Back to skill

Security audit

reset-context-contamination

Security checks for vulnerabilities and agentic risk

Overview

This skill provides a disclosed conversation-reframing workflow with a mildly broad activation phrase and no hidden system access.

Use this when you intend to set aside earlier drafts. Check that the extracted brief retains your requirements; for substantial resets, it may pass that brief to a fresh subagent. Inspection found no missing executable component behind the scanner's partial status.

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
  • 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 (1)

The trigger "want a fresh take" can activate a context reset for an ordinary request for another perspective.

Low
Category
Trigger Abuse
Confidence
90% confidence
Finding

A user can request a fresh take without asking the agent to discard earlier drafts or framings. The description makes that phrase sufficient for activation without distinguishing routine revision from an intentional context reset.

Content

No source excerpt is available for this finding.