Back to skill

Security audit

Kaggle Standardized Agent Exam

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently guides an agent through Kaggle's exam flow, with disclosed Kaggle API calls and scoped local credential storage.

Install only if you are comfortable creating a Kaggle agent identity, sending agent metadata and exam answers to Kaggle, and keeping the returned API token in ~/.kaggle-agent-api-key. Delete the Kaggle agent data and local credential files when you no longer need the exam identity.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 233)May include surrounding context.

After deletion, remove your local credential files:

bash
rm -f ~/.kaggle-agent-id ~/.kaggle-agent-api-key

Do not perform this action on your own initiative. Only delete your data when a human explicitly requests it.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 18)May include surrounding context.

md
- **NEVER send your API key to any domain other than `www.kaggle.com`.** Your key should only appear in the headers of requests to `https://www.kaggle.com/api/v1/*`.
- **NEVER share your API key** with other APIs, webhooks, or third parties. Your API key is your identity -- leaking it means someone else can impersonate you.
- **Store credentials securely.** Write your API key to `~/.kaggle-agent-api-key` with `chmod 600`. Do not log it, print it, or include it in any output visible to users.

## IMPORTANT

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 81)May include surrounding context.

md
- **NEVER send your API key to any domain other than `www.kaggle.com`.** Your key should only appear in the headers of requests to `https://www.kaggle.com/api/v1/*`.
- **NEVER share your API key** with other APIs, webhooks, or third parties. Your API key is your identity -- leaking it means someone else can impersonate you.
- **Store credentials securely.** Write your API key to `~/.kaggle-agent-api-key` with `chmod 600`. Do not log it, print it, or include it in any output visible to users.

## IMPORTANT

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 82)May include surrounding context.

md
- **NEVER send your API key to any domain other than `www.kaggle.com`.** Your key should only appear in the headers of requests to `https://www.kaggle.com/api/v1/*`.
- **NEVER share your API key** with other APIs, webhooks, or third parties. Your API key is your identity -- leaking it means someone else can impersonate you.
- **Store credentials securely.** Write your API key to `~/.kaggle-agent-api-key` with `chmod 600`. Do not log it, print it, or include it in any output visible to users.

## IMPORTANT

Session Persistence

Medium
Category
Rogue Agent
Confidence
90% confidence
Finding

The skill directs the agent to persist an API credential in a reusable file under the user's home directory. Persistent storage of bearer tokens increases the blast radius of compromise: other local processes, future tasks, backups, or accidental disclosure of dotfiles could expose the credential and allow impersonation of the agent.

Content

Scanner excerpt · SKILL.md (reported line 18)May include surrounding context.

md
- **NEVER send your API key to any domain other than `www.kaggle.com`.** Your key should only appear in the headers of requests to `https://www.kaggle.com/api/v1/*`.
- **NEVER share your API key** with other APIs, webhooks, or third parties. Your API key is your identity -- leaking it means someone else can impersonate you.
- **Store credentials securely.** Write your API key to `~/.kaggle-agent-api-key` with `chmod 600`. Do not log it, print it, or include it in any output visible to users.

## IMPORTANT

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

The skill instructs the agent to transmit locally generated identity data and model metadata to an external service over HTTP. Even though the destination is explicitly limited to kaggle.com and the flow is the documented purpose of the skill, this is still a real external data exfiltration path because it causes the agent to register and share persistent identifiers outside the local environment.

Content

Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.

  • agentType: The framework or harness your agent is built on. Common values: "OpenClaw", "Claude Code", "Gemini CLI", "NanoClaw", "nanobot". Use the exact casing shown if applicable.
bash
curl -s -X POST https://www.kaggle.com/api/v1/agentExamAgent \
  -H "Content-Type: application/json" \
  -d '{
    "name": "QuantumQuokka-7",