opencode-orchestrator 1.2.55 → 1.2.58
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 +2 -34
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -10,40 +10,6 @@
|
|
|
10
10
|
|
|
11
11
|
---
|
|
12
12
|
|
|
13
|
-
## ⚡ Why Build a Custom Orchestrator?
|
|
14
|
-
|
|
15
|
-
**TL;DR: A lightweight, self-contained orchestration system is far more reliable and maintainable than relying on OpenCode's built-in APIs.**
|
|
16
|
-
|
|
17
|
-
### 🎯 The Problem with OpenCode's Built-in APIs
|
|
18
|
-
|
|
19
|
-
OpenCode provides basic async session functionality through its `prompt_async` endpoint. However, this approach has significant limitations:
|
|
20
|
-
|
|
21
|
-
❌ **No Control**: Cannot customize parallel processing logic
|
|
22
|
-
❌ **Server Dependency**: API may change when OpenCode updates
|
|
23
|
-
❌ **Performance Bottleneck**: Cannot optimize from plugin side
|
|
24
|
-
❌ **Maintenance Nightmare**: Must update plugin every time OpenCode updates
|
|
25
|
-
|
|
26
|
-
### ✅ Our Solution: Custom Lightweight Orchestrator
|
|
27
|
-
|
|
28
|
-
We built our own orchestration system that delivers:
|
|
29
|
-
|
|
30
|
-
🚀 **90%+ CPU utilization** via work-stealing queues (vs 50-70% with OpenCode's approach)
|
|
31
|
-
🛡️ **99.95% sync accuracy** via MVCC + Mutex
|
|
32
|
-
⚡ **10x faster tool calls** (5-10ms vs 50-100ms)
|
|
33
|
-
💾 **60% memory reduction** via pooling
|
|
34
|
-
🔒 **Zero resource leaks** via RAII pattern
|
|
35
|
-
|
|
36
|
-
### 🔑 Key Benefits
|
|
37
|
-
|
|
38
|
-
1. **Full Control**: Complete authority over concurrency, session management, and state synchronization
|
|
39
|
-
2. **High Performance**: Work-stealing queues, session pooling, Rust connection pool
|
|
40
|
-
3. **Reliability**: Circuit breaker, resource pressure detection, auto-recovery
|
|
41
|
-
4. **Independence**: Minimal impact from OpenCode updates - no more "update plugin every time" headaches
|
|
42
|
-
|
|
43
|
-
[📖 Read the full analysis: Why We Built a Custom Orchestrator Instead of Using OpenCode's APIs →](docs/WHY_CUSTOM_ORCHESTRATOR.md)
|
|
44
|
-
|
|
45
|
-
---
|
|
46
|
-
|
|
47
13
|
## ⚡ Quick Start
|
|
48
14
|
|
|
49
15
|
```bash
|
|
@@ -57,6 +23,8 @@ Inside an OpenCode environment:
|
|
|
57
23
|
|
|
58
24
|
---
|
|
59
25
|
|
|
26
|
+
---
|
|
27
|
+
|
|
60
28
|
## 🚀 Engine Workflow
|
|
61
29
|
|
|
62
30
|
OpenCode Orchestrator utilizes a **Hub-and-Spoke Topology** with **Work-Stealing Queues** to execute complex engineering tasks through parallel, context-isolated sessions.
|
package/package.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"name": "opencode-orchestrator",
|
|
3
3
|
"displayName": "OpenCode Orchestrator",
|
|
4
4
|
"description": "Distributed Cognitive Architecture for OpenCode. Turns simple prompts into specialized multi-agent workflows (Planner, Coder, Reviewer).",
|
|
5
|
-
"version": "1.2.
|
|
5
|
+
"version": "1.2.58",
|
|
6
6
|
"author": "agnusdei1207",
|
|
7
7
|
"license": "MIT",
|
|
8
8
|
"repository": {
|