Back to skill

Security audit

Reptile Pet Health Diagnosis Tool | 爬行类宠物健康诊断分析工具

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because scanners report several privacy, security, and scope concerns beyond its stated reptile-video health analysis purpose.

Consider the reported risks before installing, especially possible credential disclosure, shared access to report history, and video transmission without clear notice. The overall verdict is suspicious, not malicious; the scan was marked partial because a referenced local path was missing, so the evidence does not support treating the skill as fully reviewed or safe.

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/crawl_analysis.py
Finding

Implicit identity resolution discards the sender identity and reuses a shared account

Content
View full analysis

Impact

A second sender can receive the first sender's historical analysis records and any report links or media references returned in those records. Their submissions can also become mixed in the same cloud history.

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
98% confidence
Finding

Video analysis and report output match the declared purpose. However, persistent camera monitoring and camera registration management extend materially beyond analyzing user-provided videos. Historical report retrieval and its explicit trigger phrases are also absent from the description. The supplied chunk contains command examples rather than implementation, so it establishes exposed command workflows but does not verify their underlying execution.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The supplied chunk's substantive functionality is user-account database management. It defines a sys_user table containing names, email, birthday, demographic fields, and tokens, and exposes methods to read and modify those records. It contains no video handling, health-check API invocation, disease analysis, or report generation. Although persistence can support a diagnostic application, this chunk implements a substantial account-management capability not represented in the declared description, and its connection to the stated diagnostic workflow is not shown. No token transmission or other exfiltration is demonstrated.

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: "crawl-analysis"
description: "Triggers diagnostic analysis when users provide video URLs or files for reptiles such as lizards, snakes, and spiders. Supports uploading local videos or network URLs to call the server-side API for health checks. It analyzes features like scales, skin, and body appearance to identify potential diseases and generate a Pet Safety Guardian health report. | 爬行类宠物健康诊断分析工具,当用户提供蜥蜴、蛇、蜘蛛等爬行宠物的视频 URL 或文件需要分析时,触发本技能进行爬行宠物健康诊断分析;支持通过上传本地视频或网络�

The HTTP helper can send service credentials to arbitrary external destinations, beyond the server-side health API required by the skill.

High
Category
Excessive Agency
Confidence
99% confidence
Finding

The helper accepts arbitrary absolute HTTP or HTTPS URLs and attaches service access tokens, API keys, and authorization tokens without checking the destination. A caller-controlled destination can therefore receive those credentials directly; supporting network video URLs does not justify authenticating to arbitrary media hosts.

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.

History-report triggers include generic phrases without requiring reptile-health context.

Medium
Category
Trigger Abuse
Confidence
94% confidence
Finding

Generic phrases such as querying historical reports automatically trigger this skill's cloud-history request without requiring reptile-health context. An unrelated request could consequently retrieve and display private pet-health reports, especially because identity selection is automatic.

Content

No source excerpt is available for this finding.

The skill exposes continuous camera monitoring and camera management beyond its declared video-submission analysis scope.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

The manifest at L003 describes diagnosing user-provided video files or URLs. However, L185–L192 instruct invoking commands to add or remove monitoring cameras by serial number and query 24/7 monitoring history, introducing ongoing surveillance and device management rather than analysis of submitted media.

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 that --export-env-only will not perform analysis, but execution never checks that flag. When an input is supplied, the normal analysis path still runs, potentially uploading private video despite the caller's explicit non-analysis selection.

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 disclosing the upload to the user.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The code reads the entire local file, packages it as an upload, and passes it to the API analysis method. There is no confirmation, user-facing notice, or comment explaining that the video is transmitted for remote processing; requesting analysis of a local file does not itself disclose this privacy-relevant behavior.

Content

No source excerpt is available for this finding.

The documented environment-export-only mode still performs video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The --export-env-only argument promises not to execute analysis, but its parsed value is never consulted. With a video input present, execution proceeds to the analysis helper, potentially transmitting private media when the caller intended only environment export.

Content

No source excerpt is available for this finding.

The documented workspace-only database restriction is not enforced for paths containing parent-directory components.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

Both containment checks use a textual prefix without resolving parent-directory components or symbolic links. An absolute path such as /data/../outside.db passes these checks and lets SQLite open or create a database outside the intended data directory if an attacker can influence db_path.

Content

No source excerpt is available for this finding.

The generic file-opening helper silently truncates existing files.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

FileUtil.open opens every path in write mode, which immediately destroys existing contents even before anything is written. There is no docstring, comment, confirmation, or action-specific log disclosing this destructive behavior; exception tracing does not disclose successful truncation.

Content

No source excerpt is available for this finding.

HTTP requests can silently register a remote account and persist authentication tokens.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

An ordinary HTTP request can trigger a login request with silent registration enabled and persist the resulting authentication tokens through the DAO. Although account provisioning may support the health service, hiding remote account creation and credential persistence inside a generic request helper creates undisclosed privacy and authorization side effects.

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"):

Loading configuration silently creates directories and files, including during module import.

Low
Category
Output Handling
Confidence
92% confidence
Finding

YamlUtil.load creates a missing directory and writes a configuration file without a prompt, log, or explanatory comment or docstring. BaseEnum.init_subclass invokes this loader when the enum classes are defined (L044–L059), so merely importing the module can modify the filesystem without disclosing that side effect.

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 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/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()

The ai_chat docstring claims subprocess-based agent invocation, but the implementation never invokes the agent.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L26 claims that this method invokes openclaw agent through subprocess and returns an analysis result or session ID at L33–L34. However, command construction and subprocess execution are commented out (L48–L66); the replacement empty dictionary at L68 causes attribute access at L70 to fail, and the exception handler exits without returning a 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