Back to skill

Security audit

powerdrill-data-analysis

Security checks across malware telemetry and agentic risk

Overview

This Powerdrill skill appears purpose-built for remote data analysis, but it can upload local files and delete remote datasets or sessions without strong user guardrails.

Review before installing. Use this skill only with a Powerdrill account you trust, avoid uploading secrets or regulated data unless authorized, and require explicit user confirmation before any upload, query submission, dataset deletion, or session deletion.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill clearly requires environment variables, network access, and an external client script, yet it does not declare permissions or capabilities up front. This weakens user and platform visibility into what the skill can access, increasing the risk of unintended credential use or silent data exfiltration through remote API calls.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README explicitly encourages uploading local files and sending natural-language analysis queries to Powerdrill, but it does not clearly warn users that file contents and prompts are transmitted to a third-party external service. This can lead to accidental disclosure of sensitive, regulated, or proprietary data because users may assume the analysis occurs locally or may not appreciate the privacy and compliance implications.

Missing User Warnings

Low
Confidence
83% confidence
Finding
The README instructs users to run cleanup to delete remote sessions and datasets, but it does not warn that these deletions are performed against the external service and may be irreversible. This creates risk of unintended data loss, especially when users treat cleanup as a harmless local housekeeping step.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger language is broad enough to activate on many generic data-related requests, including uploads, querying datasets, and destructive operations like deletion. Overbroad activation can cause the agent to invoke this skill in situations where the user did not specifically intend to send data to Powerdrill, which raises the chance of privacy violations or unintended remote actions.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The documentation instructs users to upload local files and submit analysis queries to a third-party remote API, but it does not prominently warn that local data will leave the host and be processed externally. In a data-analysis skill, this context makes the issue more dangerous because users are likely to provide sensitive business documents, spreadsheets, or reports without understanding the transmission and retention implications.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
def delete_dataset(dataset_id: str) -> dict:
    """DELETE /v2/team/datasets/{id} - Delete a dataset."""
    user_id, api_key = _get_env()
    resp = _session.delete(
        f"{BASE_URL}/v2/team/datasets/{dataset_id}",
Confidence
92% confidence
Finding
DELETE /v2/team/datasets/{id}

Tool Parameter Abuse

High
Category
Tool Misuse
Content
def delete_session(session_id: str) -> dict:
    """DELETE /v2/team/sessions/{id} - Delete a session."""
    user_id, api_key = _get_env()
    resp = _session.delete(
        f"{BASE_URL}/v2/team/sessions/{session_id}",
Confidence
88% confidence
Finding
DELETE /v2/team/sessions/{id}

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.