network-ai 4.0.4 → 4.0.5
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 +61 -3
- package/package.json +3 -1
package/README.md
CHANGED
|
@@ -4,7 +4,7 @@
|
|
|
4
4
|
|
|
5
5
|
[](https://github.com/jovanSAPFIONEER/Network-AI/actions/workflows/ci.yml)
|
|
6
6
|
[](https://github.com/jovanSAPFIONEER/Network-AI/actions/workflows/codeql.yml)
|
|
7
|
-
[](https://github.com/jovanSAPFIONEER/Network-AI/releases)
|
|
8
8
|
[](https://www.npmjs.com/package/network-ai)
|
|
9
9
|
[](https://clawhub.ai/skills/network-ai)
|
|
10
10
|
[](https://nodejs.org)
|
|
@@ -60,7 +60,60 @@ console.log(result.data); // "Hello, World! Your task: welcome"
|
|
|
60
60
|
|
|
61
61
|
That's it. No config files, no setup wizards. Add more agents, swap frameworks, layer on security -- all optional.
|
|
62
62
|
|
|
63
|
-
##
|
|
63
|
+
## Demos
|
|
64
|
+
|
|
65
|
+
### One command to run everything
|
|
66
|
+
|
|
67
|
+
```bash
|
|
68
|
+
npm run demo
|
|
69
|
+
```
|
|
70
|
+
|
|
71
|
+
Interactive menu or use flags directly:
|
|
72
|
+
|
|
73
|
+
```bash
|
|
74
|
+
npm run demo -- --07 # Full AI showcase (OPENAI_API_KEY required)
|
|
75
|
+
npm run demo -- --08 # Control-plane stress demo (no API key)
|
|
76
|
+
npm run demo -- --both # Both sequentially
|
|
77
|
+
npm run demo -- --both --silent-summary # Both, highlights only
|
|
78
|
+
```
|
|
79
|
+
|
|
80
|
+
---
|
|
81
|
+
|
|
82
|
+
### Demo 07 — Full AI Showcase *(requires `OPENAI_API_KEY`)*
|
|
83
|
+
|
|
84
|
+
8-agent pipeline that builds a Payment Processing Service from scratch with full governance:
|
|
85
|
+
|
|
86
|
+
- **FSM state machine** — each agent phase is gated by `JourneyFSM` state transitions
|
|
87
|
+
- **AuthGuardian tokens** — scoped HMAC-signed tokens required at every gate
|
|
88
|
+
- **FederatedBudget** — per-agent token ceilings, hard cut-off on overspend
|
|
89
|
+
- **QualityGateAgent** — AI-assisted security scan + content safety on generated code
|
|
90
|
+
- **Fixer + debugger agents** — two-pass automated remediation loop
|
|
91
|
+
- **Deterministic 10/10 scoring** — 8 objective gates, no LLM score parsing
|
|
92
|
+
- **Cryptographic audit trail** — every write signed to `data/audit_log.jsonl`
|
|
93
|
+
|
|
94
|
+
```bash
|
|
95
|
+
npm run demo -- --07
|
|
96
|
+
```
|
|
97
|
+
|
|
98
|
+
---
|
|
99
|
+
|
|
100
|
+
### Demo 08 — Control-Plane Stress Demo *(no API key)*
|
|
101
|
+
|
|
102
|
+
No LLM calls. Runs in ~2 seconds. Shows the synchronization and governance primitives under load:
|
|
103
|
+
|
|
104
|
+
- **`LockedBlackboard`** — atomic `propose → validate → commit` workflow with file-system mutex
|
|
105
|
+
- **Priority preemption** — high-priority (3) write overwrites low-priority (0) write on same key
|
|
106
|
+
- **FSM timeout** — state times out after 700 ms, FSM hard-stops
|
|
107
|
+
- **Live compliance violations** — TOOL_ABUSE (6 rapid writes), TURN_TAKING (5 consecutive actions), RESPONSE_TIMEOUT (1.2 s sleep), JOURNEY_TIMEOUT — all captured by `ComplianceMonitor`
|
|
108
|
+
- **`FederatedBudget`** — per-agent ceilings tracked without any AI backend
|
|
109
|
+
|
|
110
|
+
```bash
|
|
111
|
+
npm run demo -- --08
|
|
112
|
+
```
|
|
113
|
+
|
|
114
|
+
---
|
|
115
|
+
|
|
116
|
+
### Demo 05 — Code Review Swarm *(requires `OPENAI_API_KEY`)*
|
|
64
117
|
|
|
65
118
|
[](https://youtu.be/UyMsNhaw9lU)
|
|
66
119
|
|
|
@@ -68,7 +121,12 @@ That's it. No config files, no setup wizards. Add more agents, swap frameworks,
|
|
|
68
121
|
|
|
69
122
|
> The demo shows a code review swarm, but Network-AI is not a code review tool. The same orchestration pattern works for research pipelines, data processing, content moderation, customer support routing, document analysis, financial workflows — any task where multiple agents need to coordinate without stepping on each other.
|
|
70
123
|
|
|
71
|
-
|
|
124
|
+
```bash
|
|
125
|
+
npm run demo -- --07 # or run directly:
|
|
126
|
+
npx ts-node examples/05-code-review-swarm.ts
|
|
127
|
+
```
|
|
128
|
+
|
|
129
|
+
Supports 4 modes: built-in code review, paste your own code, system design document, or custom role for any content type. Examples `01`–`03` run without any API key. `04-live-swarm.ts` runs a 10-agent live AI research swarm (3 parallel analyst waves + synthesizer).
|
|
72
130
|
|
|
73
131
|
## Why This Exists -- The Multi-Agent Race Condition Problem
|
|
74
132
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "network-ai",
|
|
3
|
-
"version": "4.0.
|
|
3
|
+
"version": "4.0.5",
|
|
4
4
|
"description": "AI agent orchestration framework for TypeScript/Node.js - plug-and-play multi-agent coordination with 12 frameworks (LangChain, AutoGen, CrewAI, OpenAI Assistants, LlamaIndex, Semantic Kernel, Haystack, DSPy, Agno, MCP, OpenClaw). Built-in security, swarm intelligence, and agentic workflow patterns.",
|
|
5
5
|
"main": "dist/index.js",
|
|
6
6
|
"types": "dist/index.d.ts",
|
|
@@ -29,6 +29,7 @@
|
|
|
29
29
|
"setup": "npx ts-node setup.ts",
|
|
30
30
|
"setup:check": "npx ts-node setup.ts --check",
|
|
31
31
|
"setup:example": "npx ts-node setup.ts --example",
|
|
32
|
+
"demo": "npx ts-node examples/demo-runner.ts",
|
|
32
33
|
"blackboard": "python scripts/blackboard.py",
|
|
33
34
|
"check-permission": "python scripts/check_permission.py",
|
|
34
35
|
"validate-token": "python scripts/validate_token.py",
|
|
@@ -82,6 +83,7 @@
|
|
|
82
83
|
},
|
|
83
84
|
"devDependencies": {
|
|
84
85
|
"@types/node": "^25.3.0",
|
|
86
|
+
"dotenv": "^17.3.1",
|
|
85
87
|
"openai": "^6.23.0",
|
|
86
88
|
"ts-node": "^10.9.2",
|
|
87
89
|
"typescript": "^5.9.3"
|