# riskmandate.ai - [RiskMandate — the reviewed level: two sessions, and a professional signs it](https://riskmandate.ai/abp-reviewed.md) - [RiskMandate — the behaviour-policy vault for A browser extension](https://riskmandate.ai/abp-vault-browser-extension.md) - [RiskMandate — the behaviour-policy vault for ChatGPT in the browser](https://riskmandate.ai/abp-vault-chatgpt-web.md) - [RiskMandate — the behaviour-policy vault for Claude Code, confirmations off](https://riskmandate.ai/abp-vault-claude-code-cli-confirmations-off.md) - [RiskMandate — the behaviour-policy vault for Claude Code on your machine](https://riskmandate.ai/abp-vault-claude-code-cli.md) - [RiskMandate — the behaviour-policy vault for Claude Code on the web](https://riskmandate.ai/abp-vault-claude-code-web.md) - [RiskMandate — the behaviour-policy vault for Claude Desktop](https://riskmandate.ai/abp-vault-claude-desktop.md) - [RiskMandate — the behaviour-policy vault for Claude's Gmail connector](https://riskmandate.ai/abp-vault-claude-gmail-connector.md) - [RiskMandate — the behaviour-policy vault for Microsoft 365 connector (Claude)](https://riskmandate.ai/abp-vault-claude-m365-connector.md) - [RiskMandate — the behaviour-policy vault for Claude in the browser, connectors on](https://riskmandate.ai/abp-vault-claude-web-connectors.md) - [RiskMandate — the behaviour-policy vault for Dropbox MCP server](https://riskmandate.ai/abp-vault-dropbox-mcp.md) - [RiskMandate — the behaviour-policy vault for GitHub Actions](https://riskmandate.ai/abp-vault-github-actions.md) - [RiskMandate — the behaviour-policy vault for Gmail, read-only scope](https://riskmandate.ai/abp-vault-gmail-readonly.md) - [RiskMandate — the behaviour-policy vault for Google Drive, read-only scope](https://riskmandate.ai/abp-vault-google-drive-readonly.md) - [RiskMandate — the behaviour-policy vault for Google Workspace MCP servers](https://riskmandate.ai/abp-vault-google-workspace-mcp.md) - [RiskMandate — the behaviour-policy vault for n8n, owner API key](https://riskmandate.ai/abp-vault-n8n-owner-api-key.md) - [RiskMandate — the behaviour-policy vault for A scheduled job](https://riskmandate.ai/abp-vault-scheduled-job.md) - [RiskMandate — the Agent Behaviour Policy](https://riskmandate.ai/abp.md) - [RiskMandate — Accepted is not acceptable](https://riskmandate.ai/acceptable.md) - [Risk Mandate — agents act, and someone has to own the risk](https://riskmandate.ai/acceptance.md) - [RiskMandate — after payment: a debrief for the store team](https://riskmandate.ai/after-payment.md) - [RiskMandate — which Agent Behaviour Policy next?](https://riskmandate.ai/agent-behaviour-policy-next.md) - [RiskMandate — Agent Behaviour Policies, one vault per application](https://riskmandate.ai/agent-behaviour-policy.md) - [RiskMandate — For Agents](https://riskmandate.ai/agents.md) - [RiskMandate — an approval prompt is not a human in the loop](https://riskmandate.ai/article-approval-prompts.md) - [A deleted meeting comes back. An edited one does not.](https://riskmandate.ai/article-calendar-edits-cannot-be-undone.md) - [The pilot worked. Then somebody asked what else it could do.](https://riskmandate.ai/article-pilots-do-not-stay-in-production.md) - [One agent, and every desk it reaches. How to read the risk propagation visualiser.](https://riskmandate.ai/article-risk-propagation-visualiser.md) - [RiskMandate — in this session, the agent holds the union of everything it has ever been allowed to do](https://riskmandate.ai/article-union-of-every-session.md) - [RiskMandate — what an Agent Behaviour Policy is, and why one agent needs one](https://riskmandate.ai/article-what-is-an-abp.md) - [Who owns what in AI. And how accountability holds on the way up.](https://riskmandate.ai/article-who-owns-what-in-ai.md) - [RiskMandate — articles](https://riskmandate.ai/articles.md) - [RiskMandate — Brand guidelines](https://riskmandate.ai/brand.md) - [briefs-register.json](https://riskmandate.ai/briefs-register.json) - [RiskMandate — the brief register](https://riskmandate.ai/briefs.md) - [RiskMandate — the business case for agentgateway](https://riskmandate.ai/business-case-agentgateway.md) - [RiskMandate — the business case for Cedar](https://riskmandate.ai/business-case-cedar.md) - [RiskMandate — the business case for Cilium network policy](https://riskmandate.ai/business-case-cilium.md) - [RiskMandate — the business case for Falco](https://riskmandate.ai/business-case-falco.md) - [RiskMandate — the business case for gVisor](https://riskmandate.ai/business-case-gvisor.md) - [RiskMandate — the business case for Keycloak](https://riskmandate.ai/business-case-keycloak.md) - [RiskMandate — the business case for Langfuse](https://riskmandate.ai/business-case-langfuse.md) - [RiskMandate — the business case for LangGraph and LangChain human-in-the-loop](https://riskmandate.ai/business-case-langgraph-hitl.md) - [RiskMandate — the business case for LiteLLM](https://riskmandate.ai/business-case-litellm.md) - [RiskMandate — the business case for Open Policy Agent](https://riskmandate.ai/business-case-open-policy-agent.md) - [RiskMandate — the business case for OpenBao](https://riskmandate.ai/business-case-openbao.md) - [RiskMandate — the business case for OpenFGA](https://riskmandate.ai/business-case-openfga.md) - [RiskMandate — the business case for OWASP Coraza](https://riskmandate.ai/business-case-owasp-coraza.md) - [RiskMandate — the business case for OWASP Threat Dragon and pytm](https://riskmandate.ai/business-case-owasp-threat-dragon.md) - [RiskMandate — the business case for PostgreSQL point-in-time recovery](https://riskmandate.ai/business-case-postgresql-pitr.md) - [RiskMandate — the business case for Agent Behaviour Policy](https://riskmandate.ai/business-case-riskmandate-abp.md) - [RiskMandate — the business case for Squid](https://riskmandate.ai/business-case-squid.md) - [RiskMandate — the business case for Unleash](https://riskmandate.ai/business-case-unleash.md) - [RiskMandate — the business case for Velero](https://riskmandate.ai/business-case-velero.md) - [RiskMandate — business cases, by the risk they change](https://riskmandate.ai/business-cases.md) - [RiskMandate — Agent permission games, live](https://riskmandate.ai/demo-agent-permission-games.md) - [RiskMandate — Agentic Browser Isolation, live](https://riskmandate.ai/demo-agentic-browser-isolation.md) - [RiskMandate — File security, live](https://riskmandate.ai/demo-file-security.md) - [RiskMandate — Licence to Operate, live](https://riskmandate.ai/demo-licence-to-operate.md) - [RiskMandate — RiskGraph Explorer, live](https://riskmandate.ai/demo-risk-graph-explorer.md) - [RiskMandate — the field demo, live](https://riskmandate.ai/demo-risk-mandate-field.md) - [RiskMandate — Live demos](https://riskmandate.ai/demos.md) - [RiskMandate — Design options (internal)](https://riskmandate.ai/design-options.md) - [RiskMandate — Give feedback after a demo](https://riskmandate.ai/feedback.md) - [You run agents today](https://riskmandate.ai/for-corporate.md) - [You are a founder](https://riskmandate.ai/for-founders.md) - [You are a startup](https://riskmandate.ai/for-startups.md) - [RiskMandate — Know the risk. Name the owner. Own the mandate.](https://riskmandate.ai/grant-gap.md) - [RiskMandate — How it works: from a prompt to a graph with provenance](https://riskmandate.ai/how-it-works.md) - [RiskMandate — Know what your agents can do](https://riskmandate.ai/index.md) - [Make your agents insurable](https://riskmandate.ai/insurance.md) - [RiskMandate — can you insure a software program?](https://riskmandate.ai/insure-a-program.md) - [RiskMandate — Half an hour of your advice, by voice](https://riskmandate.ai/interview-founder-marketing.md) - [What buying a behaviour policy would look like — RiskMandate Lab 02](https://riskmandate.ai/lab-abp-flow.md) - [Changes we are asking of the behaviour-policy site — RiskMandate Lab 03](https://riskmandate.ai/lab-abp-requests.md) - [RiskMandate — Lab 05 · the commit author is a free text field](https://riskmandate.ai/lab-commit-author.md) - [The grant is user-shaped, not data-shaped — RiskMandate Lab 01](https://riskmandate.ai/lab-connector-grants.md) - [lab-editions.json](https://riskmandate.ai/lab-editions.json) - [RiskMandate — Lab 06 · every routable address is in the grant](https://riskmandate.ai/lab-network-reach.md) - [RiskMandate — Lab 04 · the shape collector](https://riskmandate.ai/lab-shape-collector.md) - [What you are actually buying — an Agent Behaviour Policy as a vault — RiskMandate Lab 07](https://riskmandate.ai/lab-vault-delivered.md) - [RiskMandate — the Lab](https://riskmandate.ai/lab.md) - [RiskMandate — Library](https://riskmandate.ai/library.md) - [Licence to Operate](https://riskmandate.ai/licence-to-operate.md) - [riskmandate.ai llms-full.txt](https://riskmandate.ai/llms-full.txt) - [riskmandate.ai llms.txt](https://riskmandate.ai/llms.txt) - [RiskMandate — OWASP, as a graph](https://riskmandate.ai/owasp-graph.md) - [RiskMandate — thank you: level 1, what happens now](https://riskmandate.ai/paid-t1.md) - [RiskMandate — thank you: level 2, what happens now](https://riskmandate.ai/paid-t2.md) - [RiskMandate — thank you: level 3, what happens now](https://riskmandate.ai/paid-t3.md) - [RiskMandate — thank you: level 4, what happens now](https://riskmandate.ai/paid-t4.md) - [RiskMandate — Partners](https://riskmandate.ai/partners.md) - [RiskMandate — Who can pull the plug?](https://riskmandate.ai/plug.md) - [RiskMandate — Pricing: four levels, £10 to £1,500](https://riskmandate.ai/pricing.md) - [RiskMandate — privacy: what this site collects, and how to check](https://riskmandate.ai/privacy.md) - [RiskMandate — questions we were asked](https://riskmandate.ai/questions.md) - [RiskMandate — RAMM, the Risk Acceptance Maturity Model](https://riskmandate.ai/ramm.md) - [RiskMandate — CISO XYZ: the shape of a reviewer page](https://riskmandate.ai/reviewer-ciso-xyz.md) - [RiskMandate — Dinis Cruz, who runs the review](https://riskmandate.ai/reviewer-dinis-cruz.md) - [reviewers.json](https://riskmandate.ai/reviewers.json) - [RiskMandate — who runs your review](https://riskmandate.ai/reviewers.md) - [RiskMandate — How long will you accept this risk?](https://riskmandate.ai/scenarios.md) - [RiskMandate — The post images](https://riskmandate.ai/statics.md) - [Startup Summit Lisbon 2026 — RiskMandate materials](https://riskmandate.ai/summit-booth.md) - [RiskMandate at Startup Summit Lisbon, 17–18 September 2026](https://riskmandate.ai/summit.md) - [Synthetic users — five people who do not exist, reading this site](https://riskmandate.ai/synthetic-users.md) - [RiskMandate — try it: write a behaviour policy for your own agent in twenty minutes](https://riskmandate.ai/try-it.md) - [uk-support.json](https://riskmandate.ai/uk-support.json) - [UK support, in the open](https://riskmandate.ai/uk-support.md) - [RiskMandate — version record](https://riskmandate.ai/versions.md) - [Brief B1 — power user and tester of Agent Behaviour Policies](https://riskmandate.ai/work-abp-power-user.md) - [Working with us — briefs for collaborators and their agents](https://riskmandate.ai/work.md) - [agent-content.json](https://riskmandate.ai/.well-known/agent-content.json) - [The console](https://riskmandate.ai/admin/index.md) - [The agents' front door](https://riskmandate.ai/admin/agents/index.md) - [Start here](https://riskmandate.ai/admin/agents/00-start-here/index.md) - [The map](https://riskmandate.ai/admin/agents/01-map/index.md) - [The Agent Behaviour Policy, condensed](https://riskmandate.ai/admin/agents/02-abp-model/index.md) - [Where we are, and what is next](https://riskmandate.ai/admin/agents/03-state-and-next/index.md) - [Rules of engagement for agents working in parallel](https://riskmandate.ai/admin/agents/04-rules-of-engagement/index.md) - [Workflows](https://riskmandate.ai/admin/agents/05-workflows/index.md) - [riskmandate.ai — read this first](https://riskmandate.ai/admin/agents/claude-md/index.md) - [brief-from-debrief](https://riskmandate.ai/admin/agents/commands/brief-from-debrief/index.md) - [handover](https://riskmandate.ai/admin/agents/commands/handover/index.md) - [merge-to-dev](https://riskmandate.ai/admin/agents/commands/merge-to-dev/index.md) - [new-page](https://riskmandate.ai/admin/agents/commands/new-page/index.md) - [new-vault](https://riskmandate.ai/admin/agents/commands/new-vault/index.md) - [onboard](https://riskmandate.ai/admin/agents/commands/onboard/index.md) - [release](https://riskmandate.ai/admin/agents/commands/release/index.md) - [research-vault](https://riskmandate.ai/admin/agents/commands/research-vault/index.md) - [Briefs written here](https://riskmandate.ai/admin/briefs/index.md) - [Architecture Brief — Decoupling site structure from content](https://riskmandate.ai/admin/briefs/architecture__structure-content-decoupling/index.md) - [Vaults in vaults: the application vault is data, the renderer lives once](https://riskmandate.ai/admin/briefs/architecture__vaults-in-vaults-for-behaviour-policies/index.md) - [The policy is a graph, and every stakeholder gets a projection of it](https://riskmandate.ai/admin/briefs/direction__abp-as-a-graph-and-stakeholder-views/index.md) - [Direction change — the Agent Behaviour Policy becomes the primitive](https://riskmandate.ai/admin/briefs/direction__abp-at-the-centre/index.md) - [A grant is a union of capabilities; what the reader needs is the consequences — and the assets that make each one real](https://riskmandate.ai/admin/briefs/direction__consequences-assets-and-the-vault-as-a-website/index.md) - [The first MVP vault: oc433z3m becomes the product, and the reading app is rebuilt around it](https://riskmandate.ai/admin/briefs/direction__mvp-vault-and-the-reading-app/index.md) - [The grant has storeys: what the credential permits, what the client exposes, what practice allows — and the vault opens on the audience, not on the data](https://riskmandate.ai/admin/briefs/direction__the-grant-has-storeys-and-the-vault-opens-on-the-audience/index.md) - [The menu after the ABP turn: five entries, one product first, and eleven pages out of the nav](https://riskmandate.ai/admin/briefs/direction__the-menu-and-the-page-list-after-the-abp-turn/index.md) - [Nobody buys a policy until they have made a small one, so the next phase is users](https://riskmandate.ai/admin/briefs/direction__the-next-phase-is-users/index.md) - [A policy per use case, and the £500 level is a prompt the customer runs](https://riskmandate.ai/admin/briefs/direction__use-case-driven-policies-and-the-prompt-workflow/index.md) - [How riskmandate.ai actually works](https://riskmandate.ai/admin/briefs/how-the-website-works/index.md) - [Implementation Brief — Scenarios pilot (first decoupled content area)](https://riskmandate.ai/admin/briefs/implementation__scenarios-pilot/index.md) - [LinkedIn company page — every field, ready to paste](https://riskmandate.ai/admin/briefs/marketing--linkedin-company-page/index.md) - [The agents' front door: onboard in ten minutes, and work in parallel without undoing each other](https://riskmandate.ai/admin/briefs/process__agent-onboarding-and-parallel-work/index.md) - [Connector grants: the open questions, written to be handed to an agent](https://riskmandate.ai/admin/briefs/research__connector-grants-open-questions/index.md) - [The Design Studio MVPs, read against the workflow that sells a behaviour policy](https://riskmandate.ai/admin/briefs/review__design-studio-mvps-vs-the-selling-workflow/index.md) - [The first measured behaviour policy: an owner API key on a self-hosted n8n, read against the model](https://riskmandate.ai/admin/briefs/review__first-measured-abp-n8n-owner-key/index.md) - [Five synthetic users read the new home page: the word is fixed, the vocabulary is split](https://riskmandate.ai/admin/briefs/review__five-synthetic-users-read-the-new-home-page/index.md) - [The vault pages, read against the vault: show the thing, and stop copying it](https://riskmandate.ai/admin/briefs/review__vault-pages-vs-the-vault/index.md) - [Lisbon messaging — we now have something to sell in the room](https://riskmandate.ai/admin/briefs/summit__lisbon-2026-messaging/index.md) - [Startup Summit Lisbon 2026 — strategy, materials, and the build list](https://riskmandate.ai/admin/briefs/summit__lisbon-2026-strategy/index.md) - [A behaviour policy for everybody the lead talks to, made by an agent from a zip](https://riskmandate.ai/admin/briefs/workflow__abp-vaults-for-people-we-know/index.md) - [Buying a behaviour policy for Claude on one Gmail mailbox: the workflow, run once](https://riskmandate.ai/admin/briefs/workflow__buying-a-policy-for-claude-on-gmail/index.md) - [The reviewer manifest: one source for who does the work, addressed to the store team](https://riskmandate.ai/admin/briefs/workflow__the-reviewer-manifest-and-who-owns-a-person-page/index.md) - [The memo queue](https://riskmandate.ai/admin/memos/index.md) - [The records](https://riskmandate.ai/admin/records/index.md) - [Tooling and the gate](https://riskmandate.ai/admin/tooling/index.md) - [The vaults](https://riskmandate.ai/admin/vaults/index.md) - [The board](https://riskmandate.ai/admin/work/index.md) - [T01 — One page per behaviour](https://riskmandate.ai/admin/work/T01/index.md) - [T02 — The grant as edges, one per path](https://riskmandate.ai/admin/work/T02/index.md) - [T03 — Metrics and outward links on the 23 behaviours](https://riskmandate.ai/admin/work/T03/index.md) - [T04 — Views per audience, and projections regenerated from a shipped prompt](https://riskmandate.ai/admin/work/T04/index.md) - [T05 — Settle the open questions in the five connector vaults](https://riskmandate.ai/admin/work/T05/index.md) - [T06 — The next five connector vaults, then the three business functions](https://riskmandate.ai/admin/work/T06/index.md) - [T07 — Lab 03: bring the request list against the model site up to date](https://riskmandate.ai/admin/work/T07/index.md) - [T08 — Bring docs/how-the-website-works.md to the current site](https://riskmandate.ai/admin/work/T08/index.md) - [T09 — Stop deploying the vault: inputs in the repository, the product in the vault](https://riskmandate.ai/admin/work/T09/index.md) - [T10 — The two Voice Debrief use-case vaults, and a use-case axis on the library](https://riskmandate.ai/admin/work/T10/index.md) - [T11 — Consequences and assets: what follows when a capability meets what is in the deployment](https://riskmandate.ai/admin/work/T11/index.md) - [T12 — The standards mini-graphs inside the vault: GDPR, the EU AI Act and ATT&CK as nodes to link to](https://riskmandate.ai/admin/work/T12/index.md) - [Task briefs — written to be picked up by one agent each](https://riskmandate.ai/admin/work/about-task-briefs/index.md) - [In-flight work — one file per branch](https://riskmandate.ai/admin/work/about-work-files/index.md) - [2026-09-15__transcript__website-repositioning-strategy.txt](https://riskmandate.ai/assets/briefs/2026-09-15__transcript__website-repositioning-strategy.txt) - [2026-09-15__voice-memo__abp-graph-and-stakeholder-views__otter-transcript.txt](https://riskmandate.ai/assets/briefs/2026-09-15__voice-memo__abp-graph-and-stakeholder-views__otter-transcript.txt) - [2026-09-15__voice-memo__use-case-driven-abp-strategy__otter-transcript.txt](https://riskmandate.ai/assets/briefs/2026-09-15__voice-memo__use-case-driven-abp-strategy__otter-transcript.txt) - [2026-09-22__transcript__the-named-professional-and-the-assignment.txt](https://riskmandate.ai/assets/briefs/2026-09-22__transcript__the-named-professional-and-the-assignment.txt) - [2026-09-24__transcript__abp-vaults-for-people-we-know.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__abp-vaults-for-people-we-know.txt) - [2026-09-24__transcript__business-case-by-risk-reduced.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__business-case-by-risk-reduced.txt) - [2026-09-24__transcript__calendar-edits-cannot-be-undone.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__calendar-edits-cannot-be-undone.txt) - [2026-09-24__transcript__graph-visualisations-for-the-role-ownership-article.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__graph-visualisations-for-the-role-ownership-article.txt) - [2026-09-24__transcript__how-it-technically-works.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__how-it-technically-works.txt) - [2026-09-24__transcript__owasp-and-open-source-first.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__owasp-and-open-source-first.txt) - [2026-09-24__transcript__pilots-do-not-stay-in-production.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__pilots-do-not-stay-in-production.txt) - [2026-09-24__transcript__red-travels-up-and-controls-leave-green-risks.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__red-travels-up-and-controls-leave-green-risks.txt) - [2026-09-24__transcript__risks-flow-upwards-and-the-accepted-level.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__risks-flow-upwards-and-the-accepted-level.txt) - [2026-09-24__transcript__uk-support-in-the-open.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__uk-support-in-the-open.txt) - [2026-09-24__transcript__who-owns-what-in-ai-and-abp-interview-questions.txt](https://riskmandate.ai/assets/briefs/2026-09-24__transcript__who-owns-what-in-ai-and-abp-interview-questions.txt) - [2026-09-25__transcript__an-article-on-the-risk-propagation-visualiser.txt](https://riskmandate.ai/assets/briefs/2026-09-25__transcript__an-article-on-the-risk-propagation-visualiser.txt) - [Every Routable Address Is In The Grant: Do Not Attack Anyone Is The One Rule Everybody Signs, And It Is The One With No Barrier](https://riskmandate.ai/assets/briefs/v0.33.70__dev-brief__every-routable-address-is-in-the-grant.md) - [The Commit Author Is A Free Text Field: A Prompt Shifts The Odds, And Every Documented Fix Was Architectural](https://riskmandate.ai/assets/briefs/v0.33.70__dev-brief__the-commit-author-is-a-free-text-field.md) - [The Grant Is User Shaped And Not Data Shaped: Start With The Connectors, And The Template Vault Is The Product](https://riskmandate.ai/assets/briefs/v0.33.70__dev-brief__the-grant-is-user-shaped-and-not-data-shaped.md) - [This Is The Tier One Application Nobody Could Find: The Connector List Alone Is Twenty Bits, So Compute Locally And Submit Banded](https://riskmandate.ai/assets/briefs/v0.33.70__dev-brief__this-is-the-tier-one-application-nobody-could-find.md) - [The Urgency Is Not A Deadline But A State: You Already Connected It, And A Distributed Skill Cannot Carry A Control](https://riskmandate.ai/assets/briefs/v0.33.70__strategy-brief__the-urgency-is-not-a-deadline-but-a-state.md) - [The Offer Is Built And The Button Is Not: Each Level Is The Level Below Plus One Thing, And The Post Sale Page Does Not Exist Yet](https://riskmandate.ai/assets/briefs/v0.33.71__strategy-brief__the-offer-is-built-and-the-button-is-not.md) - [OFL.txt](https://riskmandate.ai/assets/fonts/OFL.txt) - [Where this model comes from](https://riskmandate.ai/business-case/model/PROVENANCE.md) - [graph.json](https://riskmandate.ai/business-case/owasp/graph.json) - [index.json](https://riskmandate.ai/vaults/index.json) - [AGENT BEHAVIOUR POLICY — Claude Code on the web, with one repository attached](https://riskmandate.ai/vaults/claude-code-web/AGENT-BEHAVIOUR-POLICY.md) - [AGENTS.md — how to treat the behaviour policy files beside this one](https://riskmandate.ai/vaults/claude-code-web/AGENTS.md) - [DELTA — what it can do that you did not ask for](https://riskmandate.ai/vaults/claude-code-web/DELTA.md) - [GRANT — everything the agent can do](https://riskmandate.ai/vaults/claude-code-web/GRANT.md) - [LICENCE TO OPERATE — the organisation authorises the agent, under this behaviour policy, for an interval](https://riskmandate.ai/vaults/claude-code-web/LICENCE-TO-OPERATE.md) - [MANDATE — what the agent is authorised and expected to do](https://riskmandate.ai/vaults/claude-code-web/MANDATE.md) - [Claude Code on the web, with one repository attached — an Agent Behaviour Policy, as a vault](https://riskmandate.ai/vaults/claude-code-web/README.md) - [Agent Behaviour Policy — skill](https://riskmandate.ai/vaults/claude-code-web/SKILL.md) - [delta.json](https://riskmandate.ai/vaults/claude-code-web/data/delta.json) - [grant.json](https://riskmandate.ai/vaults/claude-code-web/data/grant.json) - [mandate.json](https://riskmandate.ai/vaults/claude-code-web/data/mandate.json) - [validity.json](https://riskmandate.ai/vaults/claude-code-web/data/validity.json) - [The vocabulary this vault was computed against](https://riskmandate.ai/vaults/claude-code-web/data/vocabulary/README.md) - [Grant check — 15 September 2026, from inside the shape](https://riskmandate.ai/vaults/claude-code-web/history/grant-check--2026-09-15.md) - [index.json](https://riskmandate.ai/vaults/claude-code-web/history/index.json) - [MAP-A-GRANT.md — a prompt for an agent that holds a connector or a credential](https://riskmandate.ai/vaults/gmail-readonly/MAP-A-GRANT.md) - [0.1.0.md](https://riskmandate.ai/versions/0.1.0.md) - [0.1.1.md](https://riskmandate.ai/versions/0.1.1.md) - [0.10.0.md](https://riskmandate.ai/versions/0.10.0.md) - [0.10.1.md](https://riskmandate.ai/versions/0.10.1.md) - [0.11.0.md](https://riskmandate.ai/versions/0.11.0.md) - [0.11.1.md](https://riskmandate.ai/versions/0.11.1.md) - [0.11.2.md](https://riskmandate.ai/versions/0.11.2.md) - [0.12.0.md](https://riskmandate.ai/versions/0.12.0.md) - [0.12.1.md](https://riskmandate.ai/versions/0.12.1.md) - [0.12.2.md](https://riskmandate.ai/versions/0.12.2.md) - [0.13.0.md](https://riskmandate.ai/versions/0.13.0.md) - [0.14.0.md](https://riskmandate.ai/versions/0.14.0.md) - [0.2.0.md](https://riskmandate.ai/versions/0.2.0.md) - [0.3.0.md](https://riskmandate.ai/versions/0.3.0.md) - [0.4.0.md](https://riskmandate.ai/versions/0.4.0.md) - [0.4.1.md](https://riskmandate.ai/versions/0.4.1.md) - [0.4.2.md](https://riskmandate.ai/versions/0.4.2.md) - [0.4.3.md](https://riskmandate.ai/versions/0.4.3.md) - [0.4.4.md](https://riskmandate.ai/versions/0.4.4.md) - [0.4.5.md](https://riskmandate.ai/versions/0.4.5.md) - [0.5.0.md](https://riskmandate.ai/versions/0.5.0.md) - [0.5.1.md](https://riskmandate.ai/versions/0.5.1.md) - [0.5.2.md](https://riskmandate.ai/versions/0.5.2.md) - [0.5.3 — Scenarios split into two pages](https://riskmandate.ai/versions/0.5.3.md) - [0.5.4.md](https://riskmandate.ai/versions/0.5.4.md) - [0.5.5.md](https://riskmandate.ai/versions/0.5.5.md) - [0.5.6 — Risk Scenarios nav dropdown](https://riskmandate.ai/versions/0.5.6.md) - [0.6.0 — Pages-only nav, a Pricing page, and RAMM](https://riskmandate.ai/versions/0.6.0.md) - [0.7.0 — RAMM on the homepage + the interactive pyramid](https://riskmandate.ai/versions/0.7.0.md) - [0.8.0 — Homepage links to the interactive scenarios](https://riskmandate.ai/versions/0.8.0.md) - [0.9.0 — Mobile navigation fix](https://riskmandate.ai/versions/0.9.0.md) - [0.9.1.md](https://riskmandate.ai/versions/0.9.1.md) - [0.9.2.md](https://riskmandate.ai/versions/0.9.2.md) - [0.9.3.md](https://riskmandate.ai/versions/0.9.3.md) - [0.9.4.md](https://riskmandate.ai/versions/0.9.4.md) - [1.0.0.md](https://riskmandate.ai/versions/1.0.0.md) - [1.1.0.md](https://riskmandate.ai/versions/1.1.0.md) - [1.10.0.md](https://riskmandate.ai/versions/1.10.0.md) - [1.11.0.md](https://riskmandate.ai/versions/1.11.0.md) - [1.12.0.md](https://riskmandate.ai/versions/1.12.0.md) - [1.13.0.md](https://riskmandate.ai/versions/1.13.0.md) - [1.14.0.md](https://riskmandate.ai/versions/1.14.0.md) - [1.15.0.md](https://riskmandate.ai/versions/1.15.0.md) - [1.16.0.md](https://riskmandate.ai/versions/1.16.0.md) - [1.17.0.md](https://riskmandate.ai/versions/1.17.0.md) - [1.18.0.md](https://riskmandate.ai/versions/1.18.0.md) - [1.19.0.md](https://riskmandate.ai/versions/1.19.0.md) - [1.19.1.md](https://riskmandate.ai/versions/1.19.1.md) - [1.19.2.md](https://riskmandate.ai/versions/1.19.2.md) - [1.19.3.md](https://riskmandate.ai/versions/1.19.3.md) - [1.2.0.md](https://riskmandate.ai/versions/1.2.0.md) - [1.2.1.md](https://riskmandate.ai/versions/1.2.1.md) - [1.20.0.md](https://riskmandate.ai/versions/1.20.0.md) - [1.20.1.md](https://riskmandate.ai/versions/1.20.1.md) - [1.20.2.md](https://riskmandate.ai/versions/1.20.2.md) - [1.21.0.md](https://riskmandate.ai/versions/1.21.0.md) - [1.22.0.md](https://riskmandate.ai/versions/1.22.0.md) - [1.22.1.md](https://riskmandate.ai/versions/1.22.1.md) - [1.23.0.md](https://riskmandate.ai/versions/1.23.0.md) - [1.24.0.md](https://riskmandate.ai/versions/1.24.0.md) - [1.24.1.md](https://riskmandate.ai/versions/1.24.1.md) - [1.24.2.md](https://riskmandate.ai/versions/1.24.2.md) - [1.25.0.md](https://riskmandate.ai/versions/1.25.0.md) - [1.25.1.md](https://riskmandate.ai/versions/1.25.1.md) - [1.25.2.md](https://riskmandate.ai/versions/1.25.2.md) - [1.25.3.md](https://riskmandate.ai/versions/1.25.3.md) - [1.25.4.md](https://riskmandate.ai/versions/1.25.4.md) - [1.26.0.md](https://riskmandate.ai/versions/1.26.0.md) - [1.26.1.md](https://riskmandate.ai/versions/1.26.1.md) - [1.26.2.md](https://riskmandate.ai/versions/1.26.2.md) - [1.27.0.md](https://riskmandate.ai/versions/1.27.0.md) - [1.27.1.md](https://riskmandate.ai/versions/1.27.1.md) - [1.27.2.md](https://riskmandate.ai/versions/1.27.2.md) - [1.27.3.md](https://riskmandate.ai/versions/1.27.3.md) - [1.28.0.md](https://riskmandate.ai/versions/1.28.0.md) - [1.29.0.md](https://riskmandate.ai/versions/1.29.0.md) - [1.29.1.md](https://riskmandate.ai/versions/1.29.1.md) - [1.29.2.md](https://riskmandate.ai/versions/1.29.2.md) - [1.29.3.md](https://riskmandate.ai/versions/1.29.3.md) - [1.29.4.md](https://riskmandate.ai/versions/1.29.4.md) - [1.29.5.md](https://riskmandate.ai/versions/1.29.5.md) - [1.3.0.md](https://riskmandate.ai/versions/1.3.0.md) - [1.30.0.md](https://riskmandate.ai/versions/1.30.0.md) - [1.30.1.md](https://riskmandate.ai/versions/1.30.1.md) - [1.31.0.md](https://riskmandate.ai/versions/1.31.0.md) - [1.31.1.md](https://riskmandate.ai/versions/1.31.1.md) - [1.32.0.md](https://riskmandate.ai/versions/1.32.0.md) - [1.32.1.md](https://riskmandate.ai/versions/1.32.1.md) - [1.32.2.md](https://riskmandate.ai/versions/1.32.2.md) - [1.32.3.md](https://riskmandate.ai/versions/1.32.3.md) - [1.33.0.md](https://riskmandate.ai/versions/1.33.0.md) - [1.34.0.md](https://riskmandate.ai/versions/1.34.0.md) - [1.34.1.md](https://riskmandate.ai/versions/1.34.1.md) - [1.34.2.md](https://riskmandate.ai/versions/1.34.2.md) - [1.34.3.md](https://riskmandate.ai/versions/1.34.3.md) - [1.34.4.md](https://riskmandate.ai/versions/1.34.4.md) - [1.34.5.md](https://riskmandate.ai/versions/1.34.5.md) - [1.34.6.md](https://riskmandate.ai/versions/1.34.6.md) - [1.34.7.md](https://riskmandate.ai/versions/1.34.7.md) - [1.34.8.md](https://riskmandate.ai/versions/1.34.8.md) - [1.4.0.md](https://riskmandate.ai/versions/1.4.0.md) - [1.5.0.md](https://riskmandate.ai/versions/1.5.0.md) - [1.6.0.md](https://riskmandate.ai/versions/1.6.0.md) - [1.7.0.md](https://riskmandate.ai/versions/1.7.0.md) - [1.8.0.md](https://riskmandate.ai/versions/1.8.0.md) - [1.9.0.md](https://riskmandate.ai/versions/1.9.0.md) - [index.json](https://riskmandate.ai/versions/index.json) - [source-snapshot.json](https://riskmandate.ai/versions/source-snapshot.json)