mcpField
I
intelligence-transfer

intelligence-transfer

github.com/ruvnetAgent workflows80.0k683 installsUpdated today
Repository

About intelligence-transfer

intelligence-transfer is a skill for sharing learned patterns between projects or machines through IPFS. It can publish local patterns, fetch and merge a peer's patterns, or mirror patterns from another project; patterns are public by default.

Key features

  1. 01Publishes patterns to IPFS
  2. 02Loads patterns by CID
  3. 03Merges patterns with local state
  4. 04Mirrors patterns from another project
  5. 05Resolves conflicts by recency

Use cases

  1. 01Seed a new project with prior patterns
  2. 02Share learnings after a milestone
  3. 03Compare patterns when debugging regressions

MCP support

It calls MCP tools to inspect patterns and transfer them to or from IPFS. The skill transfers intelligence-side patterns, not AgentDB rows.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/ruvnet/ruflo --skill intelligence-transfer

After installing, type/intelligence-transferin Claude Code to confirm it loaded.

Details

From the project README

Intelligence Transfer

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

Why this exists

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."

Prerequisite

# 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.

Workflows

Publish current project's patterns
# 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.

Fetch + apply a peer's patterns
# 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).

Mirror an entire project's patterns
# 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.

When to use this skill

  • 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.

When NOT to use

  • 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.

Caveats

  • 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).

Related

  • ruflo-agentdb ADR-0001 §"Namespace convention" — defines pattern namespace that this transfer reads from
  • neural-train skill — produces the patterns that this skill ships

Comments

Markdown supported