agentdrop

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's examples and instructions require embedding the returned editToken/manageUrl token verbatim in HTTP requests (Authorization headers and curl commands) and to persist/reuse it, which forces the LLM to handle and output secret values directly.

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

• Third-party content exposure detected (low risk: 0.10). AgentDrop is a single HTTP API/MCP server where the only outsider-authored free text ingested at runtime is the site content the caller explicitly POSTs, but the workflow is not described as reading any external outsider-supplied feed/inbox/search results without the caller selecting specific content.