playbooks

home / skills / plurigrid / asi / pkg-memory-bridge

pkg-memory-bridge skill

/skills/pkg-memory-bridge

This skill bridges memory systems with PKG knowledge graphs to organize and retrieve personal data across Mem0, Graphiti, Solid PODs and Logseq.

npx playbooks add skill plurigrid/asi --skill pkg-memory-bridge

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
2.4 KB
---
name: pkg-memory-bridge
description: Bridge to PKG systems (Mem0, Graphiti, Solid PODs, Logseq) for individuated information indices
version: 1.0.0
---


# PKG Memory Bridge Skill

Connects music-topos to external Personal Knowledge Graph systems.

## GF(3) Triads

```
shadow-goblin (-1) ⊗ pkg-memory-bridge (0) ⊗ gay-mcp (+1) = 0 ✓  [Memory Trace]
temporal-coalgebra (-1) ⊗ pkg-memory-bridge (0) ⊗ agent-o-rama (+1) = 0 ✓  [Temporal KG]
keychain-secure (-1) ⊗ pkg-memory-bridge (0) ⊗ pulse-mcp-stream (+1) = 0 ✓  [Auth + Stream]
```

## Supported Systems

| System | API | Use Case |
|--------|-----|----------|
| Mem0 | `pip install mem0ai` | LLM agent memory |
| Graphiti | MCP Server | Temporal knowledge graph |
| Solid POD | REST/SPARQL | Decentralized personal data |
| Logseq | Local DB | Block-level PKB |

## Quick Integration

```python
from mem0 import Memory
m = Memory()
m.add("User prefers GF(3) balanced triads", user_id="bmorphism")
results = m.search("color conservation", user_id="bmorphism")
```

## Graphiti MCP

```bash
# Add to .mcp.json
{"mcpServers": {"graphiti": {"command": "uvx", "args": ["graphiti-mcp"]}}}
```

## Key Researchers

- Krisztian Balog (PKG ecosystem)
- Gordon Bell (MyLifeBits/memex)
- Mem0 team (Prateek Chhikara, Taranjeet Singh)
- Zep/Graphiti (temporal KG)



## Scientific Skill Interleaving

This skill connects to the K-Dense-AI/claude-scientific-skills ecosystem:

### Graph Theory
- **networkx** [○] via bicomodule
  - Universal graph hub

### Bibliography References

- `general`: 734 citations in bib.duckdb



## SDF Interleaving

This skill connects to **Software Design for Flexibility** (Hanson & Sussman, 2021):

### Primary Chapter: 10. Adventure Game Example

**Concepts**: autonomous agent, game, synthesis

### GF(3) Balanced Triad

```
pkg-memory-bridge (○) + SDF.Ch10 (+) + [balancer] (−) = 0
```

**Skill Trit**: 0 (ERGODIC - coordination)

### Secondary Chapters

- Ch3: Variations on an Arithmetic Theme

### Connection Pattern

Adventure games synthesize techniques. This skill integrates multiple patterns.
## Cat# Integration

This skill maps to **Cat# = Comod(P)** as a bicomodule in the equipment structure:

```
Trit: 0 (ERGODIC)
Home: Prof
Poly Op: ⊗
Kan Role: Adj
Color: #26D826
```

### GF(3) Naturality

The skill participates in triads satisfying:
```
(-1) + (0) + (+1) ≡ 0 (mod 3)
```

This ensures compositional coherence in the Cat# equipment structure.

Overview

This skill provides a bridge between an agent's internal music-topos memory model and external Personal Knowledge Graph (PKG) systems like Mem0, Graphiti, Solid PODs, and Logseq. It exposes connectors and lightweight adapters so an individuated information index can be stored, queried, and synchronized across decentralized and local PKG backends. The goal is to let agents persist temporal, block-level, and semantic traces for long-term retrieval and reasoning.

How this skill works

The skill implements shims and clients for each target: a Mem0 client for agent memory, an MCP adaptor for Graphiti temporal graphs, a REST/SPARQL client for Solid PODs, and a local DB accessor for Logseq blocks. It normalizes memory records into an interoperable schema and handles authentication, streaming updates, and temporal annotations. Users call the bridge to add, search, or sync items; the bridge routes operations to the configured backend(s) and reconciles results into a unified index.

When to use it

  • Persist agent memories and retrieval indexes to an external PKG
  • Maintain a temporal knowledge graph for event sequencing and reasoning
  • Share individuated data across decentralized Solid PODs or local Logseq stores
  • Stream agent-authenticated events and logs to Graphiti or Mem0 for long-term analysis

Best practices

  • Model memories as small, contextual records with timestamps and provenance metadata
  • Configure only the backends you need; avoid parallel writes to incompatible schemas
  • Use agent-scoped user_ids or POD identifiers to keep individuated indices separated
  • Enable streaming/auth adapters for Graphiti when you need temporal ordering and event replay
  • Regularly validate SPARQL/REST mappings for Solid PODs to prevent drift

Example use cases

  • Agent logs conversation traces to Mem0 for personalized long-term dialog tuning
  • A research agent stores event sequences in Graphiti to reconstruct experiments with temporal context
  • Personal data from a Solid POD is indexed for private retrieval by an assistant agent
  • Local Logseq blocks hold granular notes and links; the bridge exposes them as a searchable agent memory

FAQ

Which backends require additional installation?

Mem0 requires the mem0ai client library; Graphiti needs an MCP server process; Solid PODs require valid POD credentials; Logseq access can use local DB connectors.

Can I sync the same record to multiple systems?

Yes. The bridge supports multi-target writes but recommends using provenance metadata to avoid accidental duplication and to reconcile conflicts.

Is the bridge opinionated about schema?

It normalizes to a minimal interoperable schema (content, timestamp, provenance, tags) and leaves richer domain modeling to the backend.