T08 · Insecure Dependencies
- Location
SKILL.md:32- Finding
Unpinned Third-Party Installer Executed Through npx
- Content
View full analysis
- Remediation
View remediation
Security audit
Security checks for vulnerabilities and agentic risk
This is a straightforward neon image generator, with notable but disclosed risks around API token handling and install provenance.
Prefer the ClawHub install path or a pinned, reviewed source instead of a floating npx command. Use a limited Neta token, run the script only on trusted machines, and assume your prompt, optional reference UUID, and token are sent to api.talesofai.com for generation.
SKILL.md:32Unpinned Third-Party Installer Executed Through npx
neonartgenerator.js:8API Token Exposed Through Command-Line Arguments
The README instructs users to install and run a package via npx skills without pinning a specific version. This allows whatever version is current at execution time to be fetched and executed, which creates a supply-chain risk if the upstream package is compromised, replaced, or updated with malicious behavior.
Referencing npx skills again without a pinned version repeats the same supply-chain exposure: users may execute a different package version than the author reviewed. In security-sensitive environments, dynamic retrieval and execution of latest code materially increases the chance of compromise.
The skill declares only the Bash tool and omits any explicit permission or allowed-tools scoping, even though its documented behavior depends on network access to the Neta API. That mismatch can cause the agent runtime to grant broader-than-necessary capabilities or leave network use insufficiently constrained, increasing the risk of unintended outbound requests or abuse if the skill is modified or invoked unexpectedly.
Using npx skills add blammectrappora/neon-art-generator without pinning a specific version introduces a supply-chain risk: future installs may pull a changed or compromised package. An attacker who gains control of the upstream package or a later release could distribute malicious code to users who follow the installation instructions.
The hardcoded external domain indicates outbound communication to api.talesofai.com, which differs from the skill branding and token acquisition domain shown to the user. That mismatch makes the data flow more dangerous because it can mislead users into trusting one provider while actually sending content and authentication material to another service.
};
}
const res = await fetch('https://api.talesofai.com/v3/make_image', {
method: 'POST',
headers: HEADERS,
body: JSON.stringify(body),
The hardcoded external domain indicates outbound communication to api.talesofai.com, which differs from the skill branding and token acquisition domain shown to the user. That mismatch makes the data flow more dangerous because it can mislead users into trusting one provider while actually sending content and authentication material to another service.
};
}
const res = await fetch('https://api.talesofai.com/v3/make_image', {
method: 'POST',
headers: HEADERS,
body: JSON.stringify(body),
The polling request repeatedly transmits the same authentication header to the external service and confirms continued interaction with the third-party API. While polling itself is normal functionality, it extends the exposure of credentials and task metadata to an external domain and reinforces the undisclosed cross-service data sharing risk in this skill.
for (let attempt = 0; attempt < maxAttempts; attempt++) {
await new Promise(r => setTimeout(r, 2000));
const pollRes = await fetch(`https://api.talesofai.com/v1/artifact/task/${taskUuid}`, {
headers: HEADERS,
});