Back to skill

Security audit

lgpd-incident-response

Security checks for vulnerabilities and agentic risk

Overview

The scanners classify this skill as benign because they found no flagged behavior in the AIG scan and report no concerns in most dimensions, while noting a possible operational warning gap; coverage is incomplete because several referenced paths were not found in the bundle.

The verdict is benign, but benign does not mean risk-free. Consider whether containment steps that revoke keys or isolate resources fit your incident-response process, and review the expected service impact before applying them. The scan is incomplete: several local path references did not match bundled files, so evidence about those references is limited.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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)

Containment instructions omit a warning that credential revocation and resource isolation can disrupt legitimate services.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The runbook instructs the agent to revoke keys and sessions and isolate compromised routes, instances, or buckets without warning that these actions can interrupt production workloads or lock out legitimate users. The evidence-preservation instruction addresses forensic integrity, but does not disclose these operational impacts.

Content

No source excerpt is available for this finding.