@sulala/agent-os 0.1.29 → 0.1.30
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 +12 -0
- package/dashboard-dist/assets/index-Bu-NzOKR.js +83 -0
- package/dashboard-dist/assets/index-LNX1FrFY.css +1 -0
- package/dashboard-dist/index.html +2 -2
- package/dist/cli.js +599 -76
- package/dist/index.js +596 -73
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -2,6 +2,14 @@
|
|
|
2
2
|
|
|
3
3
|
Lightweight **Bun-based Agent Operating System** — micro-agents, installable skills, workflows, and a web dashboard.
|
|
4
4
|
|
|
5
|
+
## Showcase
|
|
6
|
+
|
|
7
|
+
| Dashboard | Chat | Graphs |
|
|
8
|
+
| --------- | ---- | ------ |
|
|
9
|
+
|  |  |  |
|
|
10
|
+
| **Skills** | **Schedules** | **Memory** |
|
|
11
|
+
|  |  |  |
|
|
12
|
+
|
|
5
13
|
## How lightweight is Sulala?
|
|
6
14
|
|
|
7
15
|
Compared to typical AI agent frameworks, **Sulala is extremely small** — the core repo is only on the order of **~100 KB**, and even with dependencies it stays around **sub‑MB scale**.
|
|
@@ -148,9 +156,13 @@ curl -X POST http://127.0.0.1:3010/api/agents/run \
|
|
|
148
156
|
| `OPENROUTER_API_KEY` | OpenRouter API key |
|
|
149
157
|
| `AGENT_OS_AGENTS_DIR` | Override agents directory (default: ~/.agent-os/agents/) |
|
|
150
158
|
| `AGENT_MEMORY_DB_PATH` | SQLite path for memory (default: ~/.agent-os/database.db) |
|
|
159
|
+
| `AGENT_OS_MAX_WORKERS` | Max concurrent task workers (default: 2× CPU). Set to `1` to serialize queue runs and reduce 429 rate limits. |
|
|
160
|
+
| `AGENT_OS_TASKS_DIR` | Task log directory (default: ~/.agent-os/tasks) |
|
|
151
161
|
|
|
152
162
|
API keys can also be set in **Dashboard → Settings → AI Provider** and are stored in `~/.agent-os/config.json`.
|
|
153
163
|
|
|
164
|
+
**Request limit (429):** The runtime retries LLM calls up to 3 times with backoff on rate limit. For pipelines, you can also run graphs from **Schedules** (enqueued) and set `AGENT_OS_MAX_WORKERS=1` so only one task runs at a time.
|
|
165
|
+
|
|
154
166
|
## Project structure
|
|
155
167
|
|
|
156
168
|
```
|