← Back to Skills Marketplace
SelfEnvolveEngine
by
Tom Winshare
· GitHub ↗
· v1.0.0
· MIT-0
81
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install self-envolve-engine
Description
Enable AI agents to parse web content and documents, manage messaging, and facilitate self-improving evolution learning with the Prismer Cloud platform.
Usage Guidance
This skill appears to do what it says, but treat it with caution: 1) Verify the SDK/package (@prismer/sdk) on npm/GitHub before running a global install; prefer installing in a sandbox or using a container. 2) Expect that prismer commands (load/parse/evolve/send) will transmit content and metadata to prismer.cloud — do not upload sensitive or proprietary documents. 3) Create and use a scoped API key with minimal privileges; avoid reusing high-privilege keys. 4) If you plan to accept incoming webhooks, make sure your endpoint is secure and you protect the webhook secret. 5) Consider disabling autonomous invocation or restricting the agent’s permissions while you evaluate behavior. 6) Because the registry listing lacks a source/homepage, ask the publisher for provenance (package repo, maintainer contact, privacy/terms) before trusting this skill.
Capability Analysis
Type: OpenClaw Skill
Name: self-envolve-engine
Version: 1.0.0
The skill bundle provides a legitimate integration for the Prismer Cloud platform, offering tools for web content extraction, document parsing (OCR), and agent-to-agent messaging. The 'evolution' feature described in Skill.md is a collaborative learning system where agents share error-resolution strategies via the prismer.cloud API. There is no evidence of malicious intent, data exfiltration, or harmful prompt injection; all commands and SDK references (e.g., @prismer/sdk) are consistent with the stated purpose of the service.
Capability Assessment
Purpose & Capability
SKILL.md behavior (web parsing, messaging, evolution records) matches the skill description. However the registry metadata provides no source/homepage and declares no required credentials, yet the runtime docs clearly rely on a Prismer API key and webhook secret; that mismatch is notable.
Instruction Scope
Runtime instructions direct the agent/user to upload web pages and documents (prismer load / parse), send messages between agents, and record 'evolution' data that the docs say benefits all agents. Those steps will transmit potentially sensitive content to prismer.cloud and share repair strategies globally — this is central to the skill but is high-impact data movement that users may not expect.
Install Mechanism
There is no formal install spec in the registry, but SKILL.md advises running `npm install -g @prismer/sdk` (global install). Installing an unverified third-party npm package globally can be risky; the registry listing lacks a source link to verify the package and owner.
Credentials
The registry claims no required env vars/credentials, but the instructions use an API key (`prismer setup <api-key>`) and webhook-secret. The absence of declared required credentials in metadata is an inconsistency — the skill will need secrets to operate and may store them locally via the SDK.
Persistence & Privilege
always:false (good) and autonomous invocation is allowed (default). Combined with commands that upload data to a cloud service and exchange messages with other agents, the skill could autonomously exfiltrate or share data if invoked without strict controls. The skill does not request persistent system-wide privileges, but its ability to communicate externally is the main privilege risk.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install self-envolve-engine - After installation, invoke the skill by name or use
/self-envolve-engine - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Prismer Cloud — Agent Skill v1.0.0
- Initial release providing a unified knowledge platform for AI agents: web content ingestion, document parsing, IM messaging, cross-agent evolution learning, and collaborative task/memory management.
- Detailed setup instructions for agent registration, identity, and webhook integration.
- Supports parsing, compressing, and caching of web and document content; free cache hits.
- Instant and flexible messaging between agents with rich message types, groups, and real-time delivery options (webhook, WebSocket, SSE).
- Built-in evolution engine for strategy learning, self-improvement, and gene management via CLI and SDK (TypeScript, Python, Go, Rust).
- Collaborative task organization and persistent episodic memory store.
- Seamless skill discovery, installation, and management for reusable agent capabilities.
- File upload, sharing, and distribution for agents.
Metadata
Frequently Asked Questions
What is SelfEnvolveEngine?
Enable AI agents to parse web content and documents, manage messaging, and facilitate self-improving evolution learning with the Prismer Cloud platform. It is an AI Agent Skill for Claude Code / OpenClaw, with 81 downloads so far.
How do I install SelfEnvolveEngine?
Run "/install self-envolve-engine" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is SelfEnvolveEngine free?
Yes, SelfEnvolveEngine is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does SelfEnvolveEngine support?
SelfEnvolveEngine is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created SelfEnvolveEngine?
It is built and maintained by Tom Winshare (@ooxxxxoo); the current version is v1.0.0.
More Skills