promptic

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflows and API calls explicitly retrieve user-created content from the third-party Promptic platform (e.g., "prompt = client.get_deployed_prompt('comp_...')" in the Prompt Optimization Workflow and the generalJudge/evaluator config.messages in references/api.md), which the agent is expected to read and can materially change its prompts/behavior, so it exposes the agent to untrusted third-party content.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime examples call the Promptic API (default endpoint https://promptic.eu) — e.g., client.get_deployed_prompt(...) and printing/using the returned prompt — which fetches external prompt text at runtime that can directly control agent instructions, so https://promptic.eu is flagged.