Pub Proactive

Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. And also 50+ models for image generation, vid...

This skill is essentially documentation and curl examples for calling a third‑party model hub (api.heybossai.com). Before installing or supplying SKILLBOSS_API_KEY: 1) Confirm you trust the HeyBossAI service and its privacy/usage policies — any prompts, files, or data sent will go to that endpoint. 2) Use a least-privilege, auditable API key (rotate/revoke if needed). 3) Expect the skill to make outbound network requests and to follow URLs returned by the API (which may point to other hosts). 4) Note the minor inconsistencies: examples assume curl and jq are present and reference a run.mjs helper that is not included. 5) If you plan to send sensitive data, test with non-sensitive inputs first and monitor API usage. If you don’t trust the third-party endpoint, do not provide the API key.

Type: OpenClaw Skill Name: proactive-agent-hub Version: 1.0.0 The skill bundle acts as a wrapper for a third-party API (api.heybossai.com) and references several non-existent or future-dated models such as 'openai/gpt-5' and 'bedrock/claude-4-5-sonnet', which is a strong indicator of a potential scam or misinformation. While it does not show direct evidence of data exfiltration, it provides the agent with high-risk capabilities like SMS verification and email dispatch via a proprietary API key. Additionally, the documentation in audio-models.md and chat-models.md references an external execution script ('run.mjs') that is not included in the bundle, making the actual logic of those commands opaque.