iam-architect
Architect-level identity and access design in the cloud — human versus machine identities, role and permission boundaries, federation over keys, least privilege that can actually be maintained, break-glass access, and the audit trail that proves who did what. Use when designing accounts, roles, or service credentials.
- License 许可
- MIT
- 最近更新
- 2026-11-13
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- iam-architect
- license
- MIT
- metadata
- { "crusor-path": "cloud-and-devops-architect/identity-architect/iam-architect", "crusor-domain": "cloud-and-devops-architect", "crusor-origin": "original", "crusor-sources": "docs.aws.amazon.com/IAM/latest/UserGuide; cloud.google.com/iam/docs/overview; learn.microsoft.com/entra/identity", "crusor-triggers": "iam, roles, permissions, least privilege, service account, federation, oidc, break glass, access review, policy boundary", "crusor-verified-on": "2026-10-30" }
- description
- Architect-level identity and access design in the cloud — human versus machine identities, role and permission boundaries, federation over keys, least privilege that can actually be maintained, break-glass access, and the audit trail that proves who did what. Use when designing accounts, roles, or service credentials.
- compatibility
- AWS IAM, Entra ID, GCP IAM, any OIDC federation
03 SKILL.md 正文
IAM Architect
Identity is the real perimeter. Network controls are advisory once a credential is stolen; the permission set is what actually limits the damage.
When to use
- Designing roles for humans and workloads.
- Choosing between long-lived keys and federation.
- An audit or incident asks "who could have done this".
When NOT to use
- Application-level authorization (tenant roles, feature access) → web
authz-models-architect. - Secrets storage mechanics →
secrets-management-architect.
Decision 1 — Identity classes
| Class | Credential | Lifetime |
|---|---|---|
| human, interactive | SSO federation | session, hours |
| human, privileged | SSO + step-up MFA | minutes |
| workload in cloud | instance/workload identity | automatic, rotated |
| workload outside cloud | OIDC federation | minutes |
| CI/CD | OIDC to provider | minutes |
| break-glass | hardware key, sealed | manual, audited |
- Static access keys are a legacy exception requiring an owner and an expiry date.
- One identity per workload — shared identities destroy attribution and blast-radius control.
Decision 2 — Permission shape
- Start from deny; grant by role at the smallest scope that runs the workload.
- Use roles with conditions rather than wildcards: resource tags, source network, MFA presence, time.
- Permission boundaries / guardrails stop a role from escalating itself — the most important control in any delegated-admin model.
- Managed policies first, custom only with a documented reason and a review date.
- Separate planes: control-plane admin (create/delete) from data-plane access (read/write). Most humans need neither full-time.
Decision 3 — Federation
- OIDC for CI: the provider issues short-lived credentials scoped to a repo/branch.
- Verify the token's audience and subject conditions — a permissive trust policy is an open door.
- Human access via the IdP; no local IAM users with console passwords.
- Cross-account access by role assumption with an external id, not shared keys.
Decision 4 — Privileged access
- Break-glass role: no standing access, alert on use, credentials held outside the normal path, and a post-use review.
- Just-in-time elevation with approval; standing admin is the default failure.
- Session recording or command logging for production admin.
- Offboarding is automated from the IdP — a departed human's roles die with the account.
Decision 5 — Audit
- Every identity action logged to an immutable store, retained per policy.
- Access reviews quarterly: unused roles, unused permissions, stale keys.
- Alert on: new admin role, policy attached to a principal, root/owner login, key creation.
- Attribution: the log must name the human behind the workload role when possible.
Failure modes at scale
- One
deployrole with*on*, used by every pipeline. - Access keys from 2019 still active in a CI secret store.
- Trust policy accepting any repo in the org, so a fork can assume the role.
- Break-glass credentials in a shared password manager nobody audits.
- Hundreds of roles, none removable because "something might use it".
Verification receipts
- Identity inventory: class, credential type, rotation, owner.
- Least-privilege check: unused-permission report with removals applied.
- Federation trust policies reviewed for audience/subject conditions.
- Break-glass use rehearsed and alert verified.
- Audit log query demonstrating attribution for a sensitive action.
04 同仓库更多 Skills
interaction-patterns-architect
Architect-level interaction design in code — choosing established patterns over invented ones, feedback and latency behaviour, destructive-action safeguards, navigation and overlay conventions, and the state model behind every interactive surface. Use when a screen has to behave, not just look.
design-system-ops-architect
Architect-level operation of a design system in code — token pipelines and versioning, component API and variant policy, the catalogue as the contract, adoption metrics, and the governance that stops a system from forking into three. Use when a component library must survive several teams.
design-handoff-architect
Architect-level design handoff — what must be specified before implementation starts, tokens over raw values, responsive and content-limit contracts, state matrices, and the review loop that catches drift between design and code. Use when a design is about to become code.
content-microcopy-architect
Architect-level UI content — where strings live, voice and terminology consistency, error and empty-state copy as part of the interface, translation-ready authoring, and the content model a product needs before it multiplies screens. Use when the words in the interface matter as much as the pixels.
ui-ux-implementation-architect
Architect-level direction for turning a design into working UI — the handoff contract, component ownership versus screen-level layout, interaction and state conventions, design QA gates, and the review habits that keep an implementation faithful without frozen pixels. Use before any specific UI implementation skill.
test-strategy-architect
Architect-level test strategy — allocating assertions across unit, integration, contract and end-to-end layers, designing for testability through dependency boundaries, fixture and data strategy, and the flake and coverage policies that keep a suite trustworthy. Use when defining how a codebase is tested.