npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

openclaw-hrr-memory

v1.0.1

Published

Structured fact recall for OpenClaw agents using Holographic Reduced Representations. Complements RAG with instant <2ms factual lookups.

Readme

openclaw-hrr-memory

Structured fact recall for OpenClaw agents using Holographic Reduced Representations.

RAG handles 80% of memory queries. The other 20% — exact fact recall like "What is Alice's timezone?" — is where it struggles. This plugin fills that gap with instant <2ms structured lookups, zero dependencies, no embeddings API.

Install

openclaw plugins install openclaw-hrr-memory

What It Does

Parses your agent's MEMORY.md into (subject, relation, object) triples and stores them in an HRR index. Agents query facts instantly instead of searching through document chunks.

| Question type | Tool | Speed | |---------------|------|-------| | "What is Jounes's timezone?" | fact_lookup | <2ms | | "Where does Alice work?" | fact_ask | <2ms | | "What did we discuss about deployment?" | memory_search | ~200ms |

Tools

| Tool | Description | |------|-------------| | fact_lookup | Structured subject+relation query. Use first for factual questions. | | fact_ask | Natural language with stop word handling. | | fact_forget | Remove outdated facts. | | fact_rebuild | Force reindex from MEMORY.md. |

The plugin tells the agent to try fact_lookup before memory_search for direct questions via system prompt injection.

Configuration

In your OpenClaw config (~/.openclaw/openclaw.json):

{
  "plugins": {
    "entries": {
      "hrr-memory": {
        "config": {
          "memoryFiles": ["~/.openclaw/workspace/MEMORY.md"],
          "watchInterval": 30000
        }
      }
    }
  }
}

| Option | Default | Description | |--------|---------|-------------| | memoryFiles | Workspace MEMORY.md | Paths to MEMORY.md files to index | | watchInterval | 30000 | File watcher interval (ms). 0 to disable. | | enableObservations | false | Enable belief change tracking (requires hrr-memory-obs) |

Observation Layer (Optional)

Track how facts change over time with hrr-memory-obs:

cd ~/.openclaw/extensions/openclaw-hrr-memory
npm install hrr-memory-obs

Then enable in config:

{
  "plugins": {
    "entries": {
      "hrr-memory": {
        "config": {
          "enableObservations": true
        }
      }
    }
  }
}

This adds four tools:

| Tool | Description | |------|-------------| | fact_history | Temporal changelog for a subject | | fact_observations | Synthesized beliefs about knowledge changes | | fact_flags | Unflushed conflict flags | | fact_observe_write | Store observation about belief changes |

Every MEMORY.md edit is diffed against the previous state. Changed facts are recorded in a timeline, and conflicting values (e.g., timezone changed from UTC to CET) are automatically flagged.

How MEMORY.md Is Parsed

The parser extracts triples from markdown key-value patterns:

## server
- **port**: 8080
- **timezone**: CET

## jounes
- **role**: developer
- **prefers**: concise answers, dark mode

Becomes:

  • (server, port, 8080)fact_lookup subject="server" relation="port"8080
  • (jounes, role, developer)fact_ask "What is Jounes's role?"developer

The ## heading becomes the subject. Key-value lines become relations and objects.

Links

License

MIT