uploads-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes examples and guidance that pass workspace/admin tokens directly on the command line (e.g. --token up_acme_…) and tells operators to hand a token to the agent via UPLOADS_TOKEN, which would require the LLM to handle or embed secret values verbatim in generated commands or output.

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

• Third-party content exposure detected (low risk: 0.10). Source text from outsider authors is only ingested when the caller supplies arbitrary content (e.g., uploads put --url fetching a user-provided HTTPS URL, or hosted MCP put fetching contentUrl), so the runtime workflow does not passively read queue/feed items without the user selecting specific content.