A specialized safety workflow designed to triage suspicious communications and identify potential fraud or phishing attempts without interacting with dangerous content.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install suspicious-message-safety-check
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install suspicious-message-safety-check using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Suspicious Message Safety Check?
The Suspicious Message Safety Check is a critical safety prompt workflow within the Openclaw Skills ecosystem. It empowers users to calmly evaluate high-risk communications—such as urgent texts, marketplace chats, or account warnings—before taking any action that could compromise their data or finances. By focusing on risk language rather than guarantees, it provides a grounded, objective analysis of digital threats.
This skill operates as a non-interactive diagnostic tool, meaning it never clicks links, opens attachments, or contacts third parties. It serves as a protective buffer, helping users identify emotional manipulation and technical red flags while directing them toward official, verified channels for resolution.
Suspicious Message Safety Check Use Cases
- Evaluating urgent SMS alerts regarding package delivery failures or unpaid tolls.
- Triaging suspicious bank alerts or account lockout warnings that demand immediate login.
- Checking marketplace buyer requests for off-platform communication or overpayment.
- Analyzing family emergency messages or romance-based requests for money or gift cards.
- Assessing job offers, tax notices, or charity requests that include unfamiliar attachments.
How Suspicious Message Safety Check Works
- Data Intake: The user provides redacted text from a suspicious message, ensuring no sensitive personal identifiers are shared.
- Claim Extraction: The skill identifies the claimed sender, requested action, and the specific emotional or technical hooks used.
- Signal Identification: It scans for risk indicators like urgency, sender mismatch, payment pressure, or requests for sensitive codes (OTP/PIN).
- Risk Assessment: A risk level is assigned (Low, Unclear, High, or Stop Now) based on the density of scam signals.
- Verification Guidance: The skill provides a 'Verification Ladder' to help the user confirm the message's legitimacy through official, independent channels.
Suspicious Message Safety Check Setup
To implement this safety workflow within your Openclaw Skills environment, add the prompt logic to your agent configuration:
# Example of integrating the safety check logic
openclaw install safety-check-skill
Ensure users are briefed on redacting PII (Personally Identifiable Information) before pasting message content into the analysis prompt.
Suspicious Message Safety Check Data Schema & Taxonomy
The skill organizes its analysis into a structured output format to ensure clarity and actionable safety.
| Component | Description |
|---|---|
| Quick Risk Read | Summary of risk level and confidence judgment. |
| Extraction | Breakdown of sender, action, and pressure tactics. |
| Risk Signals | Detailed list of identified red flags and why they matter. |
| Next Steps | A prioritized list of safe actions for the user to take. |
| Official Ladder | Step-by-step instructions for out-of-band verification. |
Suspicious Message Safety Check Advanced Features
- Multi-Scenario Support: Tailored analysis for marketplace, family, and corporate impersonation tactics.
- Family-Friendly Explanations: Generates simplified, non-shaming summaries designed to be forwarded to less tech-savvy relatives.
- Safe Response Scripts: Provides pre-written, non-confrontational templates for declining suspicious requests.
- Post-Incident Guidance: Includes immediate remediation steps if a user has already inadvertently shared data or funds.
SKILL.md
Loading
Related Openclaw Skills
Redis Manager
A robust Docker-based Redis management tool designed to provide a shared, persisted caching layer for local development environments.
UX Research Interview Kit
A comprehensive toolkit for turning product uncertainties into disciplined, non-leading user research plans and actionable synthesis grids.
Medical Bill Audit Helper
A document organization and question-preparation tool designed to help users navigate and verify complex medical bills and insurance statements.
Micro Offer Validation Sprint
A 7-day framework for freelancers and creators to validate paid offers through customer discovery and ethical outreach before building a full product.
Codai
Codai is a centralized orchestration skill for managing local development environments, routing complex commands to specialized database and proxy managers.
PostgreSQL Manager
A specialized tool for managing a shared PostgreSQL Docker lifecycle, including container persistence, database creation, and cross-instance snapshots.
Featured*
