network-ai 4.10.2 → 4.10.4
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 -2
- package/SKILL.md +9 -40
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -2,10 +2,10 @@
|
|
|
2
2
|
|
|
3
3
|
**TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination**
|
|
4
4
|
|
|
5
|
-
[](https://network-ai.org/)
|
|
6
6
|
[](https://github.com/Jovancoding/Network-AI/actions/workflows/ci.yml)
|
|
7
7
|
[](https://github.com/Jovancoding/Network-AI/actions/workflows/codeql.yml)
|
|
8
|
-
[](https://github.com/Jovancoding/Network-AI/releases)
|
|
9
9
|
[](https://www.npmjs.com/package/network-ai)
|
|
10
10
|
[](#testing)
|
|
11
11
|
[](#adapter-system)
|
package/SKILL.md
CHANGED
|
@@ -1,40 +1,31 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: Network-AI
|
|
3
|
-
description: "Python orchestration skill:
|
|
3
|
+
description: "Local Python orchestration skill: multi-agent workflows via shared blackboard file, permission gating, token budget scripts, and persistent project context. The bundled Python scripts make no network calls and have zero third-party dependencies. Workflow delegations via the host platform's sessions_send may invoke external model APIs."
|
|
4
4
|
metadata:
|
|
5
5
|
openclaw:
|
|
6
6
|
emoji: "\U0001F41D"
|
|
7
|
-
homepage: https://
|
|
8
|
-
bundle_scope: "Python scripts only (scripts/*.py).
|
|
9
|
-
network_calls: none
|
|
10
|
-
sessions_ops: "platform-provided
|
|
7
|
+
homepage: https://network-ai.org
|
|
8
|
+
bundle_scope: "Python scripts only (scripts/*.py). All execution is local."
|
|
9
|
+
network_calls: "none from bundled scripts; platform sessions_send delegations may invoke external models"
|
|
10
|
+
sessions_ops: "platform-provided"
|
|
11
11
|
requires:
|
|
12
12
|
bins:
|
|
13
13
|
- python3
|
|
14
|
-
optional_bins: []
|
|
14
|
+
optional_bins: []
|
|
15
15
|
env:
|
|
16
|
-
SWARM_TOKEN_SECRET:
|
|
17
|
-
required: false
|
|
18
|
-
description: "Node.js MCP server only — not used by these Python scripts. The Python permission layer uses UUID-based tokens stored in data/active_grants.json."
|
|
19
|
-
SWARM_ENCRYPTION_KEY:
|
|
20
|
-
required: false
|
|
21
|
-
description: "Node.js MCP server only — not used by these Python scripts. The Python blackboard does not encrypt data at rest."
|
|
22
16
|
OPENAI_API_KEY:
|
|
23
17
|
required: false
|
|
24
|
-
description: "Not used by
|
|
25
|
-
MINIMAX_API_KEY:
|
|
26
|
-
required: false
|
|
27
|
-
description: "Not used by these Python scripts. Only used by the optional Node.js MiniMaxAdapter when running the companion npm package."
|
|
18
|
+
description: "Not used by this skill."
|
|
28
19
|
privacy:
|
|
29
20
|
audit_log:
|
|
30
21
|
path: data/audit_log.jsonl
|
|
31
22
|
scope: local-only
|
|
32
|
-
description: "Local append-only JSONL file recording operation metadata
|
|
23
|
+
description: "Local append-only JSONL file recording operation metadata. No data leaves the machine."
|
|
33
24
|
---
|
|
34
25
|
|
|
35
26
|
# Swarm Orchestrator Skill
|
|
36
27
|
|
|
37
|
-
> **Scope
|
|
28
|
+
> **Scope:** The bundled Python scripts (`scripts/*.py`) make no network calls, use only the Python standard library, and have zero third-party dependencies. Tokens are UUID-based (`grant_{uuid4().hex}`) stored in `data/active_grants.json`. Audit logging is plain JSONL (`data/audit_log.jsonl`). Workflow delegations that use the host platform's `sessions_send` may invoke external model APIs outside this skill's control.
|
|
38
29
|
|
|
39
30
|
## Setup
|
|
40
31
|
|
|
@@ -728,25 +719,3 @@ python {baseDir}/scripts/swarm_guard.py supervisor-review --task-id "task_001"
|
|
|
728
719
|
- [Blackboard Schema](references/blackboard-schema.md) - Data structure specifications
|
|
729
720
|
- [Agent Trust Levels](references/trust-levels.md) - How trust is calculated
|
|
730
721
|
- [CLI Reference](QUICKSTART.md) - Full `network-ai` CLI command reference (§ 10. CLI)
|
|
731
|
-
|
|
732
|
-
---
|
|
733
|
-
|
|
734
|
-
## Appendix: Optional Node.js Companion (npm)
|
|
735
|
-
|
|
736
|
-
> **This section describes a SEPARATE project — not part of this ClawHub skill bundle.**
|
|
737
|
-
> The Python scripts above work completely without any of this.
|
|
738
|
-
> Install only if you want MCP server integration with Claude/Cursor/VS Code.
|
|
739
|
-
|
|
740
|
-
```bash
|
|
741
|
-
npm install -g network-ai
|
|
742
|
-
npx network-ai-server --port 3001
|
|
743
|
-
```
|
|
744
|
-
|
|
745
|
-
The companion npm package (`network-ai`) provides:
|
|
746
|
-
- HMAC / Ed25519-signed audit tokens (vs UUID tokens in the Python layer)
|
|
747
|
-
- AES-256 blackboard encryption
|
|
748
|
-
- A standalone MCP server for IDE integration (Claude, Cursor, VS Code)
|
|
749
|
-
- 17 framework adapters (LangChain, AutoGen, CrewAI, DSPy, LlamaIndex, NemoClaw, APS, etc.)
|
|
750
|
-
- A full CLI (`network-ai bb`, `network-ai auth`, `network-ai budget`, `network-ai audit`)
|
|
751
|
-
|
|
752
|
-
None of the above are provided by this skill's Python scripts. No network calls are made by this skill.
|
package/package.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "network-ai",
|
|
3
|
-
"version": "4.10.
|
|
3
|
+
"version": "4.10.4",
|
|
4
4
|
"description": "AI agent orchestration framework for TypeScript/Node.js - 17 adapters (LangChain, AutoGen, CrewAI, OpenAI Assistants, LlamaIndex, Semantic Kernel, Haystack, DSPy, Agno, MCP, OpenClaw, A2A, Codex, MiniMax, NemoClaw, APS + streaming variants). Built-in CLI, security, swarm intelligence, real-time streaming, and agentic workflow patterns.",
|
|
5
|
-
"homepage": "https://
|
|
5
|
+
"homepage": "https://network-ai.org",
|
|
6
6
|
"main": "dist/index.js",
|
|
7
7
|
"types": "dist/index.d.ts",
|
|
8
8
|
"bin": {
|