Back to skill

Security audit

Neverland Farm

Security checks for vulnerabilities and agentic risk

Overview

This skill automates a Neverland virtual farm as advertised, but users should understand it can change farm state by selling items and advancing the day.

Install only if you are comfortable giving the skill a Neverland API key and letting it automatically mutate your virtual farm, including selling inventory and advancing the day. Prefer explicit Neverland-branded commands, and avoid enabling the optional cron schedule unless you want repeated unattended runs.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (12)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The README advertises autonomous state-changing actions such as automatic selling, planting, and building, but it does not clearly warn users that the skill will make irreversible in-game decisions on their behalf. In this skill context, those operations directly spend resources and alter farm state, so lack of explicit disclosure increases the chance of unintended loss or misuse even if the behavior is part of the intended functionality.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The README instructs users to run npx clawhub install neverland-farm without pinning an exact package version or integrity source. This can expose users to supply-chain risk if the referenced package or one of its transitive dependencies is updated maliciously or unexpectedly, especially because installation is the first step users are told to perform.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill advertises executable automation that uses environment variables and remote API access, but the manifest does not declare any explicit tool scope or permission boundaries. This can cause the agent platform to invoke the skill without clear least-privilege constraints, increasing the risk of unintended network actions or secret exposure through environment-backed credentials.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
76% confidence
Finding

The visible skill name, description, triggers, and user instructions are all presented only in Chinese, with no indication that users may interact in other languages or that Chinese is a required locale. Under the stated policy, a fixed language without user opt-in can be a natural-language policy violation unless clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases are generic farming terms that can match ordinary conversation unrelated to this specific game, causing accidental activation of an automation skill. In context, accidental invocation is more dangerous because the skill is designed to perform state-changing actions against a live external service using stored credentials.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% 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/farm_smart.py (reported line 65)May include surrounding context.

python
if method == "GET":
                r = requests.get(url, headers=self.headers, timeout=30)
            else:
                r = requests.post(url, headers=self.headers, json=data, timeout=30)
            r.raise_for_status()
            return r.json()
        except requests.exceptions.HTTPError as e:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script automatically sells every item found in the user's inventory without any confirmation, filtering, dry-run preview, or allowlist. Because selling changes user-owned assets and may be irreversible or strategically harmful, a trigger invoking this skill could cause unintended loss of valuable items.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill advances to the next day automatically as part of its normal flow, with no disclosure or consent despite this being an irreversible state transition. In a farm-management context this can consume time-based opportunities, alter game state, and trigger downstream changes the user did not intend.

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
88% confidence
Finding

The trigger "农场" is extremely short and broad, making unintended matching likely during normal discussion. Because this skill can automate actions against an external farm account, accidental routing may lead to unwanted operations rather than a harmless response.

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
86% confidence
Finding

The trigger "收菜" is colloquial and broadly applicable, so it may collide with ordinary farming or gaming conversation. In this skill's context, that raises the chance of unintended automation against the user's live Neverland farm.

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
86% confidence
Finding

The trigger "种田" is generic and likely to appear in unrelated chat, making overmatching plausible. Since the skill is action-oriented and uses API credentials, even low-friction accidental invocation creates avoidable risk.

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
87% confidence
Finding

The trigger "收获" is too broad to uniquely identify this skill and may fire in many benign contexts. That broad matching is risky here because the skill is not informational only; it is intended to mutate remote game state.

Content

No source excerpt is available for this finding.