microsoft-sharepoint-webhooks

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly requires echoing the incoming validationtoken query parameter back "verbatim" in the HTTP response (and thus would force any agent implementing the handler to emit that token value), which is direct verbatim handling of a secret-like value even though clientState itself is read from an env var.

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

• Third-party content exposure detected (high risk: 0.85). This skill’s required runtime workflow accepts SharePoint webhook POSTs and reads attacker-controlled outsider free text from the request query/body (e.g., validationtoken query string and JSON value array fields), which is then used as strings in the handler; even though it’s not passed to an LLM in these examples, there is a direct path for outsider text to enter application context.