Back to skill

Security audit

Project sourcing engine. Supports 6 resource types (Skill/Service/Material/Equipment/Human/Digital). Built for engineers and AI agents. Homepage: https://www.pclawai.com

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Pclaw platform workflow skill with no bundled code or persistence, but users should be aware that using it may send business details to Pclaw APIs and may create platform records.

Install only if you intend to use Pclaw sourcing workflows. Before publishing demands, resources, matches, or expo entries, confirm exactly what will be sent to Pclaw and avoid entering confidential budgets, client names, project details, or commercial terms unless you are comfortable sharing them with that service.

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
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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The standalone trigger phrase "展会" is overly generic and likely to match normal discussion about exhibitions, events, or trade shows. Because this skill includes operational routes such as listing, joining, or creating expos, broad triggering increases the chance of unintended platform actions or unnecessary collection/transmission of user intent and metadata.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The standalone trigger phrase "展会" is overly generic and likely to match normal discussion about exhibitions, events, or trade shows. Because this skill includes operational routes such as listing, joining, or creating expos, broad triggering increases the chance of unintended platform actions or unnecessary collection/transmission of user intent and metadata.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill documents external platform and API endpoints but does not warn users that publishing demands, resources, or earnings queries may send business-sensitive data to an external service. In this context, users may disclose project descriptions, budgets, deadlines, and commercial information without informed consent, creating privacy, confidentiality, and compliance risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.