Back to skill

Security audit

KARP Research by SoulDriver

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: scanners found research-related external data transmission and broad activation criteria, but treated both as aligned with the skill’s purpose; review is incomplete because a referenced local path was missing.

Consider whether you want research questions sent to the external service and whether the broad activation criteria fit your needs. The scan is partial: it inspected the listed component, but a local path-like reference was missing, so the evidence is incomplete.

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 (2)

Activation criteria extend to general research questions and incidental fact-checking without clear invocation boundaries.

Medium
Category
Trigger Abuse
Confidence
88% confidence
Finding

Whether a multi-model conclusion 'would help' is subjective, and needing to check a factual claim can occur during almost any task. These conditions do not distinguish a user-requested external research workflow from ordinary answering or internal verification, potentially causing unintended KARP invocations.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

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

1. Coverage (free, always first)

bash
curl -s https://agents.souldriver.com.au/v1/coverage \
  -H 'content-type: application/json' \
  -d '{"query": "<the question you are actually researching>", "intent": "<optional: what you are trying to find out, and why>"}'