Back to skill

Security audit

Image Quality Assessment Analysis Tool | 图像质量检测分析工具

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with medium confidence because scanners report undeclared account and token handling, remote data transmission, and behavior that may analyze or upload media even in export-only mode.

The verdict is suspicious, not a claim that the author is malicious. Consider the reported upload and identity-handling behavior before installing, especially if you would provide sensitive camera footage; the scan is partial, so the evidence does not cover every referenced item.

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 (25)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied chunk contains no camera access, image processing, or quality-detection logic. Its concrete functionality is user and token persistence, including automatic migration of the sys_user table. Although database access can support an image-analysis skill, this chunk specifically manages identity records rather than images, quality measurements, or detection results, and no connection to the declared purpose is shown. The mismatch concerns the supplied code; it does not establish that the full skill lacks image-quality functionality or that the code exfiltrates data.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

Generic formatting, filesystem, polling, and HTTP helpers can legitimately support an image-analysis skill and are not themselves mismatches. However, automatic account provisioning, persistent identity/token management, and payment-related behavior are substantial capabilities not disclosed by the camera-quality description. The HTTP wrapper also attaches authentication headers even when given an absolute caller-supplied URL, without a visible destination restriction. These behaviors are conditional on helper invocation; this chunk does not establish that they execute automatically or that footage is transmitted. As a shared utility file, it also cannot establish whether other files implement the advertised image analysis.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
87% confidence
Finding

The explicit history-listing mode is a separate user-data retrieval capability not represented in the declared footage-quality detection purpose. Identifier resolution supports that additional capability. Delegating analysis and saving results are compatible implementation details, not mismatches by themselves. Because the actual analysis component is not supplied, this chunk does not establish whether the advertised defect detectors are implemented, nor does it establish what video data is transmitted.

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: "image-quality-detection-analysis"
description: "Detects quality issues in camera footage such as black/white screens, color cast, stripes, noise, and blurriness. Suitable for security surveillance and camera self-check scenarios. | 图像质量检测分析工具,检测摄像头画面出现的全黑、全白、偏色、条纹、雪花、模糊等质量问题,适用于安防监控、摄像头自检等场景"
version: "1.0.23"
license: "MIT-0"
---

# 🖼️ Image Quality Assessment Analysis Tool | 图像质量检测分析工具
> **智能分析中枢** · 图片/视频智能分析 · 结构化报告 · 历史报告云端查�

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 analysis workflow lacks a privacy warning about remote processing of camera images and video.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The workflow automatically analyzes supplied media without clearly warning that local camera images and video are uploaded to an external service. Surveillance footage can contain identifiable people and sensitive locations, making undisclosed remote processing a significant confidentiality risk even though API use is mentioned elsewhere.

Content

No source excerpt is available for this finding.

The documented export-only mode does not prevent analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The --export-env-only flag promises not to execute analysis, but its parsed value is never checked. When media arguments are supplied, normal analysis still runs, potentially uploading sensitive surveillance footage despite the caller explicitly selecting a non-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 file contents are submitted to an analysis API without disclosure of the upload.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The implementation reads the complete local media file and passes its bytes to the analysis API without a disclosure or authorization gate in this path. Although uploading is functional for cloud analysis, camera footage is sensitive and should not cross a local-to-external trust boundary without informed authorization.

Content

No source excerpt is available for this finding.

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

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The export-only option is parsed but never used to stop the normal execution path. Supplying it alongside a video input still invokes analysis, potentially transmitting private footage when the caller explicitly requested no analysis.

Content

No source excerpt is available for this finding.

The service exposes username-based user lookup, unrelated to camera image-quality detection (L098–L101).

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

The get_user_by_username method delegates user lookup to RequestUtil. Looking up user accounts is not an obvious requirement for detecting camera image defects, and the manifest does not declare account-management functionality.

Content

No source excerpt is available for this finding.

The image-quality skill includes unrelated system-user and authentication-token database management.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The User model stores personal information and authentication material, including real names, email addresses, token, and open_token; UserDao exposes inherited read, write, update, and delete operations for these records. DAO initialization also creates tables and migrates sys_user (L180–L215), making this an implemented user-management capability rather than unused field declarations. Managing system identities and credentials is not justified by the manifest's camera image-quality detection purpose.

Content

No source excerpt is available for this finding.

ai_chat documents agent execution and analysis results, but its implementation never invokes the agent and returns no result.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L26 explicitly claims that this method invokes openclaw agent through subprocess, and L33–L34 promise an analysis result or session ID. However, the command and subprocess invocation are commented out; L68 creates an empty dictionary, and accessing result.stderr at L70 raises an exception that is caught at L81–L82, leaving the method to return None without performing the documented analysis.

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
97% confidence
Finding

At L101, opening a path with mode 'w' immediately erases any existing contents. The generically named helper has no docstring, explanatory comment, confirmation, or action log disclosing that behavior; its exception handler does not warn about successful truncation.

Content

No source excerpt is available for this finding.

The request helper silently registers remote user accounts and persists authentication state, beyond the stated image-quality analysis purpose.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

An ordinary HTTP request can trigger silent remote account registration and persist returned authentication tokens without a separate enrollment decision. This introduces undisclosed identity-management side effects beyond image-quality analysis and leaves reusable credentials in a local SQLite database without explicit access protections in this code.

Content

No source excerpt is available for this finding.

HTTP authentication silently transmits a user identifier and requests automatic account registration.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The authentication helper explicitly sets silent=1 and register=1 while transmitting the current identifier as both openId and mobile, without an explicit consent gate. The identifier can originate from configuration or a workspace file rather than a generated anonymous value, making undisclosed identity disclosure and account creation plausible.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The POST sends identifying data to a configured remote login service as an automatic side effect of another request. Its caller prioritizes API_SECRET_KEY as the username, so under the missing-token condition a configured API key can also be copied into the openId and mobile payload fields; remote image analysis does not itself justify this additional disclosure.

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 automatic history-query triggers include generic phrases unrelated to image quality.

Low
Category
Trigger Abuse
Confidence
93% confidence
Finding

Examples include generic phrases such as 'historical reports' that automatically trigger a cloud query without requiring image-quality context. An unrelated request could consequently retrieve and display surveillance-analysis history under the automatically selected identity.

Content

No source excerpt is available for this finding.

Loading configuration silently creates directories and writes missing configuration files.

Low
Category
Output Handling
Confidence
94% confidence
Finding

YamlUtil.load creates parent directories and writes a YAML file when the requested path does not exist, without a warning, log, comment, or docstring disclosing this side effect. Because BaseEnum.init_subclass calls this loader at L053 and L058, importing the module can modify the filesystem without 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 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()

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