vercel-deploy

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to "Always show" the Claim URL (and the JSON output includes "claimUrl"), which would require the LLM to output a claim link containing a secret-like code verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill intentionally packages and uploads the user's project tarball to an external, unauthenticated deploy endpoint, which enables data exfiltration of project files (including any secrets) and produces claimable preview URLs that can be abused.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The runtime workflow reads outsider-authored free text from a user-supplied project path by ingesting package.json contents via cat in detect_framework() (and then packaging the project), but it is not an external public feed/queue that an outsider can poison without targeting the deploy input.