Back to skill

Security audit

Passive Vital Signs Monitoring Tool | 无感生命体征监测分析工具

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with high confidence because scanners report behavior that does not match its stated camera based health monitoring purpose, along with risks involving sensitive data, credentials, and user accounts.

Consider avoiding installation until the publisher explains the account and credential handling, the external video processing and consent flow, and why the implementation does not clearly support its stated monitoring purpose. The scan was partial because one reference could not be resolved, so it does not establish that all behavior is understood.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/contactless_vital_signs_monitoring.py
Finding

Default identity fallback overrides the upstream sender and shares report access

Content
View full analysis

Impact

Reports created through this workflow for one sender become accessible to another sender using the same workspace default account. Health-analysis records and report links can consequently cross user boundaries.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
Findings (24)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The visible implementation does not substantiate the declared monitoring purpose. It is primarily scaffolding with a nonfunctional agent-chat helper, not a camera-based vital-sign analysis implementation. Imported base classes and API services are not supplied, so their behavior cannot be assessed. No subprocess execution or other external action should be inferred from the commented-out code.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The supplied chunk implements user-account persistence rather than camera capture, physiological signal processing, or vital-sign estimation. Database access can support a monitoring application, but this code specifically manages user profiles and tokens without showing a connection to the declared monitoring function. That materially different functionality is not represented in the description. No network transmission or malicious behavior is evident; this assessment is limited to the supplied chunk.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

Generic formatting, polling, and request helpers can be supporting implementation details. However, the automatic identity provisioning, remote account lifecycle, credential transmission, and paid-account workflow introduce material behavior not disclosed by the camera-based monitoring description. These actions occur when the relevant helpers are invoked; their presence does not establish that they execute merely on import. The absence of measurement code in this shared module alone would not establish a mismatch, nor does this chunk establish malicious intent or actual transmission of camera footage.

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: "contactless-vital-signs-monitoring-analysis"
description: "Non-contact detection of heart rate, respiration, blood oxygen, and heart rate variability. No wearable devices are required; monitoring is achieved solely through camera footage. | 无感生命体征监测分析技能,非接触检测心率、呼吸、血氧、心率变异性,无需穿戴设备,通过摄像头画面即可监测"
version: "1.0.26"
license: "MIT-0"
---

# 💗 Passive Vital Signs Monitoring Tool | 无感生命体征监测分析工具

> **智能健康/识别分析中枢** · 图片/视频智能分析 · 结构化报告 · 历史报告云端查询

---

## �

The analysis workflow lacks a privacy warning about sending facial video to an external API for health analysis.

High
Category
Output Handling
Confidence
94% confidence
Finding

Although cloud APIs are mentioned, the workflow does not clearly warn that local facial footage will leave the device or require informed authorization before transmission. Facial video associated with health inferences is particularly sensitive, making automatic analysis a significant privacy risk.

Content

No source excerpt is available for this finding.

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

High
Category
Excessive Agency
Confidence
99% confidence
Finding

The helper accepts arbitrary absolute HTTP or HTTPS URLs and then adds service tokens, the API key, and user identity without checking the destination. A caller-controlled destination can therefore receive available credentials and request data; plain HTTP also exposes them in transit, which is especially concerning for health-related requests.

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.

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

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser accepts --export-env-only and promises not to analyze, but execution never checks that flag. When supplied with an input, the command still initializes identity and invokes analysis, potentially transmitting sensitive facial footage despite the caller's explicit non-analysis choice.

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 video contents are submitted for API analysis without disclosure of the upload.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The function reads the complete local video into memory and passes its bytes as a file attachment to the analysis service without a disclosure or authorization gate. For facial health monitoring, this crosses a sensitive privacy boundary that a local file-selection action alone may not authorize.

Content

No source excerpt is available for this finding.

The documented export-only mode still executes video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The export-only flag is parsed but never consulted, so supplying a video still reaches analyze_video and the analysis service. This violates an explicit non-analysis mode and can cause unintended disclosure of sensitive footage or chargeable service activity.

Content

No source excerpt is available for this finding.

The skill includes persistent user-account and authentication-token management unrelated to camera-based vital-sign analysis.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

The User model stores account identifiers, email addresses, and authentication fields named token and open_token (L454–L461), while UserDao exposes the inherited database CRUD operations (L470–L473). Managing persistent account credentials is not a direct requirement of estimating vital signs from camera footage and is not declared in the supplied manifest.

Content

No source excerpt is available for this finding.

ai_chat documents an agent invocation but never executes it or produces analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L26 explicitly says this method invokes the openclaw agent command through subprocess. However, the invocation is commented out, and L68 substitutes an empty dictionary whose stderr access immediately raises an exception; the handler suppresses that exception and the method returns None. This contradicts the documented analysis operation rather than merely omitting an implementation detail.

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
96% confidence
Finding

At L101, opening a path in 'w' mode immediately erases any existing contents. The generic open method has no docstring, explanatory comment, confirmation, or action log warning callers about this irreversible effect; exception tracing only runs if opening fails.

Content

No source excerpt is available for this finding.

API requests can silently register a remote user account, beyond the declared vital-sign analysis purpose.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

An ordinary API request can trigger remote registration with silent=1 and register=1, transmitting the selected identity without a separate enrollment step. In a health-analysis skill, undisclosed account creation creates a meaningful privacy and consent risk, although no resulting charge is demonstrated.

Content

No source excerpt is available for this finding.

HTTP requests can silently register a remote account without disclosure.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The generic request path can silently create a remote account rather than merely perform the requested analysis. This is the same consent and privacy issue identified by SDI-1, not an independent additional exploit.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This POST sends the selected username as both openId and mobile to the configured remote login/registration service as part of automatic enrollment. The concern is undisclosed identity transmission, not networking itself; the shown payload does not establish transmission of camera footage or vital-sign measurements.

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"):

Loading configuration silently creates directories and writes a YAML file when the file is missing.

Low
Category
Output Handling
Confidence
94% confidence
Finding

YamlUtil.load modifies the filesystem without a confirmation, log message, docstring, or explanatory comment disclosing this behavior. Because BaseEnum.init_subclass calls this loader during class definition (L044–L059), merely importing this module can create configuration files, an unexpected side effect of loading configuration.

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:

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