← Back to Skills Marketplace
dial481

Penfield

by dial481 · GitHub ↗ · v2.0.0 · MIT-0
cross-platform ⚠ suspicious
2658
Downloads
5
Stars
7
Active Installs
5
Versions
Install in OpenClaw
/install penfield
Description
Persistent memory for OpenClaw agents. Store decisions, preferences, and context that survive across sessions. Build knowledge graphs that compound over time...
Usage Guidance
This skill looks functionally coherent but lacks transparency about where and how it stores user data and the npm package has no listed homepage/source to inspect. Before installing: (1) verify the openclaw-penfield npm package and author on the registry and read its source code or repository; (2) confirm the storage backend (local vs cloud), retention policy, encryption at rest/in transit, and which accounts can access memories; (3) test the package in an isolated environment (non-production) and monitor network activity to discover external endpoints; (4) prefer installing only if you can audit the package or the vendor provides clear data-handling documentation. If you cannot verify these, treat the package as higher-risk and avoid installing it in sensitive environments.
Capability Analysis
Type: OpenClaw Skill Name: penfield Version: 2.0.0 The 'penfield' skill bundle provides a comprehensive suite of tools for persistent agent memory and knowledge graph management, including memory storage, session context checkpointing, and artifact management. All tools and instructions in SKILL.md are consistent with the stated purpose of providing long-term memory via the penfield.app service. No evidence of malicious intent, unauthorized data exfiltration, or prompt injection was found.
Capability Assessment
Purpose & Capability
Name, description, and the SKILL.md tools (store/recall/connect/etc.) align: a plugin is expected to provide persistent memory and graph features. Requiring the OpenClaw plugin config flag is coherent.
Instruction Scope
SKILL.md focuses on what to store and how to use memory tools; it does not instruct the agent to read unrelated system files or extra env vars. However the runtime document does not disclose where memories/artifacts are persisted, who can access them, retention/ deletion behavior, or network endpoints used—important scope gaps for a persistence feature.
Install Mechanism
Install spec is an npm package (openclaw-penfield) which is a plausible delivery method for an OpenClaw plugin. But there is no homepage or source repository listed, so you cannot review the package before installing; npm packages from unknown/opaque authors are a moderate supply-chain risk.
Credentials
The skill requests no environment variables or credentials, which is plausible if storage is managed by the platform. However, persistent storage typically involves a backend (local disk, database, or cloud) and the absence of any declared storage credentials or description of where data goes is a transparency concern.
Persistence & Privilege
always is false and the skill is user-invocable; autonomous invocation is allowed but is the platform default. The plugin requests the normal plugin config path only. No indication it modifies other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install penfield
  3. After installation, invoke the skill by name or use /penfield
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.0
**Major update adds new tools and enhances memory graph management.** - Added `penfield_disconnect` tool for removing links between memories in the knowledge graph. - Introduced `penfield_awaken` tool to load or refresh personality configuration at session start. - Updated documentation to describe the new tools and their use cases. - No file changes detected in the codebase; update is documentation and workflow focused.
v1.1.2
Added auth warning to v1.1.2
v1.1.0
Changes to address VirusTotal security scan concerns
v1.0.7
- Changed skill name from "penfield-memory" to "penfield" for consistency. - No code or core functionality changes detected. - SKILL.md metadata updated to reflect the new skill name.
v1.0.6
- SKILL.md with comprehensive documentation on Penfield memory, search strategies, and knowledge graph usage. - Added detailed usage guides for all Penfield tools, including memory, knowledge graph, context management, and artifact handling. - Provided best practices on writing effective, searchable memories, including examples and recommended formatting. - Documented memory types and importance scoring for better organization and recall. - Listed connection and relationship types for building a knowledge graph. - Outlined search tuning, filtering, and recommended workflows for optimal agent use.
Metadata
Slug penfield
Version 2.0.0
License MIT-0
All-time Installs 7
Active Installs 7
Total Versions 5
Frequently Asked Questions

What is Penfield?

Persistent memory for OpenClaw agents. Store decisions, preferences, and context that survive across sessions. Build knowledge graphs that compound over time... It is an AI Agent Skill for Claude Code / OpenClaw, with 2658 downloads so far.

How do I install Penfield?

Run "/install penfield" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Penfield free?

Yes, Penfield is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Penfield support?

Penfield is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Penfield?

It is built and maintained by dial481 (@dial481); the current version is v2.0.0.

💬 Comments