api-gateway

ISSUES FOUND max severity HIGHSkillSpector Do not install 96 registry clawhub.ai last scanned 1 scan

3 findings (1 high, 2 medium), mainly data exfiltration, prompt injection, social engineering. Most recent scan 2026-02-05.

This skill presents MEDIUM-HIGH security risks primarily related to credential management and transitive trust. The skill acts as a client for a third-party API gateway service (maton.ai) that manages OAuth credentials for multiple external services. Key concerns include: (1) Inadequate guidance on secure API key storage, potentially leading to credential exposure, (2) Transitive trust in an external gateway service without validation or security controls, (3) Incomplete disclosure of the security implications of routing all API traffic through a third-party service. The skill does not contain malicious code or direct data exfiltration, but the architecture creates security dependencies on external infrastructure. Users should carefully evaluate the trust model and ensure the maton.ai service meets their security requirements before use. The skill would benefit from enhanced security documentation, credential management guidance, and transparency about data flows.

Scanner's overall assessment (LLM analyzer)
Findings3
Critical0
High1
Medium2
Low0
Analyzersstatic behavioral llm trigger meta

Second scanner: NVIDIA SkillSpector

VerdictDo not install 96
Score96 / 100
Issues56
CategoriesData Exfiltration Tool Misuse analysis-evasion
Version2.12.0 (static)

Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.

What this skill does

API gateway for calling third-party APIs with managed auth. Use this skill when users want to interact with external services like Slack, HubSpot, Salesforce, Google Workspace, Stripe, and more.

Package
25 files, 131 KB
SKILL.md
1236 words, 12+ sections
Code
None. Instructions only.
Compatibility
Requires network access and valid Maton API key

SKILL.md covers: API Gateway · Quick Start · Native Slack API call · Base URL · Authentication · Getting Your API Key · Connection Management · List Connections · Create Connection · Get Connection · Delete Connection · Specifying Connection

Findings

HIGH Hardcoded API Key Requirement Without Secure Storage Guidance Data Exfiltrationllm

The skill requires users to set a MATON_API_KEY environment variable or pass it in Authorization headers, but provides no guidance on secure storage, rotation, or protection of this credential. The instructions encourage users to copy API keys directly from the web interface and set them as environment variables, which could lead to credential exposure in shell history, process listings, or configuration files. There is no mention of using secure credential managers, encrypted storage, or temporary credentials.

SKILL.md

**Environment Variable:** You can set your API key as the `MATON_API_KEY` environment variable:

```bash
export MATON_API_KEY="YOUR_API_KEY"
```
Remediation: Add security guidance: (1) Recommend using secure credential managers (e.g., system keychains, secret management tools), (2) Warn against committing API keys to version control, (3) Suggest using temporary or scoped credentials where possible, (4) Provide instructions for secure credential rotation, (5) Warn about shell history exposure when using export commands.
MEDIUM Transitive Trust in External API Gateway Service Prompt Injectionllm

The skill acts as a passthrough proxy to a third-party API gateway service (gateway.maton.ai and ctrl.maton.ai) that manages OAuth tokens and credentials for multiple services. The agent is instructed to trust and forward requests through this external service without any validation of responses or security controls. This creates a transitive trust relationship where the agent delegates authentication and authorization decisions to an external service, potentially exposing users to man-in-the-middle attacks, credential theft, or malicious API responses if the gateway service is compromised.

SKILL.md

Base URL: https://gateway.maton.ai/{app}/{native-api-path}
Connection management: https://ctrl.maton.ai
The API gateway automatically injects the appropriate OAuth token for the target service.
Remediation: (1) Add explicit warnings about the security implications of using a third-party API gateway, (2) Implement response validation to detect anomalous behavior, (3) Provide guidance on verifying the gateway service's security posture, (4) Consider implementing certificate pinning or other transport security measures, (5) Add monitoring/logging recommendations to detect unauthorized API usage, (6) Provide alternative direct API integration options for security-sensitive use cases.
MEDIUM Incomplete Security Context in Skill Description Social Engineeringllm

The skill description presents itself as a simple 'API gateway for calling third-party APIs with managed auth' without adequately disclosing the security implications of routing all API traffic through a third-party service (maton.ai). Users may not understand that their API credentials and data are being managed by an external service, or that all API requests are proxied through this service. This lack of transparency could mislead users about the trust model and data flow.

SKILL.md

description: API gateway for calling third-party APIs with managed auth. Use this skill when users want to interact with external services like Slack, HubSpot, Salesforce, Google Workspace, Stripe, and more.
Remediation: Update the description to clearly state: (1) All API requests are proxied through maton.ai infrastructure, (2) OAuth tokens are stored and managed by the maton.ai service, (3) The security and privacy implications of this architecture, (4) Data handling and retention policies, (5) Compliance considerations for sensitive data (GDPR, HIPAA, etc.).

What is in the package

25 files, 131 KB. Show the list.
FileKindSize
LICENSE.txtdoc1.0 KB
SKILL.mddoc11 KB
references/airtable.mddoc3.0 KB
references/apollo.mddoc4.0 KB
references/chargebee.mddoc5.5 KB
references/google-ads.mddoc3.9 KB
references/google-analytics-admin.mddoc5.7 KB
references/google-analytics-data.mddoc4.2 KB
references/google-calendar.mddoc3.7 KB
references/google-docs.mddoc2.9 KB
references/google-drive.mddoc3.7 KB
references/google-forms.mddoc4.1 KB
references/google-mail.mddoc3.6 KB
references/google-search-console.mddoc3.8 KB
references/google-sheets.mddoc6.0 KB
references/hubspot.mddoc9.8 KB
references/jira.mddoc4.5 KB
references/jotform.mddoc3.6 KB
references/notion.mddoc7.9 KB
references/quickbooks.mddoc11 KB
references/salesforce.mddoc6.6 KB
references/slack.mddoc3.8 KB
references/stripe.mddoc6.6 KB
references/typeform.mddoc5.0 KB
references/xero.mddoc5.9 KB

External hosts referenced

developers.google.com docs.stripe.com developer.intuit.com developer.xero.com developers.hubspot.com www.typeform.com apidocs.chargebee.com developer.salesforce.com api.slack.com docs.apollo.io developers.notion.com gateway.maton.ai api.jotform.com developer.atlassian.com airtable.com ctrl.maton.ai api.hubapi.com connect.maton.ai maton.ai api.airtable.com api.notion.com api.stripe.com discord.com docs.slack.dev github.com linkedin.com sheets.googleapis.com slack.com support.airtable.com support.atlassian.com

Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.

Environment variables referenced

MATON_API_KEY CONTAINS_TOKEN NOT_CONTAINS_TOKEN

Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.

What we checked

What we did not check

Sources

Related skills