DoctorClaw Meeting Prep

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: doctorclaw-meeting-prep Version: 1.0.1 The skill bundle contains markdown instructions (SKILL.md) for an AI agent to automate meeting preparation by scanning calendars, emails, and CRM data. While the skill requires access to sensitive personal information to function, the instructions are transparent and aligned with the stated purpose. There is no evidence of data exfiltration to unauthorized third parties, malicious code execution, or prompt injection attempts. The metadata (_meta.json) contains a future-dated timestamp, which appears to be a non-malicious anomaly.

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.

What this means

If granted broad permissions, the agent may read sensitive meeting schedules, attendee relationships, deal context, emails, and tasks.

Why it was flagged

The skill explicitly depends on delegated access to calendar data and optionally private email, CRM/contact, and task systems.

Skill content
**Calendar access** ... **Email access** ... **Contact/CRM context** ... **Task system**
Recommendation

Grant the narrowest practical account scopes, choose specific calendars and data sources, and review provider OAuth/API permissions before use.

What this means

If enabled broadly, the agent could create tasks or update contact notes based on meeting summaries without a separate confirmation step.

Why it was flagged

The instructions include scheduled automatic operation and mutation of task/contact records as part of the workflow.

Skill content
set it to auto-run 30 minutes before each event ... Create any follow-up tasks mentioned ... Updates contact notes.
Recommendation

Use on-demand mode or require confirmation for task creation, contact-note updates, and any external delivery until the workflow is trusted.

What this means

Sensitive notes or inaccurate summaries could be reused in later briefs unless memory storage and retention are managed.

Why it was flagged

The skill stores and later reuses meeting notes, making persistent memory part of future meeting-prep context.

Skill content
Previous meeting notes pulled from memory ... Save meeting notes to memory for future prep briefs
Recommendation

Decide where notes are stored, avoid saving secrets or highly confidential details, and periodically review or delete stored meeting memory.

What this means

Private meeting context could be exposed to a shared chat, third-party service, or insecure file location if the delivery channel is misconfigured.

Why it was flagged

Prep briefs may contain calendar, email, CRM, and task context and can be delivered to external messaging or file destinations.

Skill content
Delivery — where to send the brief (Telegram, Discord, file, or inline) ... Send the prep brief via configured channel
Recommendation

Choose private delivery targets, avoid public/shared channels for confidential meetings, and verify where briefs are stored or sent.