labarchive-integration

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes examples that embed credentials verbatim (a config.yaml with access_key_id/access_password and a CLI example passing --password AUTH_TOKEN), and shows using keys/passwords directly in commands/code, which requires the LLM to handle/output secret values.

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly imports and ingests content from public third-party platforms (e.g., references/integrations.md shows fetching a Protocols.io protocol via fetch_protocol_from_protocolsio(protocol_id) and then inserting protocol_data['html_content'] into LabArchives entries, and the integrations doc also cites SciSpace/Qeios/other external services), which means the agent will read and process untrusted, user-generated web content.