aimeat 3.3.0 → 3.3.2

package/README.md

@@ -311,7 +311,7 @@ Tell any AI what you want and get a working app. Your node serves a canonical bu
 
 For simple one-off apps, just copy the prompt from the portal landing page, paste it into any AI chat, and you get a working HTML app that uses AIMEAT memory. No registration needed.
 
-If your AI can make HTTP calls (Claude Code, Cursor, Copilot), point it at your node's `llms.txt` and describe what you want — it reads the API docs, checks `/v1/capabilities`, and builds. If your AI chat cannot make HTTP calls (ChatGPT, Gemini, free-tier Claude), copy the app-generation prompt from your node's "Try it" page at `/v1/classic`; the AI asks you what you want and produces an HTML file you paste into the App Catalogue. Add the node as an MCP server and the AI can install extensions and publish apps directly, with no UI at all: `aimeat_app_publish`, `aimeat_extension_install`, `aimeat_cortex_install`, `aimeat_memory_*`, `aimeat_storage_*` cover the whole publish/install/inspect cycle.
+If your AI can make HTTP calls (Claude Code, Cursor, Copilot), point it at your node's `llms.txt` (the index; the full manual behind it is `llms-full.txt`) and describe what you want — it reads the API docs, checks `/v1/capabilities`, and builds. If your AI chat cannot make HTTP calls (ChatGPT, Gemini, free-tier Claude), copy the app-generation prompt from your node's "Try it" page at `/v1/classic`; the AI asks you what you want and produces an HTML file you paste into the App Catalogue. Add the node as an MCP server and the AI can install extensions and publish apps directly, with no UI at all: `aimeat_app_publish`, `aimeat_extension_install`, `aimeat_cortex_install`, `aimeat_memory_*`, `aimeat_storage_*` cover the whole publish/install/inspect cycle.
 
 ### What has been built on it
 

package/dist/.env.example

@@ -314,8 +314,17 @@ AIMEAT_PORT=40050
 # AIMEAT_GOOSE_BIN=goose
 # Its own config and session store, away from any human's goose profile on the same machine.
 # AIMEAT_GOOSE_PATH_ROOT=
-# The provider key the agent calls with. Who may spend it is decided before a turn starts.
+# The provider key the agent calls with. Every chat turn on this node is spent from it: the agent is
+# one shared process with one process-wide key, so an individual's own OpenRouter key and the
+# per-person allowance below do NOT apply to the chat. They apply to /v1/ai/* and /v1/llm. The chat
+# page says which of the two paid, and it says this one until the agent's provider is pointed at
+# this node's /v1/llm proxy.
 # AIMEAT_GOOSE_PROVIDER_API_KEY=
+# Which provider and model the agent runs on, handed to the child as GOOSE_PROVIDER and GOOSE_MODEL.
+# Empty leaves goose's own configuration alone. Setting the model here rather than there also lets
+# the node NAME the model that answered: ACP reports a stop reason and a token count and no model.
+# AIMEAT_GOOSE_PROVIDER=openrouter
+# AIMEAT_GOOSE_MODEL=
 # ── This node's own AI key, and what one person may spend of it ──
 # Empty (the default) means the node pays for nothing and everyone brings their own key, which is
 # how every node behaved before this existed. Set it and a person who has not brought a key can

package/dist/build-stamp.json

@@ -1,10 +1,10 @@
 {
   "schema_version": 1,
-  "version": "3.3.0",
-  "commit": "ff14a2711f15567c13f31d005ac6e39cbe5cf346",
+  "version": "3.3.2",
+  "commit": "7b0b1026f286f589c768e155d72f5c4abf6e83ea",
   "dirty": false,
-  "built_at": "2026-08-16T11:14:52.462Z",
-  "newest_source_mtime": 1786878872198.984,
+  "built_at": "2026-08-16T14:39:23.659Z",
+  "newest_source_mtime": 1786891136265.2168,
   "source_dirs": [
     "src",
     "bin",