kamal-deploy

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs storing and referencing plaintext secrets (including a .kamal/secrets example with a ghp_ token) and tells the agent to confirm/put registry credentials into config, which encourages requesting and embedding secret values verbatim in outputs.

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

• Third-party content exposure detected (high risk: 0.80). The skill mandates using the WebFetch tool to retrieve and extract live documentation from public kamal-deploy.org pages (e.g., https://kamal-deploy.org/docs/installation/ and related docs), so the agent will fetch and interpret third-party public website content as part of its workflow, exposing it to potential indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill mandates runtime WebFetch calls (e.g. https://kamal-deploy.org/docs/installation/) that must be fetched before answering, so external documentation fetched during runtime directly controls the agent's instructions and is a required dependency.