Security Standards

Record the concrete technical norms your organisation holds itself to, itemise each one as a requirement, and link the standard to the policies it underpins.

A security standard (a written set of concrete technical rules — minimum TLS version, encryption-at-rest requirement, key-management rule, and so on) sits one level below your policies. A policy says what the organisation requires; a standard says exactly how strong that requirement must be. This page keeps those norms in one place: each standard gets a version, an owner, a list of requirements, and links to the policies that rely on it. Aegis stores the definition; people write it and decide when it comes into force.

Who uses it

The page sits in the GOVERNANCE group of the left menu. Reading is open to every role. Authoring is treated as a governance act and is reserved for senior roles.

Role What they can do here
Admin Everything — create, edit and delete standards; manage requirements; link and unlink policies; run the AI review.
Manager The same authoring rights as an Admin.
Contributor Read only. Can browse, search, open a standard and run the AI review, but sees no New security standard button and no edit, delete or link controls.
Viewer Read only, the same as a Contributor.
The screenshot below was captured as a Contributor

The create button is therefore absent by design. With a Manager or Admin role you also see a New security standard button at the top right, and authoring controls on each standard. The whole page belongs to the Compliance module — if that module is switched off for your organisation, the menu entry does not appear at all.

What's on this screen

The page opens on the standards overview. At the top left is the heading Security Standards with a one-line description of what the page is for: record the concrete technical norms the organisation holds itself to, and link them to the policies they underpin. Directly beneath it is a filter bar holding two controls: a Search security standards... box on the left, and an All statuses drop-down beside it.

Below the filter bar, each standard appears as a card in a grid — three cards across on a wide screen, two on a medium one and a single column on a phone. A card shows the name at the top left with a coloured status badge beside it at the top right; when the name is too long to share the line, the badge drops beneath it. Then come a Version line and an Owner line (reading No owner when nobody is assigned), and a footer counting requirements and linked policies. The whole card is a link through to the detail page. At the foot of the list is a page counter with Previous and Next buttons, both greyed out when everything fits on one page. The list shows 20 standards per page.

The captured screen shows eight standards for the demo organisation: five badged Active, two Draft and one Retired. They range from an encryption and transport standard with ten requirements and five linked policies to a draft cloud-workload standard at Version 0.3 with No owner and 0 linked policies, and a retired 2019 SSL/TLS standard kept for the record. The counter reads Page 1 of 1. Before you record anything, the page shows an empty state headed No security standards yet instead of the grid — with a New security standard action for Managers and Admins only.

While the list loads you see placeholder rows. If it cannot load, a red banner reads Failed to load security standards. with a Retry link; the filter bar stays in place. If your own role could not be looked up, a banner explains that your permissions could not be loaded and some actions are hidden — reload the page rather than assume you have lost access.

Find a standard

  1. Type part of a name into the Search security standards... box. After a short pause the grid narrows to standards whose name contains what you typed, ignoring upper and lower case. The match looks at the name only, not at descriptions or requirement text, and the page counter resets to the first page.
  2. Choose a state from the All statuses drop-down to show one lifecycle stage at a time — Draft, Active or Retired. When a filter combination matches nothing, the grid is replaced by No security standards match the current filters. and the filter bar stays in place so you can clear it again.
  3. Read the status badge beside a card's name to see where that standard stands. Draft is amber, Active green and Retired grey, and each badge carries its wording as text, so you never have to rely on the colour.
  4. Select the card to open that standard. Its detail page appears, with the requirements and linked policies underneath the header.
The Security Standards overview: search box, status filter, and a grid of standard cards with status, version, owner and counts — /security-standards.
The Security Standards overview: search box, status filter, and a grid of standard cards with status, version, owner and counts — /security-standards.

What the status values mean

A standard's status is chosen by hand when you edit it. There is no enforced order, so you move a standard between the three states as your own governance process requires.

Status What it means
Draft Being written or revised. Every new standard starts here.
Active In force — the norm the organisation currently holds itself to.
Retired Superseded or withdrawn, kept for the record.

Create a standard

This task needs a Manager or Admin role. The capture above was taken as a Contributor and so does not show the create form; the steps below describe the flow without a figure.

  1. Select New security standard at the top right. A dialogue headed Define security standard opens over the list.
  2. Enter a Name — for example, Encryption & Transport Security Standard. This is the only required field, and it accepts up to 200 characters.
  3. Fill in the optional fields: a Version label such as 1.0, an Owner from your user list, and a Description setting out scope and purpose. Leaving the version blank records 1.0; leaving the owner blank shows No owner on the card.
  4. Select Create. If the name is missing, the dialogue stays open and shows Name is required under the field. Otherwise the dialogue closes, a Security standard created confirmation appears, and the standard joins the grid as a Draft. Status is deliberately not offered while creating — every standard begins as a draft, and you move it on later by editing it.

Read a standard's detail page

Opening a standard shows a Back to security standards link, then a header carrying the name, the status badge, the description, and the Version and Owner lines. To the right of the header sit the action buttons: Review & Map for everyone, plus Edit and Delete for Managers and Admins. Beneath the header are two sections, Requirements and Linked policies.

Itemise the requirements

The requirements section is where the standard's definition actually lives — each row is one concrete norm. Managing requirements needs a Manager or Admin role; other roles see the table but no action controls.

  1. In the Requirements section, use the All categories drop-down to narrow the table to one area — for example TLS Configuration, Encryption at Rest or Key Management. If the filter matches nothing, the table is replaced by No requirements match the current filter.
  2. Select Add requirement. A dialogue opens with a Category picker, a Requirement box for the norm itself (for example, TLS 1.2 minimum, TLS 1.3 preferred for all public endpoints), and an optional Rationale recording why the norm exists. The category starts on Other; change it to the closest fit.
  3. Select Add. A Requirement added confirmation appears and the row joins the table, which lists Category, Requirement, Rationale and — for those who can edit — an Actions column. The requirement count on the standard's card goes up. Long lists are paged, with their own Previous and Next buttons.
  4. Use Edit or Delete in the Actions column to revise or remove a row. Deleting asks you to confirm, then removes that norm from the standard's definition; the standard itself is untouched.

Before you add anything, the section shows the empty state No requirements yet, with an Add requirement action for those who can use it.

Link the policies it underpins

The Linked policies section ties the standard to the policies that depend on it, so your policy framework points at concrete technical norms rather than at intentions alone. Linking needs a Manager or Admin role, and the Link policy button only appears when you can also read policies and the Policies module is switched on — so a link can never lead somewhere you cannot go.

  1. Select Link policy. A picker opens with a Search policies... box and a list of your policies, each showing its title and lifecycle status.
  2. Find the policy you want and select Link policy beside it. A Policy linked confirmation appears, the picker closes, and the policy joins the section. Policies already attached are marked Already linked and cannot be added twice.
  3. To remove a link, select Unlink on a linked row and confirm. Only the connection goes — the policy itself is untouched. Each linked row shows the policy's title, which is a link through to its page in Policies, with the policy's own lifecycle status beside it.

Until a policy is linked, the section shows the empty state No policies linked yet, with a Link policy action for those who can use it.

Edit, retire or delete a standard

  1. On the detail page, select Edit. The same dialogue opens, now titled Edit security standard and carrying an extra Status field.
  2. Change what you need — including moving the standard to Active once it is in force, or to Retired when it is withdrawn — then select Save. A Security standard updated confirmation appears and the header reflects the change at once.
  3. To remove a standard altogether, select Delete and confirm. Aegis warns that the standard's requirements and policy links will no longer be visible, then returns you to the overview. Prefer Retired over deletion whenever you want to keep a record of a norm you no longer enforce.

The AI assist

The Review & Map button on a standard's detail page opens the AI Baseline & Requirement Review. It reads the standard, its requirements and your real policy catalogue, then reports on four things: baseline and version drift for the scope described, a requirement-by-requirement review with tightened interpretations, the policies those requirements should map to, and a drafted, tighter description you could adopt. Interpreted requirements keep the language they were written in.

  1. Select Review & Map. A panel opens explaining what the review covers, with a note stating that it reasons over internal records only and is a proposal for the standard's owner to review and adjust — not a binding decision.
  2. Select Run review. Progress is shown step by step as Aegis loads the record, checks for drift, reviews each requirement, matches requirements to your configured policies and drafts its findings. The text streams into the panel as it is written.
  3. Read the result. Above the text the panel shows a Confidence reading, and beneath it the evidence the review drew on, so you can judge how much weight it deserves.
  4. Take the output somewhere useful from the footer: Copy, Export DOCX or Export PDF for the text itself, Save as record to keep the review against this standard, and — for roles that may create them — Create action item or Create work items to turn a finding into tracked work. Run again repeats the review.
The review never changes the standard on its own

The analysis writes nothing back to the standard: no requirement is added or reworded, no policy is linked, and the description is not replaced. If you agree with a suggestion, you apply it yourself through Edit, Add requirement or Link policy. The footer actions do save things, but only when you choose them — Save as record keeps a copy of the review alongside the standard, and Create action item and Create work items open a form you fill in and confirm first. A person reviews and decides; the assistant never acts on its own.

Every role that can read a standard can run the review, provided your licence includes the AI assist. Each run counts against your organisation's AI credits; once they are exhausted the review stops with a plain Action failed message, and you check the remaining balance on your licence page rather than in this panel. Where your organisation has published an AI transparency notice, the review is refused until you have acknowledged that notice. Every run is recorded in the audit log.

Tips and limits

Where this connects