Back to skill

Security audit

Mood Simulator

Security checks for vulnerabilities and agentic risk

Overview

This mood simulator is not clearly malicious, but it stores and shares inferred mood and energy history with unclear user control.

Review before installing. Use it only if you are comfortable with local storage of inferred mood, energy, routine, context, and history data under the OpenClaw workspace, and with those inferences being written to local signal files for other components to consume. Prefer an updated version that documents opt-in, retention, deletion, and signal-sharing controls.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (37)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The skill appears to include undeclared filesystem read/write operations, signal or queue management, local archival, and unrelated batch/statistics/cleanup capabilities. Combining broad hidden capabilities with a benign wellness-themed description increases the chance of unauthorized data handling and execution of side effects outside user expectations.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill appears to include undeclared filesystem read/write operations, signal or queue management, local archival, and unrelated batch/statistics/cleanup capabilities. Combining broad hidden capabilities with a benign wellness-themed description increases the chance of unauthorized data handling and execution of side effects outside user expectations.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill appears to include undeclared filesystem read/write operations, signal or queue management, local archival, and unrelated batch/statistics/cleanup capabilities. Combining broad hidden capabilities with a benign wellness-themed description increases the chance of unauthorized data handling and execution of side effects outside user expectations.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 12)May include surrounding context.

md
node scripts/simulator.js <命令> [选项]

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 17)May include surrounding context.

md
node scripts/simulator.js <命令> [选项]

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 30)May include surrounding context.

md
node scripts/simulator.js <命令> [选项]

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 33)May include surrounding context.

md
node scripts/simulator.js <命令> [选项]

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 36)May include surrounding context.

md
node scripts/simulator.js <命令> [选项]

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill documentation exposes executable script entrypoints and static analysis indicates environment-capable code, yet the manifest declares no explicit tool scope or permissions. This creates an authorization and transparency gap: operators and users cannot tell that the skill may read environment-derived data or rely on capabilities beyond its stated surface area.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger conditions are broad and ambiguous, such as emotional transitions, energy changes, and focus adjustments, which can match ordinary conversation or normal user state. In context, that makes the skill more dangerous because it may activate frequently and unexpectedly, especially when combined with undisclosed persistence and signal-emission behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The natural-language description and usage text are presented in Chinese, which effectively imposes a language requirement on users without any opt-in or stated regional justification. Under the policy, forcing a specific language without user choice is a locale-policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description is written entirely in Chinese and provides no indication that users can opt into another language or locale. Under the policy criteria, forcing a specific language without user choice or documented region-specific justification is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The file's natural-language instructions and labels are entirely Chinese, and there is no indication that the skill offers a language/locale choice or that it is intentionally limited to a Chinese-language or region-specific context. SQP-3 applies to all file types and flags forced language constraints without user opt-in or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document explicitly describes inferring user state from health records, sleep, exercise, stress indicators, and historical conversation patterns without any mention of user notice, consent, minimization, or retention controls. This is privacy-sensitive profiling that can expose intimate behavioral and health-related data, creating compliance and trust risks if the skill processes such data silently.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill uses time-of-day and historical behavior patterns to classify user mood and adapt responses, but does not warn users that profiling is occurring. This is dangerous because users may not realize that contextual and historical signals are being used to infer sensitive mental and behavioral traits, undermining transparency and informed consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill specifies persistent storage of inferred emotional-state history, energy levels, timestamps, triggers, and conversation excerpts, which together form a sensitive behavioral profile. Retaining this without clear user notice, consent, minimization, and retention controls creates a meaningful privacy and profiling risk, especially because the data is inferred rather than explicitly provided and can reveal mental state patterns over time.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The integration section describes sharing inferred emotional-state data with multiple downstream systems, increasing the spread and secondary use of sensitive profile data. Without an explicit warning, purpose limitation, or sharing controls, this creates elevated risk of overcollection, unauthorized reuse, and broader privacy harm across the system.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document specifies continuous inference of user routines, emotional/energy state, conversation-derived signals, and external contextual factors, plus historical storage and ongoing learning, but does not mention consent, minimization, retention, or transparency. In a mood/emotion-tracking skill, this increases privacy risk because highly sensitive behavioral profiles can be collected and retained without clear boundaries, enabling overcollection, unexpected secondary use, or exposure of intimate personal patterns if the surrounding system is weakly governed.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

Natural-language strings, keyword analysis, and user-facing output are all defined in Chinese, and emotion detection relies on Chinese-specific keywords. This effectively forces a specific language/locale behavior without user opt-in or clear justification in the file.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script persists mood history, factors, context, and message metadata to local disk without any clear disclosure or consent mechanism. Because mood and context data are sensitive behavioral signals, silent retention increases the chance of privacy harm, unauthorized local access, and later secondary use beyond the user's expectations.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The script publishes signals through a shared signal manager, enabling cross-skill communication that is not clearly necessary for a local mood simulator. Emitting mood-related state to other components expands the data exposure surface and can allow unintended consumers to observe or act on sensitive inferred user state.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The main assessment flow not only computes mood state but also publishes state and low-energy alerts externally. In the context of an emotion-tracking skill, broadcasting inferred emotional or energy state increases privacy risk because sensitive inferences can propagate beyond the original feature boundary without a clear need-to-know basis.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill goes beyond transient mood simulation and derives persistent behavioral profiles such as chronotype, peak hours, and low hours, then stores them on disk. This creates a longer-term user profile that is more privacy-invasive than the stated feature set and can be misused for surveillance, inference of routines, or cross-feature targeting if other components access the same workspace.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code file contains user-facing comments and console output entirely in Chinese, including status messages and next-step instructions. That imposes a specific language on users without any opt-in or documented locale justification, matching the language/locale policy violation criterion.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This code uses Chinese exclusively in comments and console messages such as configuration save/load/reset notices, with no indication that the skill is region-specific or that users can opt into the locale. Under the policy, forcing a specific language without user choice is a natural-language policy violation.

Content

No source excerpt is available for this finding.