Back to skill

Security audit

Psychological Stress Assessment Skill | 心理压力评估技能

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because scanners report sensitive remote processing, automatic identity and credential handling, and gaps between its stated analysis purpose and detected behavior.

Treat this as a review decision and consider not installing it until its remote processing, account registration, credential storage, and history access are clearly explained and acceptable to you. The scan was partial, so it does not establish that all behavior was inspected.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (24)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

The supplied chunk does not substantiate the declared primary purpose; it is generic scaffolding with a disabled, broken chat helper. Imported BaseSkill and ApiService implementations were not supplied, so mental-health functionality in inherited code cannot be ruled out. No active subprocess execution, network transmission, or other undeclared resource access is demonstrated by this chunk.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

Generic formatting, polling, workspace discovery, and HTTP helpers can reasonably support the declared skill and are not independently mismatches. However, this chunk also implements an undisclosed persistent account lifecycle and remote identity transmission. Its request wrapper accepts absolute URLs and adds configured credentials regardless of destination, a capability broader than the stated mental-health analysis purpose. Debug mode can additionally expose request contents through HTTP logging. No malicious destination or actual credential theft is demonstrated, and this utility file alone cannot establish whether assessment logic exists elsewhere.

Content

No source excerpt is available for this finding.

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: "psychological-stress-assessment-analysis"
description: "Combines facial blood flow and emotional characteristics to analyze stress index, anxiety tendency, and depression tendency, suitable for mental health monitoring scenarios. | 心理压力评估技能,结合面部血流与情绪特征,分析压力指数、焦虑倾向、抑郁倾向,适用于心理健康监测场景"
version: "1.0.26"
license: "MIT-0"
---

# 🧘 Psychological Stress Assessment Skill | 心理压力评估技能
> **智能分析中枢** · 图片/视频智能分析 · 结构化报告 · 历史报告云端查询

---

## 🧭 技能概览 | Overview

| 模块

The analysis workflow lacks a privacy warning about remote processing of facial media and sensitive mental-health assessments.

High
Category
Output Handling
Confidence
95% confidence
Finding

The workflow directs facial-media analysis without an explicit user-facing notice or consent gate for transferring identifiable media to a remote service and associating results with persistent identity. Facial recordings and inferred mental-health information are particularly sensitive, making undisclosed processing materially more dangerous than ordinary media analysis.

Content

No source excerpt is available for this finding.

The request helper attaches service credentials and user identity to arbitrary absolute destination URLs.

High
Category
Excessive Agency
Confidence
99% confidence
Finding

Absolute HTTP and HTTPS URLs bypass the configured base URL, but the helper still adds access tokens, the API key, authorization credentials, and user identity. A caller-influenced destination can therefore receive service secrets and identifying data, while HTTP destinations additionally expose them in transit.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_read, file_write, network, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Generic report-query phrases automatically activate access to psychological assessment history.

Medium
Category
Trigger Abuse
Confidence
91% confidence
Finding

Broad history-query phrases trigger mandatory retrieval and display of sensitive psychological reports without a context or identity confirmation step. Combined with the documented fallback to a reusable local default identity, this can expose another person's history in a shared deployment, although actual cross-user access depends on backend isolation.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to perform analysis, but does not prevent it.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser advertises --export-env-only as preventing analysis, but the parsed flag is never checked before identity initialization or analysis. With media arguments supplied, a caller relying on this promise can unintentionally initiate remote processing of sensitive facial data.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · skills/smyx_analysis/scripts/skill.py (reported line 28)May include surrounding context.

python
result_json = JsonUtil.parse(result_json_pure_text, result_json_pure_text)

        result_json_common_ai_response = result_json.get("commonAiResponse") if isinstance(result_json,
                                                                                           dict) else result_json
        if result_json_common_ai_response:
            result_json = result_json_common_ai_response

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · skills/smyx_analysis/scripts/skill.py (reported line 33)May include surrounding context.

python
result_json = result_json_common_ai_response

        result_json_health_ai_response = result_json.get("healthAiResponse") if isinstance(result_json,
                                                                                           dict) else result_json
        if result_json_health_ai_response:
            result_json = result_json_health_ai_response

Local file contents are passed to the analysis API without disclosure that the file will be uploaded.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The local-file path reads the entire selected file and passes its contents to the analysis service without a disclosure or consent check in this workflow. For facial recordings used to infer mental-health characteristics, a user expecting local analysis could unknowingly disclose sensitive personal data.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to perform analysis, but supplying it does not prevent video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The --export-env-only flag promises not to analyze video, but its value is never used to stop the normal execution path. When supplied alongside input media, it still reaches the analysis helper, potentially causing unintended remote processing of sensitive recordings.

Content

No source excerpt is available for this finding.

The save() docstring promises creation of a new record, but insertion failures can trigger overwrites of existing users.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

Any insertion exception can trigger an update by supplied ID, followed by a username-or-realname lookup that copies all columns except username into an existing record. If incoming model data is attacker-influenced or incorrectly associated, this can overwrite another user's identity and authentication tokens, making the issue particularly sensitive in a mental-health application.

Content

No source excerpt is available for this finding.

FileUtil.open silently truncates existing files without disclosing its destructive behavior.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

Opening a path in 'w' mode immediately erases any existing contents, even before the caller writes anything. The generic method name does not disclose this behavior, and there is no confirmation, action log, or explanatory docstring/comment; exception tracing does not warn about successful truncation.

Content

No source excerpt is available for this finding.

Assessment requests can silently register a remote user and persist account credentials, beyond the described analysis functionality.

Medium
Category
Excessive Agency
Confidence
91% confidence
Finding

When credentials are unavailable, an ordinary request can silently register a remote account using the current identifier and persist returned tokens in the local database. This introduces undisclosed identity sharing and credential retention beyond assessment functionality; the mental-health context increases the sensitivity of linking an identifier to this service.

Content

No source excerpt is available for this finding.

Ordinary HTTP requests can silently trigger remote account registration using the current user's identifier.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The authentication fallback explicitly sends silent=1 and register=1 to the remote login endpoint using the current user's identifier. Embedding account creation inside a general request helper hides a consequential external side effect from users seeking only a psychological assessment.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The POST transmits the current identifier as both openId and mobile to a remote registration endpoint without a visible consent gate. Network access is not inherently unsafe, but this automatic identity transmission creates the same privacy issue as the silent-registration findings; this call does not itself demonstrate transmission of assessment results.

Content

Scanner excerpt · skills/smyx_common/scripts/util.py (reported line 561)May include surrounding context.

python
"source": ConstantEnum.DEFAULT__SKILL_HUB_NAME
            }
            try:
                _response = requests.post(_url, json=_data)
                if _response.status_code == 200:
                    _response_json = _response.json()
                    if _response_json and _response_json.get("success"):

Configuration loading silently creates directories and files, including during module import (L017–L020).

Low
Category
Output Handling
Confidence
90% confidence
Finding

YamlUtil.load creates parent directories and writes a configuration file when the requested path is absent, without a prompt, log, comment, or docstring disclosing this side effect. BaseEnum.init_subclass calls this loader as classes are defined (L044–L059), so merely importing the module can modify the filesystem.

Content

No source excerpt is available for this finding.

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 243)May include surrounding context.

python
username = getattr(model, "username", None)
            if username:
                column_names = self.__model__.__table__.columns.keys()
                update_data = {key: getattr(model, key) for key in column_names if key != "username"}
                return self.update_by_username(username, **update_data)

            return None

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 353)May include surrounding context.

python
column_names = self.__model__.__table__.columns.keys()

            for key in column_names:
                value = getattr(model, key)
                setattr(instance, key, value)

            session.commit()

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 326)May include surrounding context.

python
if filters:
                for key, value in filters.items():
                    query = query.filter(getattr(self.__model__, key) == value)

            if offset:
                query = query.offset(offset)

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 441)May include surrounding context.

python
if filters:
                for key, value in filters.items():
                    query = query.filter(getattr(self.__model__, key) == value)

            return query.scalar()
        finally:

ai_chat claims to invoke an analysis agent, but its execution path never invokes one.

Low
Category
Excessive Agency
Confidence
100% confidence
Finding

The docstring at L26 explicitly states that the function invokes the openclaw agent through subprocess. However, the invocation at L61–L66 is commented out, and L68 substitutes an empty dictionary; accessing result.stderr then raises an exception that is caught, so the function returns None without performing the documented agent analysis.

Content

No source excerpt is available for this finding.

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/util.py (reported line 30)May include surrounding context.

python
elif isinstance(cur, dict):
                        cur = cur.get(part, default)
                    else:
                        cur = getattr(cur, part)
                return cur
            except Exception:
                return default