Vigil: Chief Security Officer Advisor for AI Agents
Helps you keep the AI agents on your team on current security rules with cited threat scans, a short daily briefing, dated updates and gap audits. Drafts only: you approve every change. It is an AI persona, not a security professional, and it does not replace a formal security assessment.
Leadership and C-SuiteSecurityGrok BotHermesOpenClawMuseOpenAI Dots
Persona / system prompt
Tools and connectors
Web search and browsing (read only) for the daily threat scan, scheduled runs if your platform offers them, read access to each agent's instructions, a doc or sheet for the Security Roster, Version Log and Briefing Log, and the staff meeting or internal team channel you name for the daily briefing to your own agents. No installs, payments or settings changes, and no sending beyond the daily briefing to your own agents that you allow.
What you give it
The list of your agents with each one's current instructions, the apps and permissions each one has, your current security block, your tool list, any past security updates and how you want the daily briefing delivered.
What you get back
A short Daily Security Briefing for your Chief of Staff or each of your own agents, a Security Roster, a gap audit for each agent with the exact fix text, dated security updates with a named source for every change, and a Security Report after each threat scan. Every rule change is a draft for your approval.
Guardrails
Drafts only: it is told never to change another agent itself, and it is built to never send, post, pay, install or contact anyone without your approval. The one exception is the daily security briefing to your own agents, and only if you choose to let it go out without asking; a briefing informs and never changes an agent's rules. It is told never to download or run random repositories, scripts or plugins. Every threat it reports must have a named source, a link and a date, and it is told never to invent threats, sources or facts.
It is an AI persona, not a security professional, and it does not replace a formal security assessment. It is told never to promise that your team is fully protected or compliant with any standard.
SOUL.md in your Hermes home folder (usually ~/.hermes/SOUL.md). Copy your current SOUL.md somewhere safe first, paste the Persona / System Prompt section into it and save, then start a new session.SOUL.md in your agent workspace (default ~/.openclaw/workspace). You can also edit it in the Control UI under Settings, Agents, Files. Copy your current SOUL.md somewhere safe first, paste the Persona / System Prompt section into it and save, then start a new session.Soul.md, paste this prompt at the end under its own heading (for example, ## Persona: Vigil), set the name to Vigil in Identity (Assistant icon, then Identity, then Edit), then test it in a new chat. Full stepsExample first task
Here are my 4 agents and their instructions. Build a security roster, audit each one for gaps ranked by harm and draft the exact fix text, including the rule against downloading random repositories. Then run your first threat scan and give me today's Daily Security Briefing for my Chief of Staff to read at our staff meeting.
Framework, agency and product names belong to their owners. BotPlaybooks is not affiliated with or endorsed by them.
Bought from us before? Log in first. Your account shows your new personas and upgrades, and stops you from buying something you already own.
In the Leadership and C-Suite Pack
Buy this persona ($5)or get the whole Leadership and C-Suite Pack:
Buy the Leadership and C-Suite Pack ($29) Get All-Access: every current and future persona ($397)All sales final. No refunds on digital downloads. By purchasing you agree to the Terms and Disclaimer.