Back to skill

Security audit

Fish Respiratory Rate (Gill Opening / Closing) Monitor | 鱼类呼吸频率(鳃盖开合)监测

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with high confidence because its scanners report account and credential handling, broad data and notification behavior, and controls that may not honor a request to avoid analysis, beyond the stated fish respiratory-monitoring purpose.

Consider avoiding installation unless you can verify the account, credential, file, and remote-service behavior and confirm that export-only mode prevents analysis. The Skillspector scan was partial because one local reference could not be resolved, so the available evidence is incomplete; the suspicious verdict is not a claim that the author acted maliciously.

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/smyx_fish_respiratory_rate_monitor_analysis.py
Finding

Environment-provided sender identity is discarded in favor of a workspace-shared account

Content
View full analysis

Impact

In a shared-workspace deployment, another sender can receive reports produced under the shared account, and new analyses can be attributed to that same account. This is client-side account conflation, not a demonstrated server-side authentication bypass.

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 (26)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied chunk implements generic database and user-management functionality rather than the declared fish-respiration analysis. Persistence could support an analysis system, but this code specifically manages user identities, personal data, and tokens without any visible connection to aquarium observations or respiratory measurements. Those capabilities are not represented in the description. This assessment is limited to the supplied chunk; it does not establish what other skill files implement or imply that stored tokens are exfiltrated.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied chunk contains no camera ingestion, gill-motion detection, respiratory-rate calculation, or hypoxia-threshold logic. Generic utility functions could support such a system, and their presence alone would not establish a mismatch. However, automatic remote account registration, persistent identity management, credential-bearing requests, and payment-service instructions introduce a materially broader account-service workflow not disclosed in the description. These capabilities are conditional on calling the relevant methods; the chunk does not establish that they execute automatically on import or that the complete skill lacks video-analysis code elsewhere.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied chunk does not implement the declared primary purpose. Its only substantive method concerns an inactive agent-command integration and fails before returning an analysis result. There is no active subprocess invocation or evidence of malicious activity. Imported base classes and services are not supplied, so their behavior cannot be assessed; this finding is limited to the visible code.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

The supplied chunk implements a generic API-management interface rather than video analysis, gill-motion detection, respiratory-rate calculation, or threshold-based warnings. Networking could support the declared purpose, but this chunk provides user lookup and broad remote-data operations with no demonstrated connection to aquarium monitoring. Endpoint destinations, authentication, and actual invocation are not shown, so the code does not establish malicious activity or specific data transfers.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
84% confidence
Finding

Generic AI-analysis endpoints could support the declared workflow, and the absence of fish-analysis logic in a configuration-only chunk is not itself a mismatch. However, the explicit health-order report export URL points to a materially different resource with no evident connection to fish respiration monitoring. This supports a mismatch finding, limited to endpoint configuration rather than demonstrated data access or export.

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: "smyx-fish-respiratory-rate-monitor-analysis"
description: "Through fixed cameras on aquariums, the system analyzes fish gill-cover opening / closing motion video, detects periodic gill opening and closing, and calculates respiratory rate (breaths per minute). | 通过鱼缸固定摄像头,分析鱼类的鳃盖开合运动视频,检测鳃盖的周期性开启和闭合,计算呼吸频率(次/分钟)。当呼吸频率超过正常阈值(例如 > 80 次/分钟,具体依品种和水温而定)时,输出'缺氧预警',提示用户检查水质(溶氧量)、水温或鱼的健康状态。"
version: "1.0.23"
license: "MIT-0"

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.

Level 4 alerts notify all household/management contacts without an explicit privacy warning or notification opt-in.

Medium
Category
Output Handling
Confidence
87% confidence
Finding

The Level 4 policy directs notifications to all household or management contacts without defining an authorized recipient list or explicit opt-in. Even for aquarium monitoring, this can disclose facility or household activity and generate unwanted messages; the general deployment-authorization statement does not specifically authorize contact-wide notification.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to execute analysis, but analysis still runs.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser accepts --export-env-only and promises not to analyze, but execution never checks the flag. When valid input is also supplied, the normal analysis path runs, potentially uploading footage and incurring service charges despite the user's explicit non-execution 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 to an analysis API without disclosing the upload.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The local-file path reads the complete video and passes it to the remote analysis operation without an upload-specific notice or consent check. Although the broader workflow mentions an API, users selecting a local file are not clearly told that its contents leave the device, and aquarium footage may include people or private surroundings.

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 parsed but never used to stop execution. Supplying it with input still invokes remote analysis, violating the promised non-analysis mode and potentially causing unintended media disclosure or charges.

Content

No source excerpt is available for this finding.

The service exposes username-based user lookup, which is unrelated to fish respiratory-rate analysis.

Medium
Category
Excessive Agency
Confidence
93% confidence
Finding

At L098–L101, get_user_by_username forwards a supplied username to a user-lookup API. Looking up user accounts is not justified by the manifest's stated functions of analyzing gill movement, calculating respiratory rate, and issuing hypoxia warnings; this file does not show that the capability is invoked during analysis.

Content

No source excerpt is available for this finding.

The skill includes persistent user-profile and credential management unrelated to fish respiratory-rate analysis.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The User model stores personal information and token/open_token values, and UserDao exposes inherited database CRUD operations for those records (L448–L473). DAO initialization also automatically creates tables and migrates sys_user (L180–L215), making this an implemented persistence capability rather than an unused field declaration. Managing human identities and credentials is not justified by the manifest's video-analysis and hypoxia-warning purpose.

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

The generically named open helper always uses mode 'w', which immediately erases an existing file's contents. There is no docstring, explanatory comment, confirmation, or user-facing log disclosing truncation; the exception handler only runs after a failure.

Content

No source excerpt is available for this finding.

The request helper silently provisions remote user accounts and persists authentication state, beyond fish respiration analysis.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

An ordinary request can implicitly invoke phoneLogin with silent registration enabled, obtain authentication tokens, and persist returned account information. This introduces undisclosed remote-account creation and credential-retention side effects beyond the stated fish-analysis task, without an explicit consent gate in this helper.

Content

No source excerpt is available for this finding.

Ordinary HTTP requests can silently transmit identity data to a separate login/registration endpoint.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

The helper sends the selected identity as both openId and mobile to a separate configured login endpoint without an explicit disclosure or consent step. The selected value can be ApiEnum.API_SECRET_KEY, so this hidden side request can disclose a credential as ordinary identity data rather than merely transmitting a generated pseudonym.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
96% confidence
Finding

This POST is a concrete external transmission of identity information during implicit login or registration. Its input can originate from API_SECRET_KEY and is copied into two JSON fields, potentially exposing that secret to the login service and its body logs; the source does not establish that the destination is malicious.

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.

Low
Category
Output Handling
Confidence
93% confidence
Finding

YamlUtil.load creates parent directories and writes a YAML file when the requested path does not exist, without a comment, docstring, log, or confirmation disclosing this side effect. BaseEnum.init_subclass calls this loader during class creation (L052–L058), so merely importing the module can modify the filesystem rather than only read 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:

ai_chat documents invoking an analysis agent, but its implementation never executes the agent or returns analysis output.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L25–L34 says this method invokes openclaw agent through subprocess and returns an analysis result or session ID. However, command construction and execution are commented out, and accessing stderr on the empty dictionary at L68–L70 raises an exception that is caught at L81–L82, causing the method to return None instead.

Content

No source excerpt is available for this finding.