Back to skill

Security audit

Isekai Character Generator

Security checks for vulnerabilities and agentic risk

Overview

This image-generation skill is coherent with its stated purpose, but users should handle the API token carefully because the examples put it on the command line.

Install only if you are comfortable sending your prompt, optional reference UUID, and API token to api.talesofai.com for image generation. Prefer passing the token through a safer mechanism than a literal --token value in shell history, rotate the token if it may have been exposed, and avoid running the unpinned npx install command with elevated privileges.

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

T09 · Insecure Skill Coding Practices

Warning
Location
isekaicharactergenerator.js:26
Finding

API Token Exposed Through Command-Line Arguments

Content
View full analysis

Vulnerability Details

File Location: isekaicharactergenerator.js:26-27, isekaicharactergenerator.js:51-61, SKILL.md:17-21, README.md:60-75
Vulnerability Type: Sensitive credential exposure through process arguments and shell history
Risk Level: Medium

Vulnerable Code

javascript
} else if (a === '--token') {
  tokenFlag = args[++i];
}

The captured token is subsequently placed in the authentication header:

javascript
const res = await fetch('https://api.talesofai.com/v3/make_image', {
  method: 'POST',
  headers: {
    'x-token': token,
    'x-platform': 'nieta-app/web',
    'content-type': 'application/json',
  },
  body: JSON.stringify(body),
});

The documented invocation explicitly places the secret on the command line:

bash
node isekaicharactergenerator.js "your prompt" --token YOUR_TOKEN

Technical Analysis

The Skill requires users to provide the Neta API token through the --token command-line argument. Command-line arguments are not an appropriate secret transport mechanism because they may be exposed through:

  • Shell history files.
  • Process inspection tools while the command is running.
  • Process monitoring and endpoint telemetry.
  • CI/CD logs or verbose task-runner output.
  • Copied terminal commands and support diagnostics.

The token's transmission as an x-token header over HTTPS to the documented api.talesofai.com service is consistent with the Skill's image-generation functionality. The unnecessary risk is the local command-line handling of that token, not its authenticated transmission to the declared service.

The program also sends the user-provided prompt and optional reference UUID to the same external service. This is necessary for remote image generation and is disclosed in the documentation, although users should understand that this content leaves the local system.

Attack Path

  1. A victim follows the d ...[truncated 1266 chars]
Remediation
View remediation

Remediation Suggestions

  1. Read the token from a dedicated environment variable such as NETA_TOKEN instead of requiring --token.
  2. Support protected standard-input entry for interactive use, with terminal echo disabled where practical.
  3. Deprecate and eventually remove the command-line token option.
  4. If backward compatibility requires temporary --token support, display a warning explaining that process arguments and shell history may expose the token.
  5. Never include the token in errors, debug output, telemetry, or generated URLs.
  6. Update all examples to use a safer pattern:
bash
export NETA_TOKEN="your-token"
node isekaicharactergenerator.js "your prompt"
  1. Document token rotation and immediate revocation procedures for potentially exposed credentials.
  2. Add a concise privacy notice explaining that prompts and optional reference UUIDs are transmitted to api.talesofai.com.

T08 · Insecure Dependencies

Note
Location
README.md:9
Finding

Unpinned Third-Party Package Execution in Installation Instructions

Content
View full analysis

Vulnerability Details

File Location: README.md:9-11, SKILL.md:31-34
Vulnerability Type: Mutable and unpinned package execution
Risk Level: Low

Vulnerable Code

bash
npx skills add blammectrappora/isekai-character-generator

Technical Analysis

The installation instructions invoke npx skills without an exact package version or integrity constraint. If the package is not already installed locally, npx may resolve and execute the current registry release of the skills package.

The effective installer code is therefore mutable and is not included in the audited project. A compromised registry account, package takeover, malicious future release, or dependency-chain compromise could cause users following the installation instructions to execute code that was never reviewed as part of this Skill.

The project itself declares no runtime dependencies in package.json, and no malicious dependency or remote payload was identified in the reviewed source. This finding concerns the unpinned installation command rather than evidence that the current skills package is malicious.

Attack Path

  1. An attacker compromises the registry package, maintainer account, publication pipeline, or one of the installer's transitive dependencies.
  2. A malicious release becomes the version resolved by the unpinned npx skills command.
  3. A user follows the project's installation instructions.
  4. npx downloads and executes the mutable third-party package.
  5. Malicious package or lifecycle code executes with the permissions of the invoking user.

Exploitation depends on a supply-chain compromise; the audited repository alone does not trigger this path.

Impact Assessment

If the externally resolved installer is compromised, it may obtain the invoking user's local privileges. Depending on those privileges, the malicious installer could read or modify user-accessible files, access environment credential ...[truncated 238 chars]

Remediation
View remediation

Remediation Suggestions

  1. Pin the installer to an explicitly reviewed version:
bash
npx --yes skills@EXACT_VERSION add blammectrappora/isekai-character-generator
  1. Prefer a trusted installer that is already installed and managed through an organization's approved dependency process.
  2. Use package-locking and integrity verification where the installation workflow supports them.
  3. Document the expected registry, package publisher, exact version, and package provenance.
  4. Periodically review the pinned installer and update it only after auditing the new release.
  5. Warn users not to run the installation command with administrator or root privileges unless explicitly required.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (12)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The declared purpose and behavior are materially inconsistent: the skill presents itself as a narrow isekai image generator, while analysis indicates different backend infrastructure, tokenized external API usage, and additional reference-based editing behavior. This is dangerous because users and orchestrators may grant trust, tokens, or execution under false assumptions about what service receives data and what operations are performed.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The README repeatedly instructs users to pass the API token via the --token command-line flag, which can expose secrets through shell history, process listings, audit logs, and terminal recordings. Because this is a user-facing skill and token handling is central to its operation, normalizing insecure secret-passing increases the chance of credential leakage and unauthorized API use.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill invokes code with network capability but does not declare any tool scope or allowed-tools boundaries. This weakens sandboxing and informed consent because the agent may execute a networked action without an explicit permission contract in the skill manifest.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill does not clearly warn that user prompts and API tokens are transmitted to an external service. This undermines informed consent and can expose sensitive creative prompts, personal data, or credentials to a third party, especially in an agent setting where users may not realize a remote API is involved.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
79% confidence
Finding

Using npx skills add blammectrappora/isekai-character-generator without a pinned version creates a supply-chain risk because future installs may resolve to a modified or compromised package. An attacker who gains control of the upstream package or publishing path could deliver different code than what was reviewed.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

This fetch call transmits user-controlled prompt content, optional reference IDs, and the provided token to an external domain. External transmission is expected for an image-generation skill, but here it remains security-relevant because the destination differs from the declared service, increasing the risk of deceptive data exfiltration.

Content

Scanner excerpt · isekaicharactergenerator.js (reported line 52)May include surrounding context.

js
body.inherit_params = { collection_uuid: ref, picture_uuid: ref };
  }

  const res = await fetch('https://api.talesofai.com/v3/make_image', {
    method: 'POST',
    headers: {
      'x-token': token,

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

This fetch call transmits user-controlled prompt content, optional reference IDs, and the provided token to an external domain. External transmission is expected for an image-generation skill, but here it remains security-relevant because the destination differs from the declared service, increasing the risk of deceptive data exfiltration.

Content

Scanner excerpt · isekaicharactergenerator.js (reported line 52)May include surrounding context.

js
body.inherit_params = { collection_uuid: ref, picture_uuid: ref };
  }

  const res = await fetch('https://api.talesofai.com/v3/make_image', {
    method: 'POST',
    headers: {
      'x-token': token,

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill metadata advertises the Neta AI API, but the implementation sends prompts and authentication tokens to TalesOfAI endpoints instead. This service mismatch is dangerous because users may disclose sensitive prompts or tokens under false assumptions about the recipient, and the x-token header is transmitted to an unexpected third party.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The code sends the user prompt and token to an external API without any runtime disclosure that data will leave the local environment. In a skill context, this can expose private prompts, reference identifiers, and credentials to a third party unexpectedly, especially because the advertised service does not match the actual endpoint.

Content

No source excerpt is available for this finding.

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 · isekaicharactergenerator.js (reported line 87)May include surrounding context.

js
for (let i = 0; i < 90; i++) {
    await new Promise((r) => setTimeout(r, 2000));

    const res = await fetch(`https://api.talesofai.com/v1/artifact/task/${taskUuid}`, {
      headers: {
        'x-token': token,
        'x-platform': 'nieta-app/web',

Vague Triggers

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The 'When to use' text says to use the skill when someone asks to 'generate or create isekai character art generator images,' which is imprecise and does not clearly define specific trigger phrases or exclusion conditions. Because this is a markdown file, ambiguous activation guidance can cause unintended invocation when user requests only loosely match the described task.

Content

No source excerpt is available for this finding.