quoroom 0.1.29 → 0.1.31

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -8,7 +8,7 @@
8
8
 
9
9
  [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
10
10
  [![npm version](https://img.shields.io/npm/v/quoroom)](https://www.npmjs.com/package/quoroom)
11
- [![Tests](https://img.shields.io/badge/tests-1083%20passing-brightgreen)](#)
11
+ [![Tests](https://img.shields.io/badge/tests-1088%20passing-brightgreen)](#)
12
12
  [![GitHub stars](https://img.shields.io/github/stars/quoroom-ai/room)](https://github.com/quoroom-ai/room/stargazers)
13
13
  [![macOS](https://img.shields.io/badge/macOS-.pkg-000000?logo=apple&logoColor=white)](https://github.com/quoroom-ai/room/releases/latest)
14
14
  [![Windows](https://img.shields.io/badge/Windows-.exe-0078D4?logo=windows&logoColor=white)](https://github.com/quoroom-ai/room/releases/latest)
@@ -59,7 +59,7 @@ The architecture draws from swarm intelligence research: decentralized decision-
59
59
 
60
60
  Quoroom is an open research project exploring autonomous agent collectives. Each collective (a **Room**) is a self-governing swarm of agents.
61
61
 
62
- - **Queen** — strategic brain, supports Claude/Codex subscriptions and OpenAI/Claude API
62
+ - **Queen** — strategic brain, supports Claude/Codex subscriptions and OpenAI/Claude/Gemini API
63
63
  - **Workers** — specialized agents that use the queen model
64
64
  - **Quorum** — agents deliberate and vote on decisions
65
65
  - **Keeper** — the human who sets goals and funds the wallet
@@ -250,7 +250,7 @@ Open **http://localhost:3700** (or the port shown in your terminal). The dashboa
250
250
  The **Clerk** tab is your global assistant for the whole local system (not a single room).
251
251
 
252
252
  - Clerk is a full assistant, not only commentary: it can reason, remember, and execute actions for the keeper
253
- - Setup paths: Claude subscription (`claude`), Codex subscription (`codex`), OpenAI API (`openai:gpt-4o-mini`), Anthropic API (`anthropic:claude-3-5-sonnet-latest`)
253
+ - Setup paths: Claude subscription (`claude`), Codex subscription (`codex`), OpenAI API (`openai:gpt-4o-mini`), Anthropic API (`anthropic:claude-3-5-sonnet-latest`), Gemini API (`gemini:gemini-2.5-flash`)
254
254
  - API keys entered in Clerk Setup are validated before saving
255
255
  - Clerk can answer and do: room lifecycle, room settings, task creation, reminders, inter-room messaging, and keeper communication
256
256
  - Clerk can act proactively through scheduled tasks/reminders and activity-driven commentary
@@ -260,9 +260,9 @@ The **Clerk** tab is your global assistant for the whole local system (not a sin
260
260
 
261
261
  API key resolution for Clerk API models:
262
262
 
263
- 1. Any room credential (`openai_api_key` or `anthropic_api_key`)
264
- 2. Clerk-saved API key (`clerk_openai_api_key` / `clerk_anthropic_api_key`)
265
- 3. Environment variable (`OPENAI_API_KEY` / `ANTHROPIC_API_KEY`)
263
+ 1. Any room credential (`openai_api_key`, `anthropic_api_key`, or `gemini_api_key`)
264
+ 2. Clerk-saved API key (`clerk_openai_api_key` / `clerk_anthropic_api_key` / `clerk_gemini_api_key`)
265
+ 3. Environment variable (`OPENAI_API_KEY` / `ANTHROPIC_API_KEY` / `GEMINI_API_KEY`)
266
266
 
267
267
  See full guide: [docs/CLERK.md](docs/CLERK.md)
268
268
 
@@ -536,10 +536,11 @@ Use your existing Claude or ChatGPT subscription, or bring an API key.
536
536
  | `codex` | OpenAI Codex CLI | Spawns CLI process | `npm i -g @openai/codex` |
537
537
  | `openai:gpt-4o-mini` | OpenAI API | HTTP REST | `OPENAI_API_KEY` |
538
538
  | `anthropic:claude-3-5-sonnet-latest` | Anthropic API | HTTP REST | `ANTHROPIC_API_KEY` |
539
+ | `gemini:gemini-2.5-flash` | Gemini API | HTTP REST | `GEMINI_API_KEY` |
539
540
 
540
541
  **CLI models** (`claude`, `codex`) — Full agentic loop with tool use via the CLI. Session continuity via `--resume`. On Windows, `.cmd` wrappers are auto-resolved to underlying `.js` scripts to bypass the cmd.exe 8191-char argument limit.
541
542
 
542
- **API models** (`openai:*`, `anthropic:*`) — Direct HTTP calls. Support multi-turn tool-calling loops. API keys resolve from: room credentials → Clerk-saved keys → environment variables. `anthropic:*` also accepts the `claude-api:` prefix.
543
+ **API models** (`openai:*`, `anthropic:*`, `gemini:*`) — Direct HTTP calls. Support multi-turn tool-calling loops. API keys resolve from: room credentials → Clerk-saved keys → environment variables. `anthropic:*` also accepts the `claude-api:` prefix. `gemini:*` uses Google's OpenAI-compatible endpoint.
543
544
 
544
545
  Workers inherit the queen's model by default, or can use a separate API model.
545
546