memory-lancedb-pro

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the user to "paste" API keys (or substitute them inline), instructs verifying them with curl commands that include Authorization headers, and tells the agent to insert actual keys into config JSON — forcing the LLM to handle and output secret values verbatim, which is an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly instructs the agent to fetch and run third-party content (e.g., the Quick Install curl to raw.githubusercontent.com, git clone of public GitHub repos, and numerous curl checks to external providers like https://api.jina.ai, https://api.openai.com, https://api.siliconflow.com and local Ollama endpoints) and to interpret those responses as part of the configuration/verification steps, so untrusted public content could materially alter tool use or next actions.