Back to skill

Security audit

Intercom Support Channel

Security checks for vulnerabilities and agentic risk

Overview

The skill is a transparent setup guide for an Intercom support channel, with sensitive permissions that are expected for its purpose and explicitly scoped by the instructions.

Before installing, review the referenced plugin version and its permissions, use a dedicated low-scope Intercom app token, start with one low-risk channel, set a real human handoff target, and monitor early conversations because the agent can send real customer-facing replies.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
70% confidence
Finding

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

Content

Scanner excerpt · SKILL.md (reported line 54)May include surrounding context.

md
teammate identity — confirm with whoever owns the support workspace before enabling it.

3. In the Intercom Developer Hub, create a **dedicated app** for the bot (don't reuse a
   broader app's token), copy its **Access token**, and grant only what you'll use:
   read/write conversations, plus read contacts and read/write tags if you want those
   features. Keep the token rotatable — you'll want revocation to be one click if
   anything looks wrong.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 53)May include surrounding context.

md
2. **Get authorization first.** This agent will reply to real customers under a real
   teammate identity — confirm with whoever owns the support workspace before enabling it.

3. In the Intercom Developer Hub, create a **dedicated app** for the bot (don't reuse a
   broader app's token), copy its **Access token**, and grant only what you'll use:
   read/write conversations, plus read contacts and read/write tags if you want those
   features. Keep the token rotatable — you'll want revocation to be one click if