integrate-flowlines-opencode-plugin

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to call Flowlines.getMemory() from the third-party Flowlines service (see "Fetch and use memory" using @flowlines/sdk) and to inject that returned string into prompts, so untrusted external memory content could be read and influence agent behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill calls Flowlines at runtime (e.g., Flowlines.getMemory and the shown API call http://api.flowlines.ai/v1/get-traces) and explicitly instructs injecting the returned memory string into agent prompts, meaning remote content from api.flowlines.ai can directly control prompts.