agent wired to **NexgenRouter**.
+NexgenR CLI is a terminal AI coding agent wired to the **NexgenR red-team** surface.
 
 ## Install
 
@@ -18,22 +18,33 @@ pnpm add -g nexgenr
 
 ## Configure
 
-NexgenR authenticates with a NexgenRouter API key:
+NexgenR signs in with an API key. There is no OAuth and no device flow:
 
 ```sh
 export NEXGENR_API_KEY="<your key>"
-# optional, defaults to https://ia.nexgen.systems/v1
-export NEXGENR_BASE_URL="https://ia.nexgen.systems/v1"
+# optional, defaults to https://redteam.nexgen.systems/v1
+export NEXGENR_BASE_URL="https://redteam.nexgen.systems/v1"
 ```
 
-Validate the key and provision the model catalog:
-
 ```sh
 nexgenr login                       # reads NEXGENR_API_KEY
 nexgenr login --api-key <your key>  # or pass it explicitly
 ```
 
-This reads `GET /v1/models`, writes `~/.nexgenr/config.toml`, and selects `nexgen-auto` as the default model.
+The same key works on both surfaces; the header is what changes:
+
+| Surface | Send the key as | Rejected |
+|---|---|---|
+| `/v1/…` inference and catalog | `Authorization: Bearer <key>` or `x-api-key: <key>` | `x-bf-vk` answers 401 here |
+| `/mcp` gateway | `x-bf-vk: <key>` or `Authorization: Bearer <key>` | — |
+
+The key must be provisioned for the `vllm` provider and the `nexgen-redteam` model.
+Without that grant `GET /v1/models` answers 200 with an empty catalog and chat fails
+with `400 modelo-desconocido`.
+
+`nexgenr login` writes `~/.nexgenr/config.toml` with the `redteam` provider and its single
+model `nexgen-redteam` (262 144 tokens of context), and `~/.nexgenr/mcp.json` with the
+gateway. It does not call the catalog: the model list is static.
 
 ## Quick start
 
@@ -47,13 +58,17 @@ nexgenr            # interactive terminal UI
 
 | Capability | Endpoint |
 |---|---|
-| Chat (OpenAI, Responses, Anthropic) | `POST /v1/chat/completions`, `/v1/responses`, `/v1/messages` |
+| Chat, streaming only | `POST /v1/chat/completions` |
 | Model catalog | `GET /v1/models` |
-| Web search | `POST /v1/tools/search` |
-| URL fetch | `POST /v1/tools/fetch` |
-| Server tools | `openrouter:*` declared in `tools[]` |
+| Health and published limits | `GET /health`, no credential |
+| MCP gateway | `POST /mcp` — 24 tools (8 Brave_Search, 16 Crawl4AI) |
+
+`POST /v1/completions` and `POST /v1/embeddings` are in the edge allowlist but the engine
+answers 404: `POST /v1/chat/completions` is the only inference route.
 
-Authentication uses `Authorization: Bearer <key>` (and `x-api-key` on the Anthropic wire).
+Limits, as published by `GET /health`: 4 requests in flight per consumer with `429` and
+`Retry-After: 10` beyond that, a 12 MiB request body, and 262 144 tokens of context. A
+prefill can take minutes, so requests stream and wait up to 3 600 s for the first byte.
 
 ## Features
 
@@ -68,7 +83,7 @@ Authentication uses `Authorization: Bearer <key>` (and `x-api-key` on the Anthro
 - Remote Control (requires a relay service).
 - CDN-based auto-update; upgrade with `npm install -g nexgenr@latest`.
 - The bundled browser UI; `nexgenr web` serves the API plus a minimal landing page.
-- Device-flow OAuth; `nexgenr login` validates an API key and writes the provider entry.
+- SSO: NexgenR signs in with an API key only.
 
 ## License
 

package/dist/main.mjs

@@ -323009,9 +323009,6 @@ init_src$8();
 init_mcp_provision();
 init_provision();
 const PROBE_TIMEOUT_MS$1 = 15e3;
-function isVirtualKey(key) {
-	return key !== void 0 && key.startsWith("sk-bf-");
-}
 function mcpHeaders(apiKey) {
 	if (apiKey === void 0 || apiKey === "") return {};
 	const headers = redteamAuthHeaders(apiKey);
@@ -323080,21 +323077,15 @@ async function probeModels(doFetch, baseUrl, apiKey, timeoutMs) {
 		detail: "no API key in config.toml — run `nexgenr login --api-key <token>`"
 	};
 	try {
-		const headers = redteamAuthHeaders(apiKey);
 		const response = await timedFetch(doFetch, `${baseUrl}/models`, {
 			method: "GET",
-			headers
+			headers: { authorization: `Bearer ${apiKey}` }
 		}, timeoutMs);
-		if (response.status === 401 || response.status === 403) return isVirtualKey(apiKey) ? {
-			name: "models",
-			kind: "probe",
-			state: "BLOQUEADO",
-			detail: `HTTP ${response.status} — a Bifrost virtual key cannot use /v1; the consumer token (64 hex) is a different credential`
-		} : {
+		if (response.status === 401 || response.status === 403) return {
 			name: "models",
 			kind: "probe",
 			state: "DENEGADO",
-			detail: `HTTP ${response.status} — the consumer token was rejected`
+			detail: `HTTP ${response.status} — the credential was rejected on /v1`
 		};
 		if (!response.ok) return {
 			name: "models",

package/package.json

@@ -1,10 +1,17 @@
 {
   "name": "nexgenr",
-  "version": "1.0.0",
+  "version": "1.0.1",
   "description": "NexgenR — terminal AI coding agent",
   "license": "MIT",
   "author": "NexgenR",
-  "homepage": "https://ia.nexgen.systems",
+  "homepage": "https://git.nexgen.systems/IANexgen/NexgenR",
+  "repository": {
+    "type": "git",
+    "url": "git+https://git.nexgen.systems/IANexgen/NexgenR.git"
+  },
+  "bugs": {
+    "url": "https://git.nexgen.systems/IANexgen/NexgenR/issues"
+  },
   "keywords": [
     "nexgenr",
     "cli",