Back to skill

Security audit

Dietary Behavior Health Analyzer | 饮食行为健康分析工具

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because scanners report behavior that exceeds or conflicts with its dietary video analysis purpose, including remote account and credential handling and a no-analysis option that may still allow analysis.

Consider the verdict's Review status carefully before installing, especially if you will provide private videos or health-related information. The scan was partial: it reports one missing local reference, so some evidence could not be assessed. These results indicate concerns, not a finding that the author is malicious.

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 (22)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The visible implementation does not substantiate the declared video-based dietary analysis capability. Its explicit helper cannot produce results because command execution is disabled and result handling raises an exception. Imported base classes and API services are not supplied, so capabilities potentially implemented there cannot be assessed. There is no evidence here of active subprocess execution or malicious resource access.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

Ordinary file, formatting, polling, and HTTP helpers can legitimately support the declared analysis and are not themselves mismatches. However, RequestUtil also invokes /sys/phoneLogin with silent=1 and register=1 when credentials are missing, and saves returned user and token information locally. This persistent account provisioning is not disclosed in the description. Additionally, absolute URLs bypass the configured base URL while still receiving authentication headers, creating a credential-forwarding capability beyond a bounded dietary-analysis integration. These behaviors occur when the relevant methods are called; this chunk does not establish that they execute during every analysis, nor does it prove malicious exfiltration. The absence of analysis logic alone is not decisive because this is a shared utility module.

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: "diet-analysis"
description: "Analyzes videos to evaluate human eating behaviors, habits, and dietary patterns. It identifies tendencies towards unhealthy eating and provides structured analysis reports along with nutritional improvement recommendations. | 饮食行为健康分析工具,针对人的饮食行为、进食习惯、饮食结构进行视频分析,识别不良饮食行为倾向,提供结构化分析报告和营养改善建议"
version: "1.0.21"
license: "MIT-0"
---

# 🍽️ Dietary Behavior Health Analyzer | 饮食行为健康分析工具
> **智能分析中枢** · 图片/视频智能分�

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 documented export-only mode does not prevent video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser accepts --export-env-only, but execution never checks it before initializing identity and invoking analysis or history retrieval. With an input supplied, a user relying on the explicit no-analysis promise can unintentionally submit sensitive eating videos for cloud processing.

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 to an API without disclosure of the upload.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The code reads the entire local input file and passes its contents to the analysis API through the files argument. There is no confirmation, log, or explanatory comment disclosing this transmission; requesting video analysis does not necessarily imply consent to upload potentially private footage.

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 --export-env-only argument is accepted but never used to control execution. Supplying it alongside a video still reaches the analysis path, violating the promised no-analysis boundary and potentially causing unintended cloud submission of private footage.

Content

No source excerpt is available for this finding.

The skill includes system-user and authentication-token persistence unrelated to dietary video analysis.

Medium
Category
Excessive Agency
Confidence
88% confidence
Finding

The User model stores authentication material in token and open_token fields, and UserDao inherits methods to read, overwrite, and delete these system-user records. DAO initialization also automatically creates tables and alters sys_user (L180–L215), extending the capability beyond storing dietary observations or reports into account and credential management that the manifest does not declare.

Content

No source excerpt is available for this finding.

FileUtil.open silently truncates existing files without disclosing its destructive write mode.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The generically named open helper always uses mode 'w', which immediately erases an existing file's contents. No docstring, comment, confirmation, or user-facing log explains this destructive behavior; exception tracing only reports failures.

Content

No source excerpt is available for this finding.

Analysis requests can silently register a remote user account and persist its credentials, beyond the declared dietary-analysis purpose.

Medium
Category
Excessive Agency
Confidence
91% confidence
Finding

An ordinary HTTP request can trigger a remote login-or-registration request with register enabled and then persist returned authentication tokens locally, without an explicit enrollment decision in this flow. This creates a remote identity and durable credentials as a hidden side effect of dietary analysis, which is especially privacy-sensitive when subsequent use concerns eating habits or health-related information.

Content

No source excerpt is available for this finding.

HTTP requests can silently register a remote account and persist authentication tokens without disclosure.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The request helper implicitly enrolls or logs in a user and saves returned tokens when credentials are absent, without a consent gate in the shown flow. This is the same underlying privacy and credential-lifecycle issue identified by SDI-1, not a separate exploit.

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 · 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"):

The history-query example includes the generic phrase “历史报告” (“historical reports”) as an automatic trigger without requiring dietary context.

Low
Category
Trigger Abuse
Confidence
94% confidence
Finding

The example makes the generic phrase “historical reports” an automatic trigger, while other instructions require immediate cloud retrieval and display. An unrelated report request could therefore cause unintended retrieval and disclosure of sensitive dietary history.

Content

No source excerpt is available for this finding.

Configuration loading silently creates directories and files, including during module import.

Low
Category
Output Handling
Confidence
93% confidence
Finding

YamlUtil.load creates parent directories and writes a YAML file when the requested configuration is missing, without a log, confirmation, or explanatory comment/docstring. BaseEnum.init_subclass calls this loader during class definition (L044–L059), so merely importing the module can modify the filesystem without disclosure.

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 documents agent execution, but the implementation never invokes the agent.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L26 explicitly says the method invokes openclaw agent through subprocess. However, command construction and execution are commented out (L48–L66); the empty dictionary at L68 instead causes an AttributeError at L70, which is caught and suppressed, so the method returns no analysis result.

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