Intelligence Transfer

智能模式转移

Cross-project pattern sharing via IPFS. Lets a different project — or a different machine — fetch and apply patterns this project has already learned.

通过IPFS实现跨项目模式共享。允许其他项目或其他机器获取并应用本项目已学习到的模式。

Why this exists

功能背景

Most learning is project-local.

hooks_transfer

is the escape hatch: publish patterns to IPFS, share the CID, and any peer can ingest them. Equivalent to "a deploy artifact for what your agents have learned."

大多数学习过程都是项目本地化的。

hooks_transfer

提供了突破限制的途径：将模式发布到IPFS，分享CID，任何节点都可以获取这些模式。这相当于“为你的Agent所学到的内容生成部署工件”。

Prerequisite

前置条件

bash

undefined

bash

undefined

Required env var (or equivalent endpoint config)

echo $PINATA_API_JWT


If unset, `hooks_transfer` returns a structured `success: false` with `error: "PINATA_API_JWT not configured"`. Configure before running this skill.

echo $PINATA_API_JWT


如果未设置该环境变量，`hooks_transfer`会返回结构化的`success: false`响应，附带错误信息`error: "PINATA_API_JWT not configured"`。请在运行该skill前完成配置。

Workflows

工作流程

Publish current project's patterns

发布当前项目的模式

bash

undefined

bash

undefined

Inspect what's stored locally first

mcp tool call neural_patterns --json -- '{"list": true}'

Publish to IPFS — returns a CID

mcp tool call hooks_transfer --json -- '{"action": "store"}'


The response includes the IPFS CID. Save it; share it with peers who need the patterns.

mcp tool call hooks_transfer --json -- '{"action": "store"}'


响应结果包含IPFS CID。请保存该CID并分享给需要这些模式的协作方。

Fetch + apply a peer's patterns

获取并应用协作方的模式

bash

undefined

bash

undefined

Pull a CID and apply locally

mcp tool call hooks_transfer --json -- '{"action": "load", "cid": "QmXyz..."}'

Verify they landed

mcp tool call hooks_intelligence_pattern-search --json -- '{"query": "<test>", "limit": 5}'


Patterns are merged with local state, not replaced. Conflicts are resolved by recency (newer wins).

mcp tool call hooks_intelligence_pattern-search --json -- '{"query": "<test>", "limit": 5}'


模式会与本地状态合并，而非替换。冲突将根据更新时间解决（较新的模式优先）。

Mirror an entire project's patterns

镜像整个项目的模式

bash

undefined

bash

undefined

Read patterns from a sibling project on disk and republish under a new CID

mcp tool call hooks_transfer --json -- '{"action": "from-project", "source": "/path/to/peer-project"}'


Useful for consolidating learnings across a monorepo or a fleet of related projects.

mcp tool call hooks_transfer --json -- '{"action": "from-project", "source": "/path/to/peer-project"}'


这对于在单体仓库或一组相关项目中整合学习成果非常有用。

When to use this skill

适用场景

Before a fresh project starts — fetch the relevant patterns from a parent project so the new project's agents start with prior knowledge instead of cold.
After a major learning milestone — publish so other projects benefit.
When debugging a regression — fetch a known-good pattern set to compare against.

新项目启动前——从父项目获取相关模式，让新项目的Agent从已有知识开始，而非从零起步。
完成重要学习里程碑后——发布模式，让其他项目受益。
调试回归问题时——获取已知可用的模式集进行对比。

When NOT to use

不适用场景

Daily — it's a heavyweight operation.
```
agentdb_consolidate
```
does the local equivalent.
For sensitive patterns — IPFS is public by default. Pinata pinning does NOT make patterns private. Strip PII (use
```
aidefence_has_pii
```
first) before publishing.

日常频繁使用——这是一项重量级操作。
```
agentdb_consolidate
```
可实现本地等效功能。
处理敏感模式——IPFS默认是公开的。Pinata固定操作不会使模式私有化。发布前请先移除个人身份信息（可先使用
```
aidefence_has_pii
```
工具检测）。

Caveats

注意事项

IPFS CIDs are content-addressed; republishing the same pattern set gives you the same CID.
Patterns are stored as JSON; they include only the embedding hashes + metadata, not raw text. Decoding requires the same SONA / MicroLoRA adapter version that produced them.
This skill does NOT publish AgentDB rows — only the intelligence-side patterns. To ship full memory, use
```
agentdb_*
```
export tools (out of scope here).

IPFS CID基于内容寻址；重新发布相同的模式集会得到相同的CID。
模式以JSON格式存储；仅包含嵌入哈希和元数据，不包含原始文本。解码需要使用生成这些模式时相同版本的SONA / MicroLoRA适配器。
该skill不会发布AgentDB中的数据行——仅传输智能侧的模式。若要迁移完整内存，请使用
```
agentdb_*
```
系列导出工具（不在本文讨论范围内）。

intelligence-transfer

Original

Translation

Intelligence Transfer

智能模式转移

Why this exists

功能背景

Prerequisite

前置条件

Required env var (or equivalent endpoint config)

Required env var (or equivalent endpoint config)

Workflows

工作流程

Publish current project's patterns

发布当前项目的模式

Inspect what's stored locally first

Inspect what's stored locally first

Publish to IPFS — returns a CID

Publish to IPFS — returns a CID

Fetch + apply a peer's patterns

获取并应用协作方的模式

Pull a CID and apply locally

Pull a CID and apply locally

Verify they landed

Verify they landed

Mirror an entire project's patterns

镜像整个项目的模式

Read patterns from a sibling project on disk and republish under a new CID

Read patterns from a sibling project on disk and republish under a new CID

When to use this skill

适用场景

When NOT to use

不适用场景

Caveats

注意事项

Related

相关内容