T08 · Insecure Dependencies
Warning
- Location
SKILL.md:9- Finding
Unpinned and Unaudited npm Dependency Installation
- Content
View full analysis
- Remediation
View remediation
Security audit
Cashclaw Whatsapp Manager
Security checks for vulnerabilities and agentic risk
Overview
The skill is mostly coherent for WhatsApp Business management, but it installs an unpinned external CLI and handles customer communications/data with limited safeguards.
Review the npm package provenance before installing, pin a known reviewed version, and run it with least-privilege access. Before using this for real customers, narrow the auto-response triggers and require privacy safeguards for support tickets, customer inquiry data, audience lists, and broadcast campaigns.
Vulnerability Patterns
- Insecure DependenciesIntroduces malicious components through unsafe dependency sources
- 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
Findings (1)
Vulnerability Patterns
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The trigger list includes very common words such as "when" and "where", which are likely to appear in many unrelated customer messages. In a WhatsApp auto-response system, this can cause incorrect automated replies, confuse users, and interfere with escalation to a human agent or the correct workflow.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The support auto-response relies on vague triggers like "help" and "problem", which overlap heavily with normal conversation and can activate in contexts that do not require support handling. This can misroute customer interactions, expose canned support prompts unnecessarily, and reduce reliability of the communication workflow.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The skill directs the operator to analyze support tickets, website content, and existing customer inquiry data to build FAQs, but it does not require notice, consent, minimization, or handling safeguards for potentially sensitive personal data. In a client communications context, this increases the risk of privacy violations, unauthorized processing, and inclusion of customer-specific information in generated templates.
- Content