Back to skill

Security audit

Weather Alert

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a normal weather-alert tool, but it needs Review because its automated alert configuration includes an unexplained execution-style notification method and its install and persistence disclosures are not tight enough.

Review before installing. Use a pinned or trusted installer, understand that location data is sent to third-party weather services, and avoid enabling `exec-event` or automated checks unless the host clearly shows what will run, how often it runs, where state is stored, and how to turn it off.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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 (2)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:12
Finding

Unpinned npm Package Is Downloaded and Executed Through npx

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:12-15; duplicated in README.md:14-17
Vulnerability Type: Unpinned executable dependency
Risk Level: Medium

Vulnerable code:

bash
# Install the skill
npx clawhub install weather-alert

Technical Analysis

The installation instructions invoke clawhub through npx without specifying an exact reviewed version or integrity value. If the package is not already available locally, npx can retrieve the currently resolved package from the configured npm registry and execute its command-line entry point.

The effective installation code can therefore change after this skill has been audited. A compromised package release, registry account, dependency, or package-resolution configuration could cause users following the documented installation procedure to execute code that was not included in the reviewed repository.

Attack Path

  1. An attacker compromises the npm package, one of its transitive dependencies, its publisher account, or the registry resolution path.
  2. The attacker publishes or serves a malicious version under the package name resolved by npx.
  3. A user follows the documented npx clawhub install weather-alert instruction.
  4. npx downloads the mutable package version and invokes its executable.
  5. Malicious package code executes with the privileges and environment of the user running the command.

Impact Assessment

Successful exploitation could provide arbitrary code execution under the installing user's account. The resulting scope could include reading or modifying files accessible to that user, accessing environment variables and developer credentials, making network requests, and installing additional user-level components. Administrator-level impact would require the user to run the command with elevated privileges or an independent privilege-escalation path.

Remediation
View remediation

Remediation Suggestions

  • Pin clawhub to an explicitly reviewed version, for example:
    bash
    npx --yes clawhub@<reviewed-version> install weather-alert
    
  • Record and verify package integrity through a lockfile or another reproducible installation mechanism.
  • Document the expected publisher and registry.
  • Prefer a locally installed, lockfile-managed CLI rather than downloading an executable during each installation.
  • Avoid recommending elevated privileges for this command.
  • Establish a dependency-update process in which version changes are reviewed and tested before documentation is updated.

T08 · Insecure Dependencies

Note
Location
README.md:24
Finding

PyYAML Dependency Is Installed Without Version or Integrity Pinning

Content
View full analysis

Vulnerability Details

File Location: README.md:24-29
Vulnerability Type: Unconstrained third-party dependency installation
Risk Level: Low

Vulnerable code:

bash
# Clone and install
git clone https://github.com/openclaw/skills.git
cd skills/weather-alert
pip install pyyaml  # optional, for config.yaml support

Technical Analysis

The installation command requests pyyaml without an exact version, lockfile, or cryptographic hash. Consequently, the package and artifacts installed can change over time without any corresponding change to this repository.

If the package publisher, package index, account, build artifact, or package-resolution configuration is compromised, pip may install an unreviewed release. A malicious source distribution could execute code during its build process, while malicious installed package code could execute when weather_alert.py imports yaml.

Attack Path

  1. An attacker compromises the PyYAML distribution channel, publisher credentials, package index, or the user's configured pip index.
  2. A malicious artifact is made available as the version selected by the unconstrained requirement.
  3. A user runs pip install pyyaml as instructed.
  4. Malicious build behavior may execute during installation, or malicious module code executes when the weather script imports yaml.
  5. The payload operates with the privileges of the user who performed the installation or launched the script.

Impact Assessment

Exploitation could result in arbitrary code execution in the installation or runtime context. The attacker could access user-readable files, environment variables, local application data, and network resources or modify files writable by that user. System-wide impact would depend on whether pip was run in a privileged or system-level environment.

Remediation
View remediation

Remediation Suggestions

  • Declare an exact reviewed PyYAML version in a dependency file.
  • Generate and verify cryptographic hashes, such as with a hash-locked requirements file:
    text
    PyYAML==<reviewed-version> --hash=sha256:<verified-hash>
    
  • Install with hash enforcement:
    bash
    python3 -m pip install --require-hashes -r requirements.txt
    
  • Use a dedicated virtual environment rather than installing into a global interpreter.
  • Restrict installation to the official, trusted package index and review dependency updates before changing the pinned version.
  • Prefer pre-reviewed binary artifacts where appropriate and avoid running package installation with administrator privileges.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (12)

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The README instructs users to run npx clawhub install weather-alert without pinning an exact package version. npx will fetch and execute the latest published package at install time, which creates a supply-chain risk if the package is compromised, typosquatted, or a malicious version is later published. In this context, the command appears in a quick-start path, which increases the chance that users will copy-paste and execute it directly.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill description emphasizes convenience features but does not clearly warn that it performs ongoing background monitoring, repeated network access, and local state writes for alert tracking. Users may enable it without understanding that it persists data and continues operating over time, which weakens informed consent and increases the chance of unsafe deployment in sensitive environments.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The installation command uses npx clawhub without pinning a specific version, which allows whatever package version is current at execution time to run on the user's system. If the package is compromised, typosquatted, or a malicious update is published, users could execute attacker-controlled code during install.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The configuration states notification.method: "exec-event", which strongly implies alert delivery may invoke an executable or system-level event handler rather than a passive notification channel. In a skill that performs recurring checks and triggers on external weather data, this creates a code-execution surface that could be abused if alert parameters, handlers, or downstream integrations are influenced by untrusted input.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Documenting exec-event as the notification method without a prominent warning obscures that alert delivery may execute system-level actions. Because this skill is designed for automated, periodic triggering based on remote data, undisclosed execution-capable notifications materially increase the risk of unintended or attacker-influenced command execution.

Content

No source excerpt is available for this finding.

Scope Creep

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The documentation lists OpenWeatherMap as a fallback data source, but the declared network permissions only include wttr.in and open-meteo.com. This mismatch can lead to undeclared outbound connections or pressure to broaden permissions later without clear review, undermining least-privilege expectations and making network behavior less transparent to users.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · references/weather-sources.md (reported line 43)May include surrounding context.

md
## Tertiary: OpenWeatherMap

**URL:** `https://api.openweathermap.org/data/2.5/weather`
**Cost:** Free tier (1000 calls/day)
**Requires:** API key
**Best for:** Historical data, detailed forecasts

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · references/weather-sources.md (reported line 5)May include surrounding context.

md
def fetch_open_meteo(lat, lon, days=7):
    """Fetch weather data from Open-Meteo API (free, no key needed)."""
    url = "https://api.open-meteo.com/v1/forecast"  # noqa: E501
    url += "?latitude=" + str(lat) + "&longitude=" + str(lon)
    url += "&current=temperature_2m,relative_humidity_2m,apparent_temperature,"
    url += "precipitation,weather_code,wind_speed_10m,wind_direction_10m,"

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · scripts/weather_alert.py (reported line 81)May include surrounding context.

python
def fetch_open_meteo(lat, lon, days=7):
    """Fetch weather data from Open-Meteo API (free, no key needed)."""
    url = "https://api.open-meteo.com/v1/forecast"  # noqa: E501
    url += "?latitude=" + str(lat) + "&longitude=" + str(lon)
    url += "&current=temperature_2m,relative_humidity_2m,apparent_temperature,"
    url += "precipitation,weather_code,wind_speed_10m,wind_direction_10m,"

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The request to the weather API forces the timezone to Europe/Prague, which imposes a specific locale behavior regardless of the user's actual region or preference. This is a natural-language/policy concern because the skill does not document this constraint as region-specific or provide an opt-in choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The document sets a default timezone of Europe/Prague, which introduces a locale-specific default in natural language without explaining why that locale is appropriate for all users. Although alternatives are listed, the file does not state that users can choose or opt in to a different timezone, so it may conflict with locale-choice policy expectations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

This code sends location data (latitude/longitude and user-provided location names) to third-party weather services via HTTP requests. Although comments name the services, there is no confirmation prompt or user-facing disclosure in runtime output warning that user-supplied location information will be sent externally.

Content

No source excerpt is available for this finding.