Back to skill

Security audit

Fish Aquatic Pet Health Diagnosis Analysis Tool | 鱼类水族宠物健康诊断分析工具

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this aquatic-pet video analysis skill suspicious because scanners report undisclosed account and personal-data handling, overly broad history triggers, and behavior that may still upload video when export-only mode is selected.

Consider the suspicious verdict and the reported HTTP transmission, video-upload behavior, unrelated history handling, and account and credential storage before installing. The scan was partial: a local reference in SKILL.md could not be resolved, so the evidence does not cover that referenced item.

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
skills/smyx_common/scripts/config-dev.yaml
Finding

Development mode transmits authentication tokens and uploaded videos over HTTP

Content
View full analysis

Impact

A network-path attacker can capture bearer tokens and video contents, alter analysis traffic, and potentially replay captured tokens to access the affected account within their validity and privileges.

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 (23)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The video-analysis commands match the declared purpose, and report formatting and file output are supporting features. However, historical-report listing is a separate user-facing capability with explicitly documented triggers outside the declared video-input workflow. The supplied chunk only invokes a Python module; it does not reveal the underlying API implementation or establish any additional resource access.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
86% confidence
Finding

The supplied chunk implements user-account storage rather than video processing, aquatic-health API requests, disease analysis, or report generation. Database access can be supporting infrastructure, but this chunk specifically exposes account and credential persistence not described in the declared purpose, and no connection to the diagnostic workflow is shown. This finding is limited to the supplied code; it does not establish that these fields are populated, that data is transmitted, or that the broader skill lacks the advertised functionality.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
86% confidence
Finding

The mandatory history-query instructions direct the agent to scripts.autism_analysis rather than the aquarium module. If that module is available, this can route a request into an unrelated, potentially sensitive report domain; otherwise it fails, and the provided evidence does not establish actual cross-domain disclosure.

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: "aquarium-analysis"
description: "When a user provides a video URL or file of aquatic pets such as goldfish, koi, betta, shrimp, crab, etc. for analysis, this skill is triggered to perform aquatic pet health diagnosis analysis. Supports uploading local videos or online video URLs, calls server-side API for aquatic pet health examination, analyzes features such as scales, fins, body color, activity level, identifies potential diseases and outputs a pet health report. | 鱼类水族宠物健康诊断分析工具,当用户提供金鱼、锦鲤、斗鱼、虾、蟹等水族宠物的视频 URL 或文件需要分析�

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.

Automatic history-query triggers include generic phrases unrelated to aquarium health.

Medium
Category
Trigger Abuse
Confidence
93% confidence
Finding

Generic phrases such as requests to query historical reports trigger a mandatory cloud lookup without requiring aquarium context. In a multi-skill conversation, this can cause unintended access to and display of aquarium history when the user meant another report category.

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 parser advertises --export-env-only as preventing analysis, but the flag is never checked. When an input is supplied, execution still reaches the analysis workflow, potentially uploading a local video or initiating remote processing despite the caller explicitly selecting a no-analysis mode.

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

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The code reads the entire local input file, packages its contents as a multipart file payload, and passes it to the analysis service. No prompt, log, docstring, or comment in this file explains the remote transmission; the comment about reading the file does not disclose uploading potentially sensitive video content.

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 --export-env-only option promises that analysis will not execute, but its parsed value is never inspected. With a video input, the command still calls analyze_video, violating the caller's explicit no-analysis choice and potentially transmitting media for cloud processing.

Content

No source excerpt is available for this finding.

The skill includes persistent management of personal demographic data unrelated to aquatic-pet video analysis.

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

The User model includes real names, email addresses, birthdays, sex, and age, and UserDao exposes persistent CRUD operations for these records (L470–L473). DAO initialization creates the schema (L180–L190), making this an implemented personal-profile storage capability rather than documentation alone. Although a user identifier or API token could support server-side diagnosis, managing these demographic attributes has no evident justification in the stated video-analysis workflow.

Content

No source excerpt is available for this finding.

FileUtil.open silently opens files in write mode, truncating existing contents without disclosure.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

At L101, opening a path with mode 'w' immediately truncates an existing file. This generically named helper has no docstring, explanatory comment, confirmation, or user-facing log disclosing that destructive behavior; exception tracing does not disclose successful truncation.

Content

No source excerpt is available for this finding.

API requests can silently register a remote user account, beyond the declared video-analysis workflow.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

An ordinary API request can automatically invoke phoneLogin with silent=1 and register=1, transmitting an identity and requesting remote account creation without an explicit provisioning step. Server-side video analysis explains network access, but does not by itself establish informed authorization for creating a persistent account.

Content

No source excerpt is available for this finding.

HTTP requests can silently trigger remote user registration and local credential persistence.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

The HTTP helper combines routine requests with automatic remote registration and saving returned account information, token, and openToken through UserDao. These hidden persistent side effects expand a video-analysis request into account provisioning and credential retention without a visible consent or retention boundary.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

External requests are expected for this skill, but the registration payload sends the selected identity as both openId and mobile. The caller prioritizes ApiEnum.API_SECRET_KEY as that identity, so when tokens are unavailable a configured secret can be copied into account-profile fields and exposed to server-side profile storage or logging beyond its intended authentication use.

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 missing configuration files, including during module import.

Low
Category
Output Handling
Confidence
91% confidence
Finding

YamlUtil.load creates parent directories and writes a YAML file when the requested path does not exist, without a prompt, log, docstring, or explanatory comment disclosing this side effect. BaseEnum.init_subclass calls this loader at L053 and L058, so defining the module's configuration classes can modify the filesystem merely by importing the module, rather than through an explicit save operation.

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:

The ai_chat docstring promises an agent invocation, but the implementation never invokes an agent and silently returns no result.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L26 states that ai_chat calls the openclaw agent through subprocess, and L33–L34 promise an analysis result or session ID. However, the subprocess invocation is commented out at L61–L66; the empty dictionary at L68 causes result.stderr at L70 to raise AttributeError, which is caught at L81–L82, leaving the method to return None without performing 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