ClawHub
Back to skill

Security audit

LinkedCamp

Security checks for vulnerabilities and agentic risk

Overview

The skill provides disclosed LinkedCamp outreach capabilities with explicit approval requirements; its broad activation keywords are a minor routing weakness.

Connect the intended account, prefer OAuth, and keep secrets out of chat. Review recipients, message text, public posts, deletions, and automated replies before approving. With the skill alone, confirmations depend on the assistant following instructions rather than enforced approval dialogs. Activated campaigns and agents can continue operating.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/setup.md (reported line 36)May include surrounding context.

not in openclaw.json:

bash
# ~/.openclaw/.env on the Gateway host
LINKEDCAMP_API_KEY=<paste the key here, never into chat>

The activation rule includes unqualified mentions of “prospects” and “sequences,” which can trigger the skill outside LinkedCamp outreach tasks.

Medium
Category
Trigger Abuse
Confidence
96% confidence
Finding

The activation rule treats generic words such as “prospects” and “sequences” as sufficient to invoke LinkedCamp, even when the user is discussing unrelated topics. Incorrect activation can redirect the assistant into an unrelated workflow and prompt an unnecessary authenticated whoami call, although the skill’s explicit confirmation requirements and credential protections substantially limit the potential harm. This is an overbroad routing weakness, not evidence of malicious intent or an established authorization bypass.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.