Back to skill

Security audit

Worldcup Predictor

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed World Cup prediction CLI with optional API enrichment and local schedule caching, with no evidence of hidden persistence, exfiltration, or destructive behavior.

Install only if you are comfortable with a Chinese-localized CLI that can read an optional BALLDONTLIE API key from ~/.openclaw/openclaw.json, call that sports API when using team-list enrichment, and update its local schedule.json cache. Do not rely on the bundled schedule as authoritative without checking the data source.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The skill is described as a match predictor, but the documented behavior also includes reading a local secrets file, making authenticated outbound API requests, and reading/writing local cache data. This mismatch undermines informed consent and can hide sensitive operations behind a benign-seeming description, making misuse or over-privileged deployment more likely.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill declares no explicit tool scope even though the documented behavior includes network access and local file writes. Without clear permission boundaries, a host may grant broader capabilities than users expect, increasing the risk of unintended data access or modification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code file includes its title, usage instructions, docstrings, and runtime messages in Chinese, and there is no indication that users may choose another language or that the tool is intentionally limited to a Chinese-speaking context. Under the language/locale policy rule, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The module docstring says update will fetch the latest schedule from Netease, and fetch_schedule_from_netease also describes scraping/browser snapshot use. In practice, cmd_update never performs any network retrieval or snapshot parsing; it simply calls build_schedule_from_netease_data() and saves a static embedded schedule list.

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 19)May include surrounding context.

md
# ── 配置 ──────────────────────────────────────────────────────────────

API_BASE = "https://api.balldontlie.io/fifa/worldcup/v1"
DATA_DIR = os.path.dirname(os.path.abspath(__file__))
SCHEDULE_CACHE = os.path.join(DATA_DIR, "schedule.json")

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 · scripts/predict.py (reported line 27)May include surrounding context.

python
# ── 配置 ──────────────────────────────────────────────────────────────

API_BASE = "https://api.balldontlie.io/fifa/worldcup/v1"
DATA_DIR = os.path.dirname(os.path.abspath(__file__))
SCHEDULE_CACHE = os.path.join(DATA_DIR, "schedule.json")

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The top-level usage documentation states standings requires manual data entry, which communicates a manual workflow. However, cmd_standings explicitly loads cached schedule data and computes standings automatically via compute_group_standings(matches), so the documentation and code intent conflict.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

This JSON schedule uses Chinese-only field values such as stadium names, group labels, and status markers throughout the file, with no indication that the locale is optional or user-selectable. Under the stated policy, forcing a specific language/locale without opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.