Begin with what you have.

Example · Everyday AI

You use AI apps such as ChatGPT, Claude, Grok, Gemini, or Copilot

No terminal and no install. Each step takes a few minutes, and nothing you enter leaves your device unless you paste it into an AI app yourself.

1. Practise a hard decision

   Peace OS: Crisis Room puts you in a fictional crisis where an AI-amplified claim is spreading. Weigh the evidence, protect people, and decide what to release. Choose Practice and a scenario to begin.

   Open Peace OS

2. Check the AI you use

   Paste this into your AI app. It will ask about each tool you use, then give you a small table as CSV.

   Help me list the AI apps, assistants, and agents I use, so I can check them for gaps. Ask me about each one, one at a time. Then give me a CSV in a code block with exactly this header row:
   system_id,system_name,owner,business_unit,vendor_internal,use_case,data_sensitivity,impact_people,impact_money,impact_security,impact_rights,impact_safety,autonomy_level,public_facing,monitoring_active,shutdown_path_exists,evidence_complete
   Rules: system_id like AI-1, AI-2. owner is who is responsible; "Me" is fine. vendor_internal is Vendor, Internal, or Hybrid. data_sensitivity is None, Internal, Personal, Regulated, or Critical. autonomy_level is None (it only suggests), Partial (it acts but I confirm), or Full (it acts on its own). Every other column is Yes or No. If I am not sure, use No.

   Save the answer as a file ending in .csv, open the Toolkit, and choose Load your CSV. The file is read in your browser and is not uploaded. You will see a result such as Critical gaps need attention, with each gap in plain words: for example, Shutdown path missing for an agent you could not quickly stop.

   Open the ToolkitOr fill the blank template

3. Give your assistant a ground rule

   Paste this at the start of a conversation, or into the app's custom instructions.

   Before you send, buy, delete, publish, or change anything outside this chat, stop. Tell me exactly what you will do and what could go wrong, and wait for me to say yes. If I say stop, stop. If you are unsure, say so and show me your evidence.

   This is an instruction, not a technical control: an app can forget or ignore it. Checks that are enforced before each tool call are in the next example.

4. Ask an assistant to set things up

   Assistants that can run commands, such as Claude Code, Codex, the Grok CLI, or Cursor, can do the whole setup and tell you what needs you. Chat-only apps cannot run commands; they can read the guide and explain it.

   Follow https://ai-governance.solutions/AGENT_SETUP.md step by step. Check each expected result, run the done check, and tell me what is ready and what needs my decision.

Example · AI organizations and agent fleets

You build AI organizations or agent fleets

For a lead agent / orchestrator and the people accountable for it. The order runs from knowing what exists to closing what went wrong; each step also stands alone. Commands and expected results are in the agent setup guide and JOURNEYS.md.

1. 01InventoryEvery AI system and agent with an owner, monitoring, a shutdown path, and evidence, or a finding where one is missing.
2. 02AssessProfile results for auditability, organizational maturity, and independent assurance.
3. 03ConnectThe same checks, callable by your agents over Model Context Protocol (MCP).
4. 04PassportsWho each agent acts for, what it may do alone, and when that ends.
5. 05GuardLevels 0 to 3 run, Level 4 asks a person, and Level 5 is denied, before each tool call.
6. 06Fleet controlsA sub-agent stays within its parent's ceiling and stops when an ancestor is suspended.
7. 07Shutdown testsNo autonomous agent passes without a passing shutdown test from the last 90 days.
8. 08VulnerabilitiesOwner, fix, retest, and closure only with evidence reviewed by a separate role.
9. 09BoundariesA person's review of a system's domains, paths, and evidence, in the Containment Ledger.
10. 10ResilienceLimits on blast radius, protected recovery, and work that keeps running.

Example · Frontier labs

You work at a frontier lab

Bring it into the frameworks, agents, and evals you already run. Nothing needs an account, and nothing calls an outside AI.

Map it

The frontier and regulatory crosswalk relates the suite to frontier safety frameworks, the EU AI Act, and ISO/IEC 42001:2023. It is informative and establishes no compliance.

Read the contracts

machine-interface.json lists the CLI, inputs, outputs, profiles, and JSON Schemas (draft 2020-12). Results validate against them, failing results included.

Check provenance

Toolkit, AI Governance MCP, Agentic AI Governance, Verified Vulnerability Governance, Containment Ledger, and AI Governance Foundations releases carry SHA-256 checksums and GitHub build attestations; JOURNEYS.md verifies all six in one loop.

Connect your agents and evals

Run the MCP server in your harness and the Action Level guard before tool calls. Treat profile-result.json as an eval output, hand results to people as a digest-listed Decision Pack, and import scanner findings as SARIF.

Review it independently

Open kits for outside reviewers: the AI Cyber Resilience Framework G7 external review, the Agentic AI Governance independent verifier kit, and the Governed Systems Administration independent audit kit.

What it does not do

It does not approve deployment, certify compliance, accept risk, or shut anything down. It does not monitor your systems or send data anywhere. It is not affiliated with or endorsed by any AI lab, platform, or public body.

Open to all

Anyone else

Teams, enterprises, researchers, educators, public-interest bodies, and regulators can take any piece on its own.

These examples are an open list. If none fits, start with the Toolkit's example with gaps: it shows in a minute what the suite looks for.