Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed video-prompting guide with expected Pruna API examples and no hidden execution, persistence, or destructive behavior.
Before installing, review the publisher and consider pinning or otherwise verifying the npx package references. When using the workflows, treat uploaded images, videos, and audio as data sent to external generation services and avoid submitting sensitive media unless that is intended.
The documentation instructs users to run npx skills add ... without pinning an exact package version or immutable source. That allows future package changes or a compromised upstream release to alter what gets installed, creating a supply-chain risk for anyone following the instructions.
The install command uses npx skills add ... without an exact pinned version, so the resolved package content may change over time. If the package or dependency chain is compromised, users could fetch and execute unexpected code during installation.
This line recommends an npx-based install of a skill package without pinning a fixed version. Because npx executes code fetched from the registry, an attacker controlling or poisoning the upstream package path could deliver malicious code to users.
The command references a package install path but does not constrain the package to a specific immutable version. This creates a supply-chain exposure where later upstream changes can silently affect installations and potentially execute malicious code.
Using npx skills add ... with an unpinned package means the installation behavior is not reproducible and depends on the current registry state. If the package account, release pipeline, or dependency tree is compromised, consumers may execute attacker-controlled code.
This install instruction relies on a floating package reference through npx, which can resolve to different code over time. That exposes users to upstream package compromise or malicious updates when they follow the documented command.
The skill documentation includes an executable install command with no exact version pin. In a security context, documentation that encourages execution of mutable remote code is a real supply-chain weakness even if the author's intent appears non-malicious.
Because the command is an npx install without immutable versioning, users are exposed to whatever the registry serves at execution time. A compromised package publisher or dependency could turn this documentation into a code-execution vector.
The line promotes a remote package install via npx but does not pin the fetched content to a specific version. This increases the chance of accidental or malicious drift and creates a practical supply-chain attack surface for users of the skill.
An unpinned npx install command allows future releases or compromised dependencies to change what code gets executed. Even in a documentation file, that is a genuine vulnerability because users may copy-paste it directly into a shell.
The command relies on mutable upstream package resolution and therefore lacks reproducibility and integrity guarantees. If exploited through package takeover or malicious release publication, it could result in arbitrary code execution during installation.
This documented npx skills add ... command does not pin an exact version, so it may execute different code over time depending on registry state. That creates a supply-chain risk with potential arbitrary code execution if the package path is compromised.
The command uses a floating package reference through npx, which is unsafe from a supply-chain perspective. Users following it may unknowingly install and execute malicious content from a tampered or malicious future release.
Because the install command is unpinned, the resulting package content is not fixed or auditable over time. This opens the door to registry compromise, dependency confusion, or malicious update attacks affecting anyone who copies the command.
This line instructs users to execute remote package installation code without an immutable version pin. In the event of upstream compromise, that could directly lead to arbitrary code execution on the user's machine.
The unpinned npx example creates a supply-chain trust issue because it resolves to whatever package version is current at run time. Documentation-driven command execution is especially risky since users often paste commands without review.
An npx command without exact versioning allows silent drift in installed code and increases exposure to malicious updates. If the package namespace or release process is compromised, users can be induced to run attacker-controlled code.
The package install reference is not immutable, so it can resolve to altered content later. In practice, this is a real vulnerability category because package managers executing remote code are common supply-chain attack targets.
This documentation line instructs execution of a package fetched at runtime without a fixed version pin. A malicious or compromised upstream release could therefore transform a benign install command into a code execution vector.
Because the command uses npx with a floating package reference, the artifact users receive is not stable or guaranteed to match what was reviewed. That is a meaningful supply-chain risk, especially in documentation intended for direct copy-paste execution.
The 'full suite' install command also relies on npx skills add ... without an immutable version pin, broadening the blast radius because it installs a larger package set. If the upstream package is compromised, users may execute malicious installer code affecting multiple components at once.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
{
"prompt": "OPEN: hold wide. MID: slow crane down, neon signs flicker. CLOSE: settle on end pose.",
"image": "https://api.pruna.ai/v1/files/START_ID",
"last_frame_image": "https://api.pruna.ai/v1/files/END_ID",
"duration": 8,
"resolution": "768p",
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
{
"prompt": "OPEN: hold wide. MID: slow crane down, neon signs flicker. CLOSE: settle on end pose.",
"image": "https://api.pruna.ai/v1/files/START_ID",
"last_frame_image": "https://api.pruna.ai/v1/files/END_ID",
"duration": 8,
"resolution": "768p",
The example explicitly specifies 'Italian radio pop' as part of the audio direction, which imposes a language/locale choice in the skill guidance without indicating user preference or opt-in. Under the policy, locale-specific output should either be user-selected or clearly justified as region-specific.
No suspicious patterns detected.