Back to skill

Security audit

Respiratory Symptom Smart Recognition Tool | 呼吸道症状智能识别工具

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a suspicious skill because its visible account, credential, and network behavior appears broader than its stated respiratory-analysis purpose, with privacy and identity-handling concerns.

Consider avoiding installation unless you can verify the skill's actual behavior and are comfortable with its account and credential handling and possible remote processing of health media. The scan is marked partial because a referenced local path was missing, so the evidence does not establish that all behavior has been assessed.

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

Error
Location
scripts/respiratory_symptom_recognition_analysis.py
Finding

Automatic identity initialization discards the upstream sender identity

Content
View full analysis

Impact

A later user can receive health-analysis records created under the shared fallback account by earlier users. New analyses are also attributed to that shared account.

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
95% confidence
Finding

The visible implementation does not substantiate the declared respiratory-monitoring purpose. It is generic skill/API scaffolding with a nonfunctional chat helper. No subprocess execution or other external-agent call actually occurs in this chunk. Imported base classes and API services are not supplied, so their behavior cannot be assessed; the mismatch finding is limited to the visible code and does not establish malicious activity.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

Basic formatting, filesystem helpers, and HTTP transport can be supporting implementation details and are not independently grounds for a mismatch. However, the supplied code also implements an undisclosed account-provisioning and credential-handling workflow, including transmitting local identity information to a remote login/registration endpoint. It accepts absolute request URLs and adds authentication headers without a visible host check. Those behaviors extend beyond the declared symptom-recognition purpose. No computer-vision processing, respiratory-event counting, or health-alert logic is present in this chunk, although such functionality could exist elsewhere.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The supplied chunk implements user-account persistence rather than respiratory-symptom analysis. Generic database plumbing could support the declared skill, but this code specifically manages identity, demographic, and token data without any demonstrated connection to symptom detection or episode tracking. It contains no computer-vision processing, symptom classification, episode-frequency calculation, or health-alert logic. The mismatch is limited to the supplied chunk; no network transmission or actual token collection is shown.

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: "respiratory_symptom_recognition_analysis"
description: "Based on computer vision, automatically detects coughing, phlegm, and wheezing frequency, counts the frequency of episodes, used for early health anomaly alerts, helping to detect respiratory diseases in a timely manner. | 呼吸道症状智能识别技能,基于计算机视觉自动检测咳嗽、咳痰、喘息频率,统计发作频次,用于健康异常早期提醒,帮助及时发现呼吸道疾病"
version: "1.0.26"
license: "MIT-0"
---

# 🫁 Respiratory Symptom Smart Recognition Tool | 呼吸道症状智能识别工具
> **智能分析中枢** · 图片/视频智�

The generic HTTP helper attaches service credentials to caller-supplied absolute URLs without restricting their destination (L572-L573, L610-L647).

High
Category
Excessive Agency
Confidence
99% confidence
Finding

Caller-supplied absolute URLs retain their destination while the helper automatically adds service tokens and the API key. If an attacker can influence the URL, they can receive these credentials at their own endpoint; HTTP destinations additionally expose them to network interception, potentially compromising health-service accounts.

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.

The analysis workflow lacks a user-facing privacy warning about remote processing of sensitive health media.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

Although API use is mentioned, the workflow does not clearly warn users before sending identifiable health footage to an external processor or explain retention and access. Automatic execution can therefore disclose sensitive medical and household media without informed authorization.

Content

No source excerpt is available for this finding.

Cloud analysis of sensitive health videos lacks a privacy warning.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

The documentation acknowledges cloud analysis but does not explain the privacy consequences of transmitting identifiable respiratory-health footage. Medical disclaimers do not substitute for a disclosure of recipients, storage, retention, and user control over sensitive media.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to run analysis, but execution ignores it and proceeds with analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser accepts --export-env-only and promises no analysis, but execution never checks the flag. When supplied with an input, a caller expecting a configuration-only operation instead triggers analysis and potentially transmits sensitive health media.

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
94% confidence
Finding

The method reads the entire local video into a multipart payload and passes it to the analysis API without a visible disclosure or consent check. In this health-monitoring context, users expecting local processing could unintentionally expose identifiable medical footage.

Content

No source excerpt is available for this finding.

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

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The --export-env-only flag is parsed but never acted upon, so supplying an input still invokes video analysis. This violates an explicit no-analysis expectation and can cause unintended remote processing of private footage.

Content

No source excerpt is available for this finding.

save() documents record creation but can silently overwrite an existing record.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

Any insertion exception triggers an update by ID and potentially a fallback matching either username or realname, copying fields including identifiers and credentials into an existing user. This can turn failed creation into unintended account modification; exploitation depends on influence over the supplied model, and account corruption could compromise health-data access.

Content

No source excerpt is available for this finding.

The skill includes account and credential management beyond respiratory episode analysis.

Medium
Category
Excessive Agency
Confidence
88% confidence
Finding

The User model stores identity details and token/open_token credentials (L454–L461), and UserDao exposes the inherited database CRUD operations for these records. Persisting symptom counts is an expected implementation detail, but managing account credentials in the shared database is not justified by the manifest's respiratory-recognition and alerting 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
98% confidence
Finding

Opening a path in 'w' mode immediately erases any existing contents, even before the caller writes anything. This generic open helper provides no confirmation, action log, docstring, or comment explaining that behavior; exception tracing does not disclose successful truncation.

Content

No source excerpt is available for this finding.

The request helper silently registers service accounts and persists account credentials, beyond the described respiratory-analysis functionality (L550-L604).

Medium
Category
Excessive Agency
Confidence
95% confidence
Finding

An ordinary HTTP operation can silently register or log in a remote account and persist returned tokens in the local database. This creates undisclosed identity and credential side effects, which are particularly sensitive for a health-related skill whose description does not disclose account provisioning.

Content

No source excerpt is available for this finding.

HTTP requests can silently trigger remote account registration without disclosure.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

The helper sends silent=1 and register=1 to a remote login endpoint as an implicit prerequisite to an unrelated HTTP request. This can create accounts and disclose an identity without an explicit provisioning decision, an important privacy concern in a respiratory-health workflow.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
94% confidence
Finding

This POST transmits the selected identity as both openId and mobile during implicit registration. The selection explicitly prioritizes API_SECRET_KEY, so when token acquisition runs with that value configured, the helper sends the API secret in identity fields rather than limiting it to a documented authentication channel; the endpoint's trust and transport configuration are not provided.

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
91% confidence
Finding

YamlUtil.load creates parent directories and writes a YAML file when the requested configuration is missing, without a prompt, log, docstring, or explanatory comment disclosing this side effect. BaseEnum.init_subclass calls this loader during class definition (L052–L058), so merely importing the module can modify the filesystem.

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 the implementation never invokes it.

Low
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 contents and subprocess invocation are commented out; the replacement dictionary at L68 causes the attribute access at L70 to raise an exception that is caught and suppressed, so the method returns None without performing the documented delegation.

Content

No source excerpt is available for this finding.