Simple Random Interaction Designer
PassAudited by VirusTotal on May 12, 2026.
Overview
Type: OpenClaw Skill Name: simple-random-interaction-designer Version: 2.0.0 The OpenClaw skill 'simple-random-interaction-designer' is benign. Its purpose is to randomly decide whether to send a casual message and, if so, suggest an interaction type and description. The Python script uses only standard libraries, generates random numbers, and outputs JSON. All interaction descriptions are hardcoded within the script and are benign, conversational prompts. While the SKILL.md instructs the agent to 'use any relevant OpenClaw-accessible tools, skills, or integrations' for data-aware interactions, this is a general agent capability instruction, and there is no evidence of malicious intent, data exfiltration, or harmful prompt injection within the provided content. The skill also includes mitigating instructions against fabricating context or using unreliable data.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A scheduled casual ping could cause the agent to look up live context from available tools before composing a message.
This authorizes follow-on tool or integration use during proactive checks. It is disclosed and aligned with data-aware message types, but the phrase is broad enough that users should ensure only intended integrations are available.
If the selected interaction is data-aware, use any relevant OpenClaw-accessible tools, skills, or integrations to fetch live context before drafting the message.
Keep OpenClaw integrations scoped to sources you are comfortable using for proactive messages, and require approval for sensitive tools if the platform supports it.
The agent may reference recent personal conversation details in a casual check-in.
The skill may reuse recent conversation details in proactive messages. The instructions limit this to clearly present context and warn not to fabricate, so this is purpose-aligned but privacy-relevant.
Prefer recent chat context when it is clearly present.
Use this only if you are comfortable with recent chat context being reflected in proactive messages; avoid enabling it in highly sensitive conversations.
Proactive messages may feel spontaneous or human-like rather than explicitly automated.
The final message is intentionally framed as natural and casual rather than explaining that it came from a random scheduled process. This is disclosed in the skill and paired with low-pressure wording, but users should be aware of the presentation choice.
Do not mention the random process, scheduled checks, or why this interaction was selected.
Install only if you want this style of casual proactive outreach; consider adding disclosure in your own agent settings if you prefer transparency.