ClawHub
Back to skill

Security audit

Product Launch Radar

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent product-news monitoring workflow that discloses aggregation, scheduled summaries, notifications, and report saving, with no hidden executable payload found.

Before installing, confirm the Feishu webhook, email recipients, cron schedule, timezone, and report path are appropriate for your team. Treat generated competitive-intelligence briefs as shareable only with the configured recipients, and use scoped tokens for the API trigger.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill description highlights aggregation and prioritization but does not clearly disclose that results may be pushed to external notification channels like Feishu or email. This can lead users to invoke the skill without understanding that summarized content, possibly including sensitive preferences or internal monitoring targets, may be transmitted outside the local environment.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The manual trigger example uses a very broad natural-language phrase ('帮我看看今天有什么新产品发布'), which can cause accidental invocation during ordinary conversation rather than through an explicit command. Because this skill performs multi-source aggregation and downstream notification/push behavior, unintended activation could consume resources, collect data unexpectedly, or distribute summaries without deliberate user intent.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The example configures a scheduled push to a Feishu group chat without showing any confirmation of recipient scope, content preview, or warning that information will be sent automatically to an external group. In a monitoring skill that aggregates potentially sensitive competitive intelligence, this can lead to unintended disclosure to the wrong audience or persistent oversharing through recurring notifications.

External Transmission

Medium
Category
Data Exfiltration
Content
**请求**:
```bash
curl -X POST https://your-openclaw-instance/api/skills/product-launch-radar/trigger \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The API example demonstrates sending authenticated requests and triggering Feishu notifications without any privacy, data handling, or recipient-scope warning. This can normalize transmitting aggregated content and metadata to third-party channels without prompting users to consider whether the payload includes confidential operational or competitive information.

Natural-Language Policy Violations

Medium
Confidence
90% confidence
Finding
The manifest description uses only Chinese text, which imposes a specific language on users without any opt-in or indication that the skill is intentionally region- or locale-specific. Under the policy, language constraints should either be optional for users or clearly justified as a documented locale limitation.

Natural-Language Policy Violations

Low
Confidence
88% confidence
Finding
This manifest file contains a locale/timezone setting expressed in natural language policy terms, and it hard-codes a specific regional timezone. Because there is no nearby indication that the timezone is configurable per user preference or that the skill is explicitly China-region-only, this can violate the policy against forcing a locale without user opt-in.

Missing User Warnings

Low
Confidence
84% confidence
Finding
The example states that a full report is saved to a local path without disclosing that the skill writes files or clarifying where data is stored and who can access it. While lower severity than network exfiltration, undocumented file creation can expose sensitive business intelligence through insecure local storage, backups, or shared workspaces.

Static analysis

No suspicious patterns detected.