javascript-sdk

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The document contains examples that place API keys and webhook URLs as literal values in generated code (e.g., apiKey: 'inf_your_key', apiKey: 'inf_...', and a Slack webhook URL), which encourages embedding secrets verbatim in outputs and thus poses a secret-exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The docs contain intentional high-risk patterns: direct eval() of external tool inputs (RCE) and automatic/manual uploading of local files (data exfiltration) that can be abused as backdoors or credential/data theft vectors.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The agent configuration (SKILL.md) includes a skill with a runtime "url" pointing to https://example.com/skills/api-docs.md which is likely fetched and injected as reusable agent context, thus it can directly control prompts at runtime.