---
title: Amazon Web Services API Integration on Truto
slug: aws
category: Cloud Storage
canonical: "https://truto.one/integrations/detail/aws/"
---

# Amazon Web Services API Integration on Truto



**Category:** Cloud Storage  
**Status:** Generally available

## Unified APIs

### Unified Cloud Infrastructure API

- **Access Grants** — Who can do what, and at which level of the hierarchy -- organisation, grouping, account, or resource.
- **Account Groupings** — A node in the account hierarchy -- an organisational unit, a management group, or a folder.
- **Accounts** — An account, subscription, or project. Enumerated across the whole organisation, not limited to the connected account.
- **Audit Trail Configs** — Whether an audit trail (change-logging configuration) exists and is switched on. Providers model this very differently: one estate-wide trail, per-scope routing, or tiered defaults.
- **Block Volumes** — Block storage volumes that attach to virtual machines (AWS EBS volumes, Azure managed disks, GCP persistent disks), with encryption at rest and where each volume is attached.
- **Cdn Distributions** — Content delivery network distributions (AWS CloudFront distributions, Azure Front Door / CDN endpoints, GCP Cloud CDN): viewer TLS, origin protection, firewall and logging. Global: region is "global".
- **Certificates** — TLS certificates managed by the provider (AWS Certificate Manager, Azure Key Vault / App Service certificates, GCP Certificate Manager): expiry, key algorithm, renewal and use. Regional: AWS certificates used by CloudFront live in us-east-1.
- **Code Repositories** — Hosted source code repositories (AWS CodeCommit; Azure Repos; GCP Secure Source Manager): default branch, encryption key and clone endpoints. Regional.
- **Compute Instances** — Virtual machines (AWS EC2 instances, Azure virtual machines, GCP Compute Engine instances), with their network exposure, metadata-service hardening, machine identity, the image they were launched from and whether a fleet-management agent manages them. Disk encryption lives on block_volumes (attachments.instance_id).
- **Container Clusters** — Container orchestration clusters that are not Kubernetes (AWS ECS clusters; Azure Container Apps environments; GCP Cloud Run has no cluster object - its services run straight in a region). Kubernetes clusters are kubernetes_clusters.
- **Container Services** — Long-running container services that keep a set of tasks running from one task definition (AWS ECS services, Azure Container Apps apps, GCP Cloud Run services).
- **Container Task Definitions** — Container task definitions: the versioned blueprint of containers, images, roles and settings that tasks and services run (AWS ECS task definitions; Azure Container Apps and GCP Cloud Run revisions). The list has one row per definition, at its latest active revision. Environment variable VALUES are never returned - only their names; secrets appear only as references.
- **Container Tasks** — Running (or recently stopped) container tasks: one running instance of a task definition (AWS ECS tasks; Azure Container Apps replicas; GCP Cloud Run job tasks). Environment variable values from start-time overrides are never returned - only names.
- **Data Snapshots** — Database and volume snapshots, including who owns the source data and who the snapshot has been shared with.
- **Data Warehouses** — Managed analytics data warehouses (AWS Redshift provisioned clusters, Azure Synapse dedicated SQL pools, Google BigQuery datasets), with encryption at rest, snapshot retention and public network access.
- **File Systems** — Managed network file systems (AWS EFS file systems, Azure file shares, GCP Filestore instances), with encryption at rest and automatic backup state.
- **Firewall Rule Sets** — Security groups, network ACLs, and firewall policies — primitives with different semantics, distinguished by scope_type.
- **Human Identities** — A user that can sign in, whether a person or a shared login. On AWS this spans both IAM users and IAM Identity Center users, which share no identifiers.
- **Instance Groups** — Groups of identical virtual machines kept at a size by the provider (AWS Auto Scaling groups, Azure virtual machine scale sets, GCP managed instance groups), with their size, placement, health checks, launch definition and members.
- **Kubernetes Clusters** — Managed Kubernetes clusters (AWS EKS clusters, Azure AKS clusters, GCP GKE clusters): control-plane version, API endpoint exposure, secrets encryption and audit logging.
- **Kubernetes Node Pools** — Managed Kubernetes worker node pools (AWS EKS managed node groups, Azure AKS node pools, GCP GKE node pools): Kubernetes and node image versions, scaling and remote access.
- **Load Balancers** — Managed load balancers (AWS Application, Network and Gateway Load Balancers; Azure Load Balancer and Application Gateway; GCP load balancers) with their listeners and each listener's TLS policy. Listeners are embedded: they have no identity outside their load balancer for posture purposes, the TLS question is asked per load balancer, and the provider reads them one load balancer at a time either way.
- **Log Destinations** — Where logs are kept, and for how long. Retention doubles as evidence for a data-retention schedule.
- **Machine Images** — Machine images the account owns (AWS AMIs, Azure managed images and gallery images, GCP custom images), with whether they are public, encrypted and deprecated. A GET reads any image the account can see by id, including Amazon and Marketplace images.
- **Managed Instances** — Machines registered with a fleet-management service (AWS Systems Manager managed nodes, Azure Arc-enabled servers, GCP VM Manager), with agent health and OS details. Use this to tell which compute_instances can be patched and inspected. On AWS stopped and terminated nodes are not listed.
- **Managed Keys** — Encryption keys held in the provider's key management service (KMS, Key Vault, Cloud KMS), including rotation and access-policy state.
- **Message Queues** — Managed message queues (SQS queues, Service Bus queues, Pub/Sub subscriptions), with server-side encryption and access-policy posture for each queue.
- **Monitoring Alarms** — Metric alarms and composite alarms (AWS CloudWatch alarms, Azure Monitor alert rules, GCP Cloud Monitoring alert policies), with what they watch, their current state, what they do when they fire, and whether a firing alarm reaches a confirmed subscriber.
- **Network Boundaries** — Virtual networks and their subnets, normalized across providers whose networks may be regional or genuinely global.
- **Network Links** — Network peerings, transit attachments, and shared-network connections between boundaries, potentially crossing account boundaries.
- **Nosql Tables** — Managed NoSQL tables (DynamoDB tables, Cosmos DB containers, Azure Table storage tables, Bigtable tables, Firestore databases), with encryption, point-in-time recovery and replication posture for each table.
- **Notification Topics** — Publish/subscribe notification topics (AWS SNS topics, Azure Event Grid topics, GCP Pub/Sub topics), with encryption at rest and how many subscriptions are confirmed. monitoring_alarms actions reference these.
- **Object Stores** — Bucket and blob storage across providers, with a normalized public-access and encryption posture for each store.
- **Password Policies** — The password policy as the provider exposes it, reported at the level it actually applies (account, tenant, org unit, etc).
- **Permission Definitions** — The contents of a custom role or policy, including wildcards and negated forms (NotAction, NotResource) preserved verbatim rather than expanded.
- **Policy Compliance States** — The evaluation result of a policy rule that reports non-compliance rather than blocking it.
- **Policy Constraints** — Preventative policies (e.g. AWS SCPs, Azure Policy) applied across the org, one row per policy per scope node, including inherited attachments.
- **Posture Findings** — One row per control evaluation per resource, from the provider's own security product (Security Hub, Defender for Cloud, Security Health Analytics).
- **Relational Databases** — Managed relational database clusters and standalone instances, with security posture (encryption, backups, network access) read at the right level.
- **Scan Coverages** — What proportion of the estate is actually being scanned. Zero findings on an unscanned estate must not be read as a clean estate.
- **Security Product States** — Whether the provider's own security product is switched on, per product, per account, per region.
- **Serverless Functions** — Serverless functions (AWS Lambda functions, Azure Functions, GCP Cloud Functions), with their runtime, execution identity, network placement and environment variable NAMES. Environment variable values are never returned, not even in remote_data.
- **Static Credentials** — Long-lived credentials held by a principal -- access keys, client secrets, certificates, SSH keys, and service account keys.
- **Threat Findings** — Runtime security alerts from the provider's threat detection product (GuardDuty, Defender alerts, Event Threat Detection): observed activity, not configuration drift.
- **Vulnerability Findings** — Package and host vulnerabilities from the provider's own scanner. Read alongside scan_coverages to know how much of the estate was actually scanned.
- **Workload Identities** — Machine identities -- roles, managed identities, service principals, and service accounts -- and who outside the estate is permitted to use them.

## MCP-ready AI tools

Truto exposes 189 tools for Amazon Web Services that AI agents can call directly.

- **list_all_amazon_web_services_organizations_accounts** — Every account in the organization, in one global sweep — never fan out per region. Account.Paths materialises grouping_path at zero extra cost, so the recursive ListParents climb is only a fallback. Map state from Account.State, not Account.Status, which retires 2026-09-09 and has no CLOSED value. JoinedTimestamp is the org-join date, not the creation date. An empty page is not a terminator: loop until NextToken is null. A standalone account returns AWSOrganizationsNotInUseException (400).
- **list_all_amazon_web_services_organizations_roots** — The organization root. Exactly one call ever — the quota 'Roots in an organization: 1' is not adjustable, and the throttle is 1 req/s. Root has no Path field (only OUs do), so synthesize o-<orgId>/r-<rootId>/. No created_at exists on Root.
- **list_all_amazon_web_services_organizations_ous** — Organizational units under one parent, walked breadth-first. Nesting is capped at 5 levels and is not adjustable, so recursion is bounded and needs no cycle guard. Preferred over ListChildren because it returns {Arn, Id, Name, Path} inline and avoids a DescribeOrganizationalUnit per OU at a 2 req/s throttle. No created_at exists on OrganizationalUnit either.
- **list_all_amazon_web_services_organizations_parents** — The direct parent (an OU or the root) of one account or OU, given as child_id. Every account or OU has exactly one parent, so this always returns a single-element list -- chain calls (feeding the returned Id back in as the next child_id) to walk up to the root instead of expecting a full ancestor chain in one call. Requires the caller to already know a starting child_id (from organizations-accounts or organizations-ous); there is no reverse lookup by name.
- **list_all_amazon_web_services_organizations_accounts_for_parent** — Accounts directly under one root or OU, given as parent_id -- the org-tree-scoped counterpart to organizations-accounts (which lists every account in the org in one global sweep). Returns only direct children: accounts nested under a child OU of this parent are not included and need a separate call per child OU. Same Account shape as organizations-accounts, including the State vs deprecated Status distinction.
- **list_all_amazon_web_services_organizations_policies** — Service control policies. Filter is required and single-valued, so RCPs are a second call with Filter=RESOURCE_CONTROL_POLICY. Management account or delegated administrator only. Filter out the AWS-managed p-FullAWSAccess or target counts are dominated by noise.
- **get_single_amazon_web_services_organizations_policy_by_id** — One SCP with its document. Policy.Content is plain JSON and must NOT be URL-decoded — the opposite of IAM's GetPolicyVersion. Applying the IAM decode here mangles the policy.
- **list_all_amazon_web_services_organizations_policy_targets** — Where one SCP is directly attached. Returns direct attachments only; inheritance must be computed from the OU tree because DescribeEffectivePolicy explicitly excludes SCPs — its PolicyType enum has no SERVICE_CONTROL_POLICY entry.
- **list_all_amazon_web_services_iam_account_authorization_details** — The primary IAM collector. One paginated sweep returns every user, group, role and policy with GroupList, AttachedManagedPolicies, inline policy documents, Tags and PermissionsBoundary — replacing four of the eight per-user calls. It does NOT cover access keys, MFA devices or login profiles. IAM policy documents here are URL-encoded (RFC 3986): decode only if the string contains %7B, never if it starts with {.
- **list_all_amazon_web_services_iam_users** — IAM users. Returns UserId, UserName, Arn, Path, CreateDate and PasswordLastUsed only — no Tags, no PermissionsBoundary. IAM users have no email and no display name anywhere in the API. PasswordLastUsed is console sign-in only, records at most one sign-in per 5-minute window, began 2014-10-20, and has a documented gap from 2018-05-03 22:50 PDT to 2018-05-23 14:08 PDT. A key-only user reports no_information forever — that is not 'never signed in'.
- **list_all_amazon_web_services_iam_roles** — IAM roles. ListRoles DOES return AssumeRolePolicyDocument — the omission list names only PermissionsBoundary, RoleLastUsed and Tags — so GetRole is needed only for last_used_at, boundary and tags. Roles have no long-term credentials and no API lists any, so has_static_keys is structurally false and key_count structurally 0.
- **list_all_amazon_web_services_iam_policies** — Managed policies. Scope=Local means customer-managed (is_custom). The response already carries DefaultVersionId, so GetPolicy can be skipped and GetPolicyVersion called directly. AttachmentCount (grants) and PermissionsBoundaryUsageCount (restricts) are separate numbers — summing them destroys a security-relevant distinction.
- **get_single_amazon_web_services_iam_account_password_policy_by_id** — The account password policy. A 404 NoSuchEntity here means no policy is configured, so AWS defaults apply — a WEAKER posture. Render it as 'not configured', never as 'unknown', and never as a pass. IAM has no lockout threshold and no breached-password list at all.
- **create_a_amazon_web_services_iam_generate_credential_report** — Starts the credential report job. WARNING: iam:GenerateCredentialReport is access level Write in the Service Authorization Reference, so it will not appear in a policy built by filtering for read access. Without it GetCredentialReport returns 410. Enabling this is a deliberate decision, not a default.
- **get_single_amazon_web_services_iam_credential_report_by_id** — The credential report: base64 CSV with sentinels N/A, no_information, not_supported and uppercase TRUE/FALSE. Returns 410 if nobody generated one. It caps at the first two access keys per user and omits service-specific credentials entirely, so it is not the source of truth for static_credentials.
- **list_all_amazon_web_services_iam_access_keys** — Access keys for one user, max 2. Access keys never expire — expires_at is structurally null, not unknown. last_used_at needs a separate GetAccessKeyLastUsed call per key; access keys are the only credential type that has last-used data at all.
- **list_all_amazon_web_services_iam_mfa_devices** — MFA devices for one user. The response carries NO device type — type must be inferred from the SerialNumber shape, and FIDO/passkey vs TOTP is not cleanly distinguishable. Mark any mfa_methods value as derived.
- **get_single_amazon_web_services_iam_login_profile_by_id** — Console password metadata for one user. A 404 NoSuchEntity is the ANSWER, not an error: it means has_console_access is false. Surfacing it as an integration error turns every programmatic-only user into noise.
- **list_all_amazon_web_services_iam_oidc_providers** — OIDC identity providers in the account. Returns ARNs only — the issuer URL and audience list need GetOpenIDConnectProvider per ARN. Not paginated.
- **get_single_amazon_web_services_iam_oidc_provider_by_id** — One OIDC provider. Url is stored WITHOUT the https:// scheme while trust-policy condition keys use the bare host — normalise both sides before joining. ClientIDList is the registered audience and can disagree with the :aud condition key; the trust policy is what actually gates the assume. A role trusting this provider with no :sub condition can be assumed by any subject from the issuer — model 'no sub condition' as its own state, never as an empty list.
- **list_all_amazon_web_services_iam_saml_providers** — SAML identity providers in the account. Returns ARNs, ValidUntil and CreateDate. Not paginated. A federated trust to one of these is an external-trust edge on any role whose trust policy names it under Principal.Federated.
- **list_all_amazon_web_services_securityhub_findings** — Lists ASFF findings via AWS Security Hub GetFindings (POST /findings). No required params beyond region/account_id; optional truto_time_filter narrows results server-side. Paginated (limit, next_cursor; AWS MaxResults capped at 100). A 404 here (or the underlying InvalidAccessException, originally HTTP 401) means Security Hub is not enabled for this account/region — that is normal DATA, not an integration failure. Check ListFindingAggregators first: if cross-region aggregation is on, calling the aggregation region returns everything and fanning out duplicates it all. Compliance.Status has only PASSED|WARNING|FAILED|NOT_AVAILABLE — there is NO NOT_APPLICABLE, and NOT_AVAILABLE conflates 'not applicable' with 'could not check'. Findings whose Config evaluation returned NOT_APPLICABLE are auto-archived after 3 days, so RecordState must explicitly include ARCHIVED (this method's fixed Filters already does) or they vanish; archived findings are permanently deleted after 30 days. Findings over 240 KB have Resource.Details silently stripped.
- **list_all_amazon_web_services_securityhub_enabled_standards** — Lists standards enabled in this region/account via AWS Security Hub GetEnabledStandards (POST /standards/get). No required params beyond region/account_id. Paginated (limit, next_cursor; AWS MaxResults capped at 100). A 404 (or the underlying InvalidAccessException, originally HTTP 401) means Security Hub is not enabled here — that is normal DATA, not an integration failure. enabled_standards is a Security Hub-only concept — the other seven AWS security services have no notion of a compliance standard, and their feature toggles must not be forced into this field. standardVersion is not a field: the version is embedded in the StandardsArn (.../cis-aws-foundations-benchmark/v/1.2.0) and must be string-parsed.
- **get_single_amazon_web_services_securityhub_hub_by_id** — Reports whether Security Hub is enabled in this region/account and which finding mapping applies, via AWS Security Hub DescribeHub (GET /accounts). No required params beyond region/account_id. A 404 here (or the underlying InvalidAccessException, originally HTTP 401) is DATA — Security Hub is simply not enabled here — not an integration failure. ControlFindingGenerator branches the whole finding mapping: SECURITY_CONTROL uses Compliance.SecurityControlId and Compliance.AssociatedStandards; STANDARD_CONTROL has those absent and needs ProductFields.StandardsArn or StandardsGuideArn and ProductFields.ControlId or RuleId. SubscribedAt is enablement time, not an evaluation time — it must not be mapped to last_evaluated_at.
- **list_all_amazon_web_services_securityhub_security_controls** — Fetches control metadata via AWS Security Hub BatchGetSecurityControls (POST /securityControls/batchGet), fetched once per collection and cached — never per finding. Required: security_control_ids, one control id like 'IAM.1' (AWS's own API accepts 1-25 ids as an array). Known limitation, verified live: this proxy method currently interpolates security_control_ids as a literal string into the request body, so AWS rejects every input — a single id, a comma-separated list, or a JSON array string — with 400 InvalidInputException ('Expected an array, value was ...'). Calls to this method fail regardless of input until the config is fixed to parse the value into a JSON array. A 404 (or InvalidAccessException) instead means Security Hub itself is not enabled. This is the only source of SeverityRating, the control's inherent risk; a finding's own Severity.Label is INFORMATIONAL on every PASSED finding because it describes the outcome, not the control.
- **list_all_amazon_web_services_guardduty_detectors** — Detector ids in this region via AWS GuardDuty ListDetectors — one detector per region per account. Returns ids only, not detector objects; hydrate each id with guardduty-detector get. No required params beyond region/account_id. Live-verified: swept 14 regions on the sandbox account and every region returned an empty list (GuardDuty is not enabled anywhere on this account) — the call itself succeeds (200, empty array), so treat an empty result as "GuardDuty not enabled here", not a failure. GuardDuty is regional: sweep every region you care about rather than assuming one region's presence or absence tells you about another.

- **list_all_amazon_web_services_guardduty_list_findings** — Finding ids, max 50 per page, via AWS GuardDuty ListFindings. Requires detector_id from guardduty-detectors. Returns ids only; call guardduty-get-findings to get the findings themselves. ListFindings returns archived and unarchived findings together unless you filter, so set truto_time_filter explicitly. The optional `truto_time_filter` query parameter is a JSON-encoded object merged into the request body at the root, so pass the whole fragment AWS expects, i.e. {"findingCriteria": {"criterion": {...}}}. Field names must be lowerCamelCase exactly as GuardDuty's JSON API defines them (findingCriteria, criterion, gte/eq/...), e.g. {"findingCriteria":{"criterion":{"severity":{"gte":7}}}}; PascalCase (FindingCriteria/Criterion/MaxResults) is rejected by AWS with "JSON could not be processed".
- **get_single_amazon_web_services_guardduty_get_finding_by_id** — Gets a finding by id via AWS GuardDuty GetFindings. Requires detector_id (from guardduty-detectors) and finding_ids (from guardduty-list-findings). Pass one finding id per call: several finding_ids values are sent as one comma-joined id, which GuardDuty rejects. severity is a number with no severityLabel field. There is no state field; the closest is service.archived, and service itself is optional. title is optional, so fall back to type. resource has no single shape: resource.resourceType selects one of about 20 sub-objects, some of them arrays.
- **list_all_amazon_web_services_guardduty_get_findings** — GuardDuty GetFindings for up to 50 finding ids on the detector in detector_id, as {result: [finding, ...], next_cursor: null}. finding_ids may be repeated or indexed keys, finding_ids[], a comma string or a JSON array string. More than 50 is GuardDuty's own 400. Needs guardduty:GetFindings.
- **get_single_amazon_web_services_guardduty_detector_by_id** — One detector's configuration via AWS GuardDuty GetDetector. Requires the detector id (path `id`) from guardduty-detectors — GetDetector rejects an id from a different account/region with "detectorId is not owned by the current account", so do not reuse an id across regions. status is the enablement signal. updatedAt is a configuration-change timestamp, NOT an evaluation timestamp — mapping it into last_evaluated_at would be a lie to the consumer. Live-verified error paths only (no live detector found in the sandbox account, see guardduty-detectors); the response shape below is from the AWS API reference, not a live response.

- **list_all_amazon_web_services_inspector_2_account_status** — Inspector v2 enablement per account via AWS Inspector2 BatchGetAccountStatus. The caller-controlled query param `truto_inspector_accounts` (JSON-encoded {"accountIds":[...]}) is OPTIONAL — live-verified: omitting it entirely still returns the calling account's own status (AWS defaults accountIds to the caller's account when empty). Live-verified live on the sandbox account (471112750856) across 14 regions: Inspector2 is DISABLED account-wide in every region tried, for every resource type (ec2, ecr, lambda, lambdaCode, codeRepository). state.status is the top-level enablement signal; resourceState breaks it down per scan type, each with its own status (and, only on error, errorCode/errorMessage). There is NO timestamp anywhere in this response, so last_evaluated_at is unreadable for Inspector — record it as such rather than substituting a config timestamp.

- **list_all_amazon_web_services_inspector_2_coverage_statistics** — Scan coverage counts per resource type via AWS Inspector2 ListCoverageStatistics, grouped by RESOURCE_TYPE (the config's body is static: groupBy=RESOURCE_TYPE, not caller-controlled). No required params beyond region/account_id. Live-verified: this operation is real and works — called live on the sandbox account across regions and got 200 with an empty array each time (Inspector2 is disabled account-wide, so there is nothing to have coverage over; an empty result here is DATA, not a failure). Counts reflect only resource types Inspector2 actually scans (EC2, ECR, Lambda, Lambda code, code repositories) — a resource type with zero enabled scanning simply has no entry, it is not reported as a zero count.

- **list_all_amazon_web_services_inspector_2_findings** — Vulnerability findings via AWS Inspector2 ListFindings. The caller-controlled query param `truto_time_filter` (JSON-encoded object merged into the request body — see the guardduty-list-findings gotcha on the same $truto_merge mechanism) is OPTIONAL; omitting it lists all findings with no filter. No required params beyond region/account_id. Live-verified: this operation is real and works — called live on the sandbox account across regions and got 200 with an empty array each time (Inspector2 is disabled account-wide, so there are no findings; an empty result here is DATA, not a failure). Not paginated by this the connector method beyond following nextToken automatically; severity/status here follow Inspector2's own model (COMPLETE/ACTIVE lifecycle, CRITICAL/HIGH/MEDIUM/LOW/INFORMATIONAL/UNTRIAGED severity) which is unrelated to GuardDuty's raw numeric severity.

- **list_all_amazon_web_services_config_recorders** — Calls AWS Config DescribeConfigurationRecorders to list configuration recorders in this region. No required params beyond region/account_id. NOT paginated. NoSuchConfigurationRecorderException is DATA — AWS Config is off — not a failure. AWS now allows only one recorder per account/region, so this list has at most one item.
- **list_all_amazon_web_services_config_recorder_status** — Calls AWS Config DescribeConfigurationRecorderStatus for whether the recorder is actually recording. No required params beyond region/account_id. NOT paginated. lastStatusChangeTime is the only real timestamp in this response - there is no configSnapshotDeliveryInfo or delivery-channel data here (that belongs to the separate, legacy DeliveryChannelStatus API, which this resource does not call); do not expect those fields. name and arn can both be absent for a service-linked recorder, which instead carries servicePrincipal.
- **list_all_amazon_web_services_config_aggregate_compliance** — Calls AWS Config DescribeAggregateComplianceByConfigRules. Requires aggregator_name (AWS's ConfigurationAggregatorName) - omit it and AWS Config rejects the literal, unresolved placeholder text as the aggregator name with a ValidationException, not the normal NoSuchConfigurationAggregatorException, which is confusing to debug. Per-(rule, account, region) compliance across the whole organization in one paged call, with no AssumeRole per account - Limit up to 1000, cursor-paginated. Counts arrive as {CappedCount, CapExceeded}, not an integer: when CapExceeded is true the count is a CEILING, and mapping it as exact publishes wrong numbers on exactly the largest rules. INSUFFICIENT_DATA is returned but cannot be used as a filter value, and the aggregate variant accepts only COMPLIANT and NON_COMPLIANT as filters despite the enum listing four.
- **list_all_amazon_web_services_config_rule_evaluation_status** — Calls AWS Config DescribeConfigRuleEvaluationStatus. No required params beyond region/account_id. Cursor-paginated, Limit up to 50 rules per page. The ONLY source of real Config timestamps for a rule. DescribeComplianceByConfigRule carries no timestamp at all, so pairing this call in is mandatory - substituting our own fetch time would be exactly the staleness lie this model exists to prevent. Use LastSuccessfulEvaluationTime (evaluation), not LastSuccessfulInvocationTime (invocation). Excludes AWS Config Custom Lambda rules from status reporting entirely - only Managed rules and Custom Policy rules appear here.
- **list_all_amazon_web_services_accessanalyzer_analyzers** — IAM Access Analyzer analyzers. status == ACTIVE is the enablement signal, and this is one of only two of the eight security services with a genuine last_evaluated_at: lastResourceAnalyzedAt.
- **get_single_amazon_web_services_macie_session_by_id** — Reports whether Macie is enabled via AWS Macie GetMacieSession (GET /macie). No required params beyond region/account_id. Verified live across 11 regions: when Macie is not enabled, AWS returns HTTP 403 with body {"message": "Macie is not enabled"} — treat that as absence-of-data, not an integration failure. status PAUSED means enabled-but-suspended, not disabled — do not treat PAUSED as 'not enabled' (per AWS docs; not independently verified live since Macie is not enabled anywhere in this sandbox account).
- **list_all_amazon_web_services_identitystore_users** — Identity Center users. ListUsers returns complete User objects — UserId, UserName, DisplayName, Emails, UserStatus and CreatedAt — so DescribeUser adds nothing for bulk sync. MFA registration state is NOT AVAILABLE through any public API (the console uses an undocumented sso-directory operation with no published wire contract). last_sign_in_at is NOT AVAILABLE either — only CloudTrail UserAuthentication events carry it. has_static_credentials and has_console_access are IAM-only concepts here.
- **list_all_amazon_web_services_sso_instances** — Identity Center instances — the bootstrap for IdentityStoreId, InstanceArn, PrimaryRegion and Regions. Capped at 10 instances; do not blindly take [0]. Calling in the WRONG region returns an empty list, not an error, which is indistinguishable from 'the customer does not use Identity Center' — sweep candidate regions rather than trusting one empty result. Note the host is sso.<region>.amazonaws.com, not sso-admin.*.
- **list_all_amazon_web_services_s_3_buckets** — All buckets. Pagination is MANDATORY, not an optimisation: unpaginated requests are rejected outright for accounts with a bucket quota above 10,000, and BucketRegion is only returned on paginated requests — without it every bucket needs an extra GetBucketLocation. BucketArn is now returned; use it as provider_ref rather than constructing arn:aws:s3:::<name>, which is wrong outside the aws partition. Directory buckets (S3 Express) are invisible here — a disjoint namespace under s3express-control.
- **list_all_amazon_web_services_s_3_bucket_public_access_block** — Bucket-level Block Public Access. NoSuchPublicAccessBlockConfiguration (404) means not configured — all four flags false — and must not be read as blocked. BlockPublicAcls and BlockPublicPolicy are WRITE-TIME GUARDS ONLY: AWS states enabling them does not affect existing policies or ACLs. Only IgnorePublicAcls and RestrictPublicBuckets change effective read access, so computing exposure from the two Block flags produces false clean results. Effective BPA is the account setting OR the bucket setting, per flag. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_policy_status** — Whether the bucket policy grants public access. This is the ONLY policy-derived signal available cross-account: GetBucketPolicy returns 405 Method Not Allowed for a caller outside the bucket owner's account regardless of IAM. Errors 404 NoSuchBucketPolicy when there is no bucket policy at all — the same error code GetBucketPolicy throws, NOT an empty document (live-verified; corrects an earlier claim here). Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_policy** — The bucket policy document. NoSuchBucketPolicy (404) means no policy, not a failure. Cross-account this returns 405 Method Not Allowed regardless of IAM — map 405 to permission_denied, never to 'no policy'. The document is plain JSON, NOT URL-encoded. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_acl** — Bucket ACL grants. Public means a Grantee.URI of http://acs.amazonaws.com/groups/global/AllUsers or .../global/AuthenticatedUsers. .../s3/LogDelivery is benign — do not flag it. AuthenticatedUsers is not anonymous (requests must be signed) but means any AWS account on earth, so keep it distinguishable rather than folding it into anonymous access. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_encryption** — Default encryption. Since January 2023 SSE-S3 is applied to every bucket by default, so 'no default encryption' is an obsolete control and this rarely 404s — the real control is SSEAlgorithm == AES256, i.e. not a customer-managed key. SSEAlgorithm has FOUR values: AES256, aws:kms, aws:kms:dsse, aws:fsx. KMSMasterKeyID may be a key id, an ARN or an ALIAS — store the raw string with a discriminator. The docs are internally inconsistent (KMSMasterKeyID vs KMSKeyID); parse both. ServerSideEncryptionConfigurationNotFoundError has no documented Errors section, so treat an unrecognised 404 as not_collected, never as a pass. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_versioning** — Versioning state. Returns an EMPTY document with no Status element at all when versioning was never enabled — absent is not the same as Suspended. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_object_lock** — Object Lock. Returns only ObjectLockEnabled (single valid value Enabled) and an optional default rule. retention_locked is NOT a real S3 field: the closest honest mapping is Mode == COMPLIANCE, which is a default for FUTURE objects and says nothing about what is in the bucket now. Legal holds and per-object retention are per object version and would need unbounded per-object calls. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_lifecycle** — Lifecycle rules. NoSuchLifecycleConfiguration (404) means no rules, not a failure. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_logging** — Server access logging. Returns an EMPTY document with no LoggingEnabled element when logging is off, rather than erroring. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_replication** — Replication rules and destination buckets. ReplicationConfigurationNotFoundError has no documented Errors section, so an unrecognised 404 must degrade to not_collected, never to a pass. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_tagging** — Bucket tags. The not-configured error code is NoSuchTagSet — singular, NOT NoSuchTagSetError — and means no tags, not a failure. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_s_3_bucket_ownership_controls** — Object ownership / ACL posture. OwnershipControlsNotFoundError has no documented Errors section, so an unrecognised 404 must degrade to not_collected, never to a pass. Never map a 403 AccessDenied this way — 'not configured' and 'cannot see' are opposite security conclusions. The collector must be region-sharded: a per-bucket GET sent to the wrong region returns 301, 307, 400 or AuthorizationHeaderMalformed.
- **list_all_amazon_web_services_rds_db_instances** — DB instances ONLY — clusters never appear here. DBClusterIdentifier is populated iff the instance belongs to a cluster: emit one row per cluster and one per standalone instance, never both. For Aurora the security posture lives on the CLUSTER (StorageEncrypted, KmsKeyId, BackupRetentionPeriod, DeletionProtection, EnabledCloudwatchLogsExports), so reading them off the instance reports the wrong object. Neptune and DocumentDB come back here too — filter by Engine. Shape asymmetries vs DBCluster: DBSubnetGroup and Endpoint are objects here but bare strings there, and the status field is DBInstanceStatus, not Status. TagList is inline, so ListTagsForResource is redundant.
- **list_all_amazon_web_services_rds_db_clusters** — DB clusters ONLY — Aurora and non-Aurora Multi-AZ. Instances never appear here. high_availability_mode cannot be a boolean: DBInstance.MultiAZ (one hidden standby), DBCluster.MultiAZ (members span AZs) and non-Aurora Multi-AZ clusters (two readable standbys) are three different failure domains. ReadReplicaDBClusterIdentifiers explicitly excludes cross-Region Aurora read replicas, so cross-region replicas are invisible from the source.
- **list_all_amazon_web_services_rds_global_clusters** — Aurora Global Databases only — the one API that returns cross-Region membership directly. Each GlobalClusterMember carries DBClusterArn, whose region segment is the residency answer, plus IsWriter and Readers, covering the writer and every secondary. It does not replace per-region enumeration: a cross-Region replica made with CreateDBCluster and ReplicationSourceIdentifier has no GlobalClusterIdentifier and never appears here, and neither do instance-level cross-Region replicas. Treat it as a cross-check on top of enumeration, not the source of truth.
- **list_all_amazon_web_services_rds_db_snapshots** — INSTANCE snapshots only — Aurora/cluster snapshots never appear here, so calling only this against an Aurora-heavy customer returns empty and would wrongly conclude they have no backups. The encryption field is Encrypted (contrast DBClusterSnapshot's StorageEncrypted). The RDS default is the mirror image of EC2's and is already correct: IncludeShared and IncludePublic both default false — leave them off, or you pull in other people's public snapshots. Automated snapshots cannot be shared at all, so skip the attribute call and set shared_with_accounts [] and is_public false by construction. Do not map location from SourceRegion — it means 'created in OR copied from'; parse the ARN instead.
- **list_all_amazon_web_services_rds_db_cluster_snapshots** — CLUSTER snapshots only. The encryption field is StorageEncrypted, NOT Encrypted — one shared mapper reads undefined here and reports every Aurora snapshot as unencrypted, a false-negative security finding at scale. DBClusterSnapshot also has no SourceRegion and no scalar AvailabilityZone. Neither RDS snapshot type carries an owner field: parse source_account_ref from the ARN.
- **list_all_amazon_web_services_rds_snapshot_attributes** — Sharing ACL for one manual instance snapshot. Skip entirely when SnapshotType is automated — automated snapshots cannot be shared, which removes over 90% of the per-item RDS calls. A manual snapshot can be shared with at most 20 accounts, and an encrypted snapshot cannot be made public, so encrypted && is_public is an impossible state worth asserting on.
- **list_all_amazon_web_services_rds_cluster_snapshot_attributes** — Sharing ACL for one manual cluster snapshot. Same automated-snapshot skip applies.
- **list_all_amazon_web_services_rds_db_parameters** — Instance parameter group values — the only path to min_tls_version, via a per-engine adapter. Three traps: this returns the GROUP's value, not the running instance's, so check ParameterApplyStatus == in-sync or a static parameter may claim TLS is enforced while the database still accepts plaintext; a parameter never set has NO ParameterValue element at all, so absent is not off and the engine default must be resolved; and enforcing TLS is not the same as setting a version floor. Use the parameter-name filter — it collapses 4-10 paginated calls per group to 1.
- **list_all_amazon_web_services_rds_db_cluster_parameters** — Cluster parameter group values. Aurora MySQL exposes tls_version as a comma-separated ALLOW-LIST to parse, not a floor, and has a second invisible audit path (server_audit_logs_upload) that never appears in EnabledCloudwatchLogsExports.
- **list_all_amazon_web_services_rds_subnet_groups** — DB subnet groups, for placing databases into the network graph. Note PubliclyAccessible true does not mean reachable and false does not mean safe — real exposure also needs route tables.
- **list_all_amazon_web_services_ec_2_regions** — Calls EC2 DescribeRegions to list AWS regions visible to the calling account. Not paginated — no cursor exists for this call. Filter to optInStatus in {opt-in-not-required, opted-in}; calling EC2 or other services in a not-opted-in region fails. Describes only the connection's own account (contrast with an Organizations-wide region enumeration). Never hardcode a region list — call this instead. All EC2 Describe* actions must be granted on Resource: "*"; this integration cannot be IAM-scoped to a subset of VPCs.
- **list_all_amazon_web_services_ec_2_vpcs** — Calls EC2 DescribeVpcs to list VPCs in the given region. Account-and-region-wide, cursor-paginated (NextToken, up to 1000 per page). Check blockPublicAccessStates: VPC Block Public Access can override an internet gateway route, so a subnet with an igw- route may still be unreachable. All EC2 Describe* actions must be granted on Resource: "*"; this integration cannot be IAM-scoped to a subset of VPCs.
- **list_all_amazon_web_services_ec_2_subnets** — Calls EC2 DescribeSubnets to list subnets in the given region. Cursor-paginated (NextToken, up to 1000 per page). Each subnet carries its vpcId and availabilityZone; join to ec2-vpcs for the parent VPC. All EC2 Describe* actions must be granted on Resource: "*"; this integration cannot be IAM-scoped to a subset of VPCs.
- **list_all_amazon_web_services_ec_2_flow_logs** — Calls EC2 DescribeFlowLogs to list VPC Flow Logs in the given region, one flat list keyed only by resourceId. Cursor-paginated (NextToken, up to 1000/page). Flow logs attach at VPC, SUBNET or ENI level, so checking resourceId == subnetId alone gives false negatives for the common "one flow log on the VPC" setup. Gate on flowLogStatus == ACTIVE AND deliverLogsStatus == SUCCESS — delivery can fail silently. Flow logs never capture DNS, IMDS, DHCP, ARP or Windows activation traffic. EC2 Describe* actions require Resource: "*"; cannot be IAM-scoped per VPC.
- **list_all_amazon_web_services_ec_2_route_tables** — Calls EC2 DescribeRouteTables to list route tables in the given region. Cursor-paginated (NextToken; MaxResults capped at 100 here, not 1000). A subnet not explicitly associated is IMPLICITLY tied to the VPC main table, and AWS omits the subnet id for implicit associations — joining only on association.subnetId marks every default-configured subnet as having no route; fall back to the main table (association.main == true). Distinguish igw- (internet), nat- (outbound only) and eigw- (IPv6 outbound only) gateway ids in routeSet. EC2 Describe* actions require Resource: "*"; cannot be IAM-scoped per VPC.
- **list_all_amazon_web_services_ec_2_network_acls** — Calls EC2 DescribeNetworkAcls to list network ACLs in the given region. Cursor-paginated (NextToken, up to 1000 per page). A genuinely different primitive from security groups: stateless, allow AND deny, ORDERED by a real ruleNumber, subnet-attached via associationSet. entrySet.protocol is a String holding a NUMBER ("-1", "6", "17"), never a protocol name, and NACL entries have no description field at all. All EC2 Describe* actions must be granted on Resource: "*"; this integration cannot be IAM-scoped to a subset of VPCs.
- **list_all_amazon_web_services_ec_2_network_interfaces** — Calls EC2 DescribeNetworkInterfaces to list elastic network interfaces (ENIs) in the given region. Cursor-paginated (NextToken, up to 1000 per page). This is the ONLY path from security groups to subnets — SecurityGroup carries no list of attached interfaces and the relationship does not exist in AWS's model. Join groups.item.groupId -> subnetId, and be honest that this is "groups observed in use right now", not configuration: an autoscaling event changes it. All EC2 Describe* actions must be granted on Resource: "*"; this integration cannot be IAM-scoped to a subset of VPCs.
- **list_all_amazon_web_services_ec_2_security_groups** — Calls EC2 DescribeSecurityGroups to list security groups in the given region. Cursor-paginated (NextToken, up to 1000/page). Stateful, ALLOW-ONLY, unordered, ENI-attached. Rules nested in ipPermissions/ipPermissionsEgress have no action, priority, enabled or logging_enabled fields — those don't exist in this model (contrast ec2-network-acls). ipProtocol "-1" means all protocols AND all ports — never default to 0-65535; for icmp, fromPort/toPort are TYPE and CODE, not ports. EC2 Describe* actions require Resource: "*"; cannot be IAM-scoped per VPC.
- **list_all_amazon_web_services_ec_2_security_group_rules** — Calls EC2 DescribeSecurityGroupRules to list individual security-group rules in the given region — the only source of a real per-rule description and a stable securityGroupRuleId (ec2-security-groups only nests rules inline, unaddressable). Cursor-paginated (NextToken, up to 1000/page). Prefix-list rules (no referencedGroupInfo, no cidrIpv4/cidrIpv6) carry no CIDRs directly — resolve via ec2-prefix-lists / GetManagedPrefixListEntries using prefixListId. EC2 Describe* actions require Resource: "*"; cannot be IAM-scoped per VPC.
- **list_all_amazon_web_services_ec_2_prefix_lists** — Calls EC2 DescribeManagedPrefixLists to list managed prefix lists (both AWS-owned, e.g. com.amazonaws.<region>.s3, and customer-managed) referenced by security-group and route-table rules in the given region. Cursor-paginated (NextToken, up to 1000 per page). Resolve the actual CIDR entries of a prefix list with GetManagedPrefixListEntries using prefixListId — this call returns only the list's metadata, not its members. All EC2 Describe* actions must be granted on Resource: "*"; this integration cannot be IAM-scoped to a subset of VPCs.
- **list_all_amazon_web_services_ec_2_prefix_list_entries** — CIDR (or ARN) entries behind one managed prefix list, via GetManagedPrefixListEntries. Requires prefix_list_id — get one from ec2-prefix-lists. Store the list's version alongside entries; a later diff can't distinguish our change from Amazon's without it. Paginated via NextToken/MaxResults. All EC2 Describe*/Get* actions must be granted on Resource: "*" — AWS does not support resource-level permissions here.
- **list_all_amazon_web_services_ec_2_vpc_peering** — VPC peering connections via DescribeVpcPeeringConnections. Transitive peering does not exist in AWS — an active peering only makes routing possible, it does not create it, so cross-check ec2-route-tables for routes targeting the peering id. Dedup on vpcPeeringConnectionId: an inter-region peering appears identically from both regions/accounts. Paginated via NextToken/MaxResults. All EC2 Describe* actions require Resource: "*" — no resource-level IAM scoping.
- **list_all_amazon_web_services_ec_2_transit_gateways** — Transit gateways via DescribeTransitGateways. The single highest-signal TGW finding lives here: options.defaultRouteTableAssociation/defaultRouteTablePropagation = "enable" is the flat-mesh default where every attached VPC can reach every other VPC. Regional resource — call once per region. Paginated via NextToken/MaxResults. All EC2 Describe* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_tgw_attachments** — Attachments to transit gateways (VPCs, VPNs, Direct Connect gateways, peering) via DescribeTransitGatewayAttachments. association.transitGatewayRouteTableId is the ONLY route-table link on this object — propagation is not present here, so two look-alike attachments can have very different blast radii. Paginated via NextToken/MaxResults. All EC2 Describe* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_tgw_route_tables** — Transit gateway route tables via DescribeTransitGatewayRouteTables — the anchor id for ec2-tgw-associations and ec2-tgw-propagations, which together determine actual reachability. Paginated via NextToken/MaxResults. All EC2 Describe* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_tgw_associations** — Attachments ASSOCIATED with one TGW route table, via GetTransitGatewayRouteTableAssociations. Half the reachability rule: A reaches B iff B's attachment is in propagations[associations-inverse(A)]. Requires transit_gateway_route_table_id — get one from ec2-tgw-route-tables. Paginated via NextToken/MaxResults. All EC2 Describe*/Get* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_tgw_propagations** — Attachments that PROPAGATE routes into one TGW route table, via GetTransitGatewayRouteTablePropagations. The other half of the reachability rule — without it, transit is unknowable from the attachment alone. Requires transit_gateway_route_table_id — get one from ec2-tgw-route-tables. Paginated via NextToken/MaxResults. All EC2 Describe*/Get* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_snapshots** — EBS snapshots via DescribeSnapshots, scoped Owner.1=self — a correctness requirement, not an optimization: without it AWS returns every public snapshot in existence, paginating effectively forever. The status field (not "state") and tagSet (not "Tags") use non-standard names. No sharing field here — use ec2-snapshot-attribute for that — and no region field; the request region is the snapshot's region. AMIs are excluded (their snapshots already appear here). Paginated via NextToken/MaxResults. All EC2 Describe* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_snapshot_attribute** — The sharing ACL for ONE EBS snapshot, via DescribeSnapshotAttribute (Attribute=createVolumePermission) — the only source of shared_with_accounts/is_public, and the most expensive call in this group: no batch form, one call per snapshot per region. The snapshot id is a PATH argument (call as ec2-snapshot-attribute <snapshot-id>), not a query param — get ids from ec2-snapshots. group == "all" (its only legal value) means public; userId entries are the shared account ids. All EC2 Describe* actions require Resource: "*".
- **list_all_amazon_web_services_ec_2_snapshot_bpa_state** — Account+region setting for EBS snapshot Block Public Access, via GetSnapshotBlockPublicAccessState — one call per region, not global. Unlike other "list" methods here, this returns a SINGLE state value, not an array of items. This changes how is_public must be read: under block-all-sharing, already-public snapshots are treated as private and aren't actually exposed, so reporting one as public is a false positive. Report both the per-snapshot ACL (ec2-snapshot-attribute) and this regional posture together. All EC2 Describe*/Get* actions require Resource: "*".
- **list_all_amazon_web_services_kms_keys** — Key ids only — {KeyArn, KeyId} and NO metadata at all, so budget 4 calls per key. Pagination is a third dialect: Limit / Marker / NextMarker / Truncated, neither EC2's nor the JSON services' NextToken. This also returns AWS-managed aws/* service keys with no server-side filter, and in a mature account they vastly outnumber customer keys — the DescribeKey cost is paid on every one before KeyManager == CUSTOMER can discard it.
- **get_single_amazon_web_services_kms_key_by_id** — Key metadata. Origin has FOUR values, not three: AWS_KMS, EXTERNAL (BYOK), AWS_CLOUDHSM and EXTERNAL_KEY_STORE (XKS). Do not read the deprecated CustomerMasterKeySpec — its documented value list is stale. Carry MultiRegion / MultiRegionConfiguration or the same logical key is reported as N unrelated keys. Call this FIRST and skip GetKeyRotationStatus for anything that is not a symmetric ENCRYPT_DECRYPT key. Works cross-account.
- **get_single_amazon_web_services_kms_key_rotation_by_id** — Rotation state. All three fields are real — KeyRotationEnabled, RotationPeriodInDays (range 90-2560, default 365) and NextRotationDate — but they are NULL, not false, for asymmetric, HMAC and custom-key-store keys, where rotation is not applicable and false would read as a misconfiguration. AWS-managed keys always rotate and it cannot be changed. Rotation is a shared property of multi-region keys, configured only on the primary. Works cross-account.
- **get_single_amazon_web_services_kms_key_policy_by_id** — The key policy, as a raw JSON string. Skip ListKeyPolicies entirely — the only valid policy name is 'default'. This does NOT work cross-account, so policy_grants_anyone is unobtainable unless we assume a role into that account. No API answers 'is this key public': naive Principal "*" matching produces FALSE POSITIVES on every normal key, because the standard idiom pairs Principal "*" with a kms:CallerAccount condition. Check for the ABSENCE of a scoping condition (kms:CallerAccount, aws:PrincipalOrgID, aws:PrincipalAccount, kms:ViaService). Grants are a second independent authorization path this does not show. Most KMS reads are also governed by the per-key policy, not just our IAM policy, so partial coverage must surface as permission_denied per key with a count, never as a failed run.
- **list_all_amazon_web_services_kms_aliases** — Key aliases. Call this ONCE PER REGION, never per key — that is one of the three mitigations that keeps a 500-key, 20-region scan from costing 10,000 requests.
- **list_all_amazon_web_services_cloudtrail_trails** — Lists CloudTrail trails visible in this account/region via AWS CloudTrail DescribeTrails. No required params beyond region/account_id. NOT paginated. `enabled` is NOT in this response — a trail can exist, be multi-region, KMS-encrypted and validated while being completely stopped, and this looks identical either way. includeShadowTrails defaults to true, so a naive multi-region sweep returns one multi-region trail once per region: dedup on TrailARN and treat the record where HomeRegion equals the region called as authoritative. Do NOT set includeShadowTrails false to avoid that — in a member account of an organization trail it returns nothing at all and we would wrongly report the account as unaudited. An absent KmsKeyId means SSE-S3, NOT unencrypted.
- **get_single_amazon_web_services_cloudtrail_trail_status_by_id** — Calls AWS CloudTrail GetTrailStatus for one trail; pass the trail name or ARN as the id (from cloudtrail-trails), plus region/account_id. The ONLY source of whether a trail is actually logging: IsLogging. One extra call per trail per region, not optional. NOT paginated. IsLogging true is still not proof logs arrive — cross-check LatestDeliveryError and LatestDeliveryTime, since a trail writing to a deleted bucket reports true with a stale timestamp.
- **get_single_amazon_web_services_cloudtrail_event_selector_by_id** — Calls AWS CloudTrail GetEventSelectors for one trail; pass the trail name or ARN as the id (from cloudtrail-trails), plus region/account_id. Data-event coverage for one trail. Classic and advanced event selectors are MUTUALLY EXCLUSIVE, so two parsers are needed. Classic is clean: three resource types (AWS::S3::Object, AWS::Lambda::Function, AWS::DynamoDB::Table) with explicit ARN lists. Advanced is a boolean predicate language (Equals/NotEquals/StartsWith/NotStartsWith/EndsWith/NotEndsWith) that no API evaluates for us, where negation can carve arbitrary holes out of a broad match and far more than three resource types are supported — an allowlist of three silently under-reports. Best-effort inference; say so. ExcludeManagementEventSources can carve out kms.amazonaws.com or rdsdata.amazonaws.com, so management coverage is not complete even when IncludeManagementEvents is true.
- **list_all_amazon_web_services_logs_log_groups** — Calls AWS CloudWatch Logs DescribeLogGroups. No required params beyond region/account_id. limit is 1-50 (small — expect heavy paging) and the connector's pagination cursor for this method EXPIRES AFTER 24 HOURS, so cursors must not be persisted. Do NOT set log_group_name_pattern: when it is set the response contains only arn, creationTime and logGroupName, silently dropping retentionInDays and kmsKeyId, the two fields we need. An absent retentionInDays means NEVER EXPIRE, not zero. An absent kmsKeyId means AWS-owned SSE, not unencrypted. deletionProtectionEnabled protects the group from deletion, not the retention period — anyone with logs:PutRetentionPolicy can still shorten retention to 1 day. logGroupClass DELIVERY keeps events one day only regardless of retentionInDays.
- **list_all_amazon_web_services_ec_2_instances** — EC2 instances, flattened to one row per instance -- DescribeInstances nests instances under reservations, but response_path unwraps reservationSet.item.instancesSet.item so each item here already IS an instance. Region-scoped XML Query API (call once per region); paginates via NextToken, up to 1000/page. XML->JSON: every scalar leaf comes back as {_text: value}, and a nested list (tagSet, groupSet, blockDeviceMapping...) collapses to a bare object instead of a one-item array when there's exactly one entry -- normalize before iterating. Fixture account returned zero instances in every region tested; schema below is derived from the AWS API reference, not a live sample.
- **list_all_amazon_web_services_ec_2_volumes** — EBS volumes. DescribeVolumes XML Query API, region-scoped; paginates via NextToken, up to 500/page. The state field is called status (creating|available|in-use|deleting|deleted|error), not state; tags are under tagSet.item, not Tags. Every scalar leaf is wrapped as {_text: value} by XML->JSON conversion, and attachmentSet.item collapses to a bare object (not a one-element array) when the volume has exactly one attachment -- normalize before iterating. Fixture account returned zero volumes in every region tested; schema below is derived from the AWS API reference, not a live sample.
- **list_all_amazon_web_services_elbv_2_load_balancers** — ALB/NLB/Gateway load balancers. DescribeLoadBalancers -- a different AWS service (elasticloadbalancing) and a different XML array convention than the ec2-* resources in this set: repeated elements use 'member', not 'item'. Region-scoped; paginates via Marker/PageSize, up to 400/page. state is a nested {code, reason} object, not a flat status string. Every scalar leaf is still {_text: value} from the XML->JSON conversion. Fixture account returned zero load balancers in every region tested; schema below is derived from the AWS API reference, not a live sample.
- **list_all_amazon_web_services_redshift_clusters** — Cluster metadata via DescribeClusters — the AWS Query/XML protocol, NOT JSON. Verified live via source (xml-js compact mode): every leaf field arrives wrapped as {_text: value} after XML-to-JSON conversion, e.g. ClusterIdentifier._text, unlike our JSON-service AWS resources. Optional ClusterIdentifier narrows to one cluster; a nonexistent id 404s with ClusterNotFound (verified live). Cursor pagination via Marker/MaxRecords, max 100. No credentials returned — MasterUserPassword is never in any Redshift read response.

- **list_all_amazon_web_services_cloudwatch_alarms** — Lists CloudWatch metric alarms in this region via AWS CloudWatch DescribeAlarms (the XML API). No required params beyond region/account_id. The integration config does not wire AWS's AlarmNames, StateValue or ActionPrefix filters, so this always returns every metric alarm in the region, paged internally via NextToken/MaxRecords (up to 100 per page) through the connector's own cursor. Composite alarms (AlarmRule instead of MetricName) are returned mixed in with metric alarms.
- **list_all_amazon_web_services_ecs_clusters** — ECS cluster ARNs only -- ListClusters returns an array of bare ARN strings, never full cluster objects (no status, running/pending task counts, capacity providers, etc.). Verified live: this fixture account has zero ECS clusters in every region tested, and the call correctly returns an empty array rather than erroring. There is no get/describe method configured for ecs-clusters in this integration, so an ARN returned here cannot be resolved to full details through this proxy. AWS JSON RPC over POST (x-amz-target header, body always {}); region-scoped, paginates via nextToken, up to 100/page.
- **list_all_amazon_web_services_ecs_services** — ECS service ARNs only -- ListServices returns bare ARN strings, never full objects (no desired/running count, task definition, deployment config, etc.). cluster is optional per the AWS API (falls back to a cluster named 'default'), but no such cluster exists in this fixture account -- confirmed live, omitting it or passing cluster=default both fail with ClusterNotFoundException. Pass a real cluster ARN/name from ecs-clusters. No get/describe method exists for ecs-services. AWS JSON RPC over POST; region-scoped, paginates via nextToken, up to 100/page.
- **list_all_amazon_web_services_ecr_repositories** — ECR repositories as full objects (ARN, URI, tag mutability, encryption, scan-on-push filters) -- DescribeRepositories, unlike the ARN/name-only list APIs elsewhere in this set (ecs-clusters, ecs-services, eks-clusters, eks-nodegroups). This config's body is always {} with no repositoryNames/registryId filter exposed, so every call returns the whole registry's repositories, unfiltered. AWS JSON RPC over POST; region-scoped, paginates via nextToken, up to 1000/page. Fixture account has zero repositories in every region tested; schema below is derived from the AWS API reference, not a live sample.
- **list_all_amazon_web_services_secretsmanager_secrets** — ListSecrets — metadata only. SecretString/SecretBinary are NEVER returned here; only GetSecretValue exposes them, and that is not one of our methods, so never describe this as a possible value leak. SortOrder (asc|desc) and IncludePlannedDeletion (bool) pass through live and are AWS-validated (confirmed: bad values 400 with ValidationException/SerializationException). Filters (AWS's {Key,Values} list) is supported by the API but not exposed as a flat query param here. Cursor pagination via NextToken/MaxResults, max 100.

- **list_all_amazon_web_services_dynamodb_tables** — ListTables. The response is a BARE ARRAY OF TABLE NAME STRINGS (TableNames) — not objects, so there is no ARN, status, billing mode, or item count here. Get those per-table via DescribeTable (a separate call, out of scope for this method). AWS ListTables takes no filter params, only pagination (ExclusiveStartTableName/LastEvaluatedTableName), max page 100.
- **list_all_amazon_web_services_lambda_functions** — Lambda functions as full configuration objects (ARN, runtime, handler, role, VPC config, layers, environment, logging config, etc.) -- ListFunctions, verified live with real functions. Region-scoped; paginates via Marker/MaxItems, up to 50/page. This fixture account's functions only exist in us-west-2 -- the default us-east-1 and every other region tested returned an empty array, so always pass region explicitly. Many fields are null unless the function uses that feature (VpcConfig, Environment, DeadLetterConfig, KMSKeyArn, ...) -- null means "not configured", not missing data.
- **list_all_amazon_web_services_eks_clusters** — EKS cluster NAMES only -- ListClusters returns an array of bare strings, never full cluster objects (no ARN, version, status, endpoint, VPC config, etc.), confirmed against the AWS API reference. This account has zero EKS clusters in every region tested, so the empty-array shape was verified live but a populated response wasn't. There is no get/describe method configured for eks-clusters in this integration, so a name returned here cannot be resolved to full details through this proxy. Region-scoped; paginates via nextToken, up to 100/page.
- **list_all_amazon_web_services_eks_nodegroups** — EKS node group NAMES only -- ListNodegroups returns an array of bare strings, never full objects (no ARN, instance types, scaling config, etc.), confirmed against the AWS API reference. cluster_name is required (path param /clusters/{cluster_name}/node-groups) and must come from eks-clusters' list; this fixture account has zero EKS clusters in any region, so no real cluster_name could be sourced and the response couldn't be verified live. No get/describe method exists for eks-nodegroups. Region-scoped; paginates via nextToken, up to 100/page.
- **list_all_amazon_web_services_efs_file_systems** — EFS file systems. DescribeFileSystems is a plain JSON REST call -- no XML {_text} wrapping, unlike the ec2-*/elbv2-* resources in this set. Region-scoped; paginates via Marker/MaxItems, up to 100/page (the config tolerates an empty-string cursor on the last page). SizeInBytes is an object with Value/Timestamp fields, not a bare number. Fixture account returned zero file systems in every region tested; schema below is derived from the AWS API reference, not a live sample.
- **list_all_amazon_web_services_iam_groups** — IAM groups in the account (ListGroups). IAM is a global service reached through a single endpoint, so region only affects which credentials/account are used, not where the data lives. Paginated with the classic Marker/IsTruncated cursor, same as iam-users and iam-roles. Does not include a group's attached/inline policies or its members — those need separate IAM calls.
- **list_all_amazon_web_services_iam_attached_user_policies** — AWS-managed and customer-managed IAM policies attached directly to one IAM user (ListAttachedUserPolicies). user_name comes from iam-users (UserName). Does not include inline (embedded) user policies (see iam-users' GetUserPolicy) or policies inherited from group membership.
- **get_single_amazon_web_services_iam_policy_version_by_id** — The full JSON policy document for one version of a customer- or AWS-managed IAM policy (GetPolicyVersion). Pass the policy ARN as the resource id (from iam-policies, Arn) and the version id as version_id (from iam-policies, DefaultVersionId, or IsDefaultVersion=true on this response). Document is returned percent-URL-encoded JSON — URL-decode it before parsing the statements. Callers must percent-encode the ARN before passing it as a path id (it contains ':' and '/').
- **list_all_amazon_web_services_iam_entities_for_policy** — Reverse lookup for an IAM managed policy: which groups, roles and users it is attached to (ListEntitiesForPolicy). policy_arn comes from iam-policies (Arn). Unlike other iam-* list methods this returns one object with three separate member arrays (PolicyGroups/PolicyRoles/PolicyUsers), not a flat list of entities.
- **get_single_amazon_web_services_iam_access_key_last_used_by_id** — When and where an IAM access key was last used (GetAccessKeyLastUsed) — the key signal for finding stale or unused long-term credentials. Pass the access key id as the resource id (from iam-access-keys, AccessKeyId). ServiceName and Region are absent from the response until the key has been used at least once.
- **list_all_amazon_web_services_config_rules** — AWS Config rule definitions in the region — name, source (AWS managed check or custom Lambda), scope and evaluation mode (DescribeConfigRules). This returns rule definitions only; use config-rule-compliance for each rule's pass/fail status.
- **list_all_amazon_web_services_config_rule_compliance** — Pass/fail compliance status for AWS Config rules in the region (DescribeComplianceByConfigRule) — COMPLIANT, NON_COMPLIANT or INSUFFICIENT_DATA per rule, with resource counts. Cross-reference ConfigRuleName against config-rules for the rule definition. Not paginated by AWS — returns every rule for the region in one call.
- **list_all_amazon_web_services_accessanalyzer_findings** — Findings from an IAM Access Analyzer analyzer — resources shared outside the account/org, or unused permissions, depending on the analyzer's type (ListFindings). analyzer_arn comes from accessanalyzer-analyzers (arn). No analyzer was configured in this account at verification time; verified live against a well-formed but nonexistent analyzer ARN, which AWS correctly rejected with 'Analyzer not found', confirming the request reaches the service correctly. Response fields below are from the AWS API reference.
- **list_all_amazon_web_services_cloudtrail_list_trails** — Lightweight inventory of every CloudTrail trail visible to this account/region — name, home region and ARN (ListTrails). Unlike cloudtrail-trails (DescribeTrails) this does not return the trail's logging configuration (S3 bucket, KMS key, multi-region flag); use cloudtrail-trails or cloudtrail-trail-status for that. No trail existed in this account at verification time (call returned 200 with an empty list).
- **get_single_amazon_web_services_guardduty_org_configuration_by_id** — Whether GuardDuty is auto-enabled for new accounts joining the organization, and this account's member/data-source status (DescribeOrganizationConfiguration). Pass a detector id (from guardduty-detectors) as the resource id. Only meaningful when called from the GuardDuty delegated-administrator account. No detector existed in this account at verification time; the call correctly reached AWS and was rejected with 'detectorId is not owned by the current account', confirming the path and signing are correct. Response fields below are from the AWS API reference.
- **list_all_amazon_web_services_securityhub_standards** — Catalog of every Security Hub compliance standard AWS offers in this region — CIS, PCI DSS, AWS Foundational Security Best Practices, and others — not just the ones enabled for this account (DescribeStandards). Use securityhub-enabled-standards to see which of these are actually subscribed; StandardsArn from this call is the id securityhub-enabled-standards' subscriptions reference.
- **get_single_amazon_web_services_organizations_account_by_id** — Details of one AWS account inside the organization — email, status, and joined method/date (DescribeAccount). Pass the 12-digit account id as the resource id (from organizations-accounts, Id). Only callable from the organization's management or delegated-administrator account; this connection's default account returned AWS's AWSOrganizationsNotInUseException at verification time — the same error every other organizations-* resource on this integration gets from this account, so this is an environment limitation rather than a config defect.
- **get_single_amazon_web_services_s_3_bucket_by_id** — One S3 bucket's ListBuckets entry (Name, BucketRegion, CreationDate, BucketArn), read with ListBuckets prefix=<id>&max-buckets=1 (s3:ListAllMyBuckets). 404 aws_bucket_not_found when the bucket is not in this account.
- **list_all_amazon_web_services_account_information** — AWS Account Management GetAccountInformation: AccountId, AccountName, AccountCreatedDate, AccountState for the calling account, or for an org member with ?truto_target_account_id= (management / delegated admin only). Needs account:GetAccountInformation.
- **list_all_amazon_web_services_iam_account_aliases** — IAM ListAccountAliases: the calling account's alias (an account has at most one). Needs iam:ListAccountAliases.
- **list_all_amazon_web_services_s_3_account_public_access_block** — S3 Control GetPublicAccessBlock: the account-wide S3 Block Public Access flags (BlockPublicAcls, IgnorePublicAcls, BlockPublicPolicy, RestrictPublicBuckets) for the connection's account. 404 NoSuchPublicAccessBlockConfiguration when none is set. Needs s3:GetAccountPublicAccessBlock.
- **get_s_3_account_public_access_block_by_id** — S3 Control GetPublicAccessBlock: the account-wide S3 Block Public Access flags (BlockPublicAcls, IgnorePublicAcls, BlockPublicPolicy, RestrictPublicBuckets) for the connection's account. 404 NoSuchPublicAccessBlockConfiguration when none is set. Needs s3:GetAccountPublicAccessBlock. GET flavour (any id, e.g. 'account'): 404 none-set is returned as 200 {"truto_not_configured": true}.
- **list_all_amazon_web_services_config_select_resource_config** — AWS Config SelectResourceConfig: runs an AWS Config advanced query (query param `expression`, SQL SELECT ... WHERE ...) against the resources Config records in this region, and returns one parsed object per match. Max 100 per page (Limit); NextToken pages. Needs config:SelectResourceConfig, and returns nothing for resource types Config does not record in the region.
- **get_single_amazon_web_services_kms_key_policy_document_by_id** — KMS GetKeyPolicy, whole response ({Policy, PolicyName}). Same call as kms-key-policy, which returns only the Policy string; this flavour returns an object so a related read can stamp the key id onto it. Needs kms:GetKeyPolicy, and the key policy must allow the caller.
- **get_single_amazon_web_services_kms_resource_tag_by_id** — KMS ListResourceTags for one key: {Tags: [{TagKey, TagValue}], Truncated}. A key holds at most 50 tags, so one page is the whole set. Needs kms:ListResourceTags.
- **list_all_amazon_web_services_dynamodb_table_list** — DynamoDB ListTables, one object per table ({TableName}) instead of the bare names dynamodb-tables returns. Needs dynamodb:ListTables.
- **get_single_amazon_web_services_dynamodb_table_by_id** — DynamoDB DescribeTable: SSEDescription (encryption), Replicas (global table regions), TableArn, status. Needs dynamodb:DescribeTable.
- **get_single_amazon_web_services_dynamodb_continuous_backup_by_id** — DynamoDB DescribeContinuousBackups: ContinuousBackupsStatus and PointInTimeRecoveryDescription (PITR status, recovery period). Needs dynamodb:DescribeContinuousBackups.
- **get_single_amazon_web_services_efs_backup_policy_by_id** — EFS DescribeBackupPolicy: BackupPolicy.Status (ENABLED, ENABLING, DISABLED, DISABLING). No policy (404 PolicyNotFound) returns {truto_not_configured: true}. Needs elasticfilesystem:DescribeBackupPolicy.
- **list_all_amazon_web_services_sqs_queue_list** — SQS ListQueues over the AWS JSON protocol, one object per queue ({QueueUrl}). Up to 1000 per page (MaxResults); NextToken pages, and AWS returns one only when a limit is given. Needs sqs:ListQueues.
- **get_single_amazon_web_services_sqs_queue_attribute_by_id** — SQS GetQueueAttributes (All) over the AWS JSON protocol for one queue URL (the id): a {Name: Value} map of strings with QueueArn, SqsManagedSseEnabled, KmsMasterKeyId, Policy, CreatedTimestamp and the rest. Needs sqs:GetQueueAttributes.
- **get_single_amazon_web_services_ec_2_instance_by_id** — EC2 DescribeInstances for one instance id (the id, i-...): the same instance item ec2-instances lists. 404 aws_no_such_entity when the instance does not exist in this account and region. Needs ec2:DescribeInstances.
- **get_single_amazon_web_services_lambda_function_by_id** — Lambda GetFunctionConfiguration for one function (the id: a function name): the $LATEST configuration, the same shape ListFunctions returns, plus State. Includes environment variable values, like lambda-functions. 404 aws_no_such_entity when the function does not exist. Needs lambda:GetFunctionConfiguration.
- **get_single_amazon_web_services_lambda_function_tag_by_id** — Lambda ListTags for one function (the id: the unqualified function ARN): { "Tags": { key: value } }. Needs lambda:ListTags.
- **get_single_amazon_web_services_cloudwatch_alarm_by_id** — CloudWatch DescribeAlarms for one alarm name (the id), metric or composite. 404 aws_no_such_entity when no alarm has that name. Needs cloudwatch:DescribeAlarms (scoped to * for composite alarms).
- **list_all_amazon_web_services_cloudwatch_composite_alarms** — CloudWatch DescribeAlarms, composite alarms only (cloudwatch-alarms returns metric alarms only, as DescribeAlarms does by default). Max 100 per page. Needs cloudwatch:DescribeAlarms scoped to *.
- **list_all_amazon_web_services_autoscaling_groups** — EC2 Auto Scaling DescribeAutoScalingGroups: every group in the region with its members, launch template or configuration, health checks and tags. Max 100 per page. Needs autoscaling:DescribeAutoScalingGroups.
- **get_single_amazon_web_services_autoscaling_group_by_id** — EC2 Auto Scaling DescribeAutoScalingGroups for one group name (the id). 404 aws_no_such_entity when no group has that name. Needs autoscaling:DescribeAutoScalingGroups.
- **list_all_amazon_web_services_sns_topics** — SNS ListTopics: topic ARNs in the region, up to 100 per page (NextToken). SNS takes no page size, so pass truto_ignore_limit=true instead of limit. Needs sns:ListTopics.
- **get_single_amazon_web_services_sns_topic_attribute_by_id** — SNS GetTopicAttributes for one topic ARN (the id): { entry: [ { key, value } ] } with SubscriptionsConfirmed, SubscriptionsPending, KmsMasterKeyId, Policy, DisplayName, Owner and the rest. 404 aws_no_such_entity when the topic does not exist; AuthorizationError is 403 aws_access_denied. Needs sns:GetTopicAttributes.
- **list_all_amazon_web_services_sns_topic_subscriptions** — SNS ListSubscriptionsByTopic for one topic (query param topic_arn, required): up to 100 subscriptions per page (NextToken). A subscription still waiting for confirmation has SubscriptionArn "PendingConfirmation". Endpoint holds the subscriber address (email, phone, URL). SNS takes no page size, so pass truto_ignore_limit=true instead of limit. Needs sns:ListSubscriptionsByTopic.
- **list_all_amazon_web_services_ec_2_images** — EC2 DescribeImages Owner=self: the account's own AMIs in the region, including deprecated (always shown to the owner) and disabled ones. MaxResults 5-1000. Needs ec2:DescribeImages.
- **get_single_amazon_web_services_ec_2_image_by_id** — EC2 DescribeImages for one AMI id (the id, ami-...): any AMI this account can see, its own, shared, Amazon or Marketplace. 404 aws_no_such_entity when it is not visible. Needs ec2:DescribeImages.
- **list_all_amazon_web_services_ssm_instance_information** — Systems Manager DescribeInstanceInformation: managed nodes in the region (EC2 i- and hybrid mi-), with ping status, agent version and platform. Stopped and terminated nodes are not returned. MaxResults 5-50. Pass Filters as an array in the query to narrow it (on the proxy URL: Filters[0][Key]=InstanceIds&Filters[0][Values][0]=i-...; SSM rejects a JSON string with SerializationException); a node id that is not managed answers 200 with an empty list and truto_not_managed. Needs ssm:DescribeInstanceInformation.
- **list_all_amazon_web_services_account_information_next** — AWS Account Management GetAccountInformation: AccountId, AccountName, AccountCreatedDate, AccountState for the calling account, or for an org member with ?truto_target_account_id= (management / delegated admin only). Needs account:GetAccountInformation.
- **list_all_amazon_web_services_ec_2_enabled_regions** — EC2 DescribeRegions (default filter: the regions enabled for the calling account). Used by GET accounts/{id} for enabled_regions. Unlike ec2-regions, an access denial answers 200 with one item { truto_read_error: { reason: "permission_denied", provider_error, provider_status } } instead of 403, so the account read can report it. Needs ec2:DescribeRegions.
- **list_all_amazon_web_services_ecs_cluster** — ECS DescribeClusters, batched: pass up to 100 cluster ARNs or names as an array, clusters[0]=...&clusters[1]=... (one call). Returns the described clusters with settings (Container Insights), configuration (ECS Exec logging and KMS key) and tags; ARNs ECS could not find are left out (truto_response_format=raw keeps failures[]). Needs ecs:DescribeClusters.
- **get_single_amazon_web_services_ecs_cluster_by_id** — ECS DescribeClusters for one cluster (id = cluster ARN or name). 404 aws_no_such_entity when ECS reports it MISSING. Needs ecs:DescribeClusters.
- **list_all_amazon_web_services_ecs_service** — ECS DescribeServices, batched: cluster (ARN or name) and up to 10 service ARNs or names as an array, services[0]=... (one call). Returns the described services with tags. Needs ecs:DescribeServices.
- **get_single_amazon_web_services_ecs_service_by_id** — ECS DescribeServices for one service: id = service ARN or name, cluster = cluster ARN or name (required unless the service is in the default cluster). 404 aws_no_such_entity. Needs ecs:DescribeServices.
- **list_all_amazon_web_services_ecs_tasks** — ECS ListTasks: task ARNs in a cluster. Filters: cluster, serviceName, family, desiredStatus (RUNNING default, or STOPPED), launchType, containerInstance, startedBy. Needs ecs:ListTasks.
- **list_all_amazon_web_services_ecs_task** — ECS DescribeTasks, batched: cluster and up to 100 task ARNs or ids as an array, tasks[0]=... (one call). Environment variable values and command lines in overrides are replaced with "[redacted]"; names are kept. Needs ecs:DescribeTasks.
- **get_single_amazon_web_services_ecs_task_by_id** — ECS DescribeTasks for one task: id = task ARN or id, cluster = cluster ARN or name. Override values redacted as in ecs-task list. Needs ecs:DescribeTasks.
- **list_all_amazon_web_services_ecs_task_definition_families** — ECS ListTaskDefinitionFamilies: family names. status=ACTIVE lists only families with an ACTIVE revision; familyPrefix filters by name. Needs ecs:ListTaskDefinitionFamilies.
- **get_single_amazon_web_services_ecs_task_definition_by_id** — ECS DescribeTaskDefinition: id = family (latest ACTIVE revision), family:revision, or task definition ARN. Returns { taskDefinition, tags }. Environment variable values, command/entryPoint/healthCheck commands, dockerLabels values, non-awslogs log options, FireLens options, Docker volume driverOpts/labels and proxy properties are replaced with "[redacted]"; names, keys and secret references (valueFrom ARNs) are kept. Needs ecs:DescribeTaskDefinition.
- **get_single_amazon_web_services_eks_cluster_by_id** — EKS DescribeCluster (id = cluster name): Kubernetes version, platform version, endpoint public/private access and CIDRs, secrets encryption, control-plane logging, authentication mode, upgrade support type, VPC config, tags. connectorConfig.activationCode is redacted. Needs eks:DescribeCluster.
- **get_single_amazon_web_services_eks_nodegroup_by_id** — EKS DescribeNodegroup: id = node group name, cluster_name = cluster name. Kubernetes version, AMI release version, AMI type, capacity type, scaling, remote (SSH) access, node role, health, tags. Needs eks:DescribeNodegroup.
- **list_all_amazon_web_services_elbv_2_listeners** — ELBv2 DescribeListeners for one load balancer (LoadBalancerArn) or for ListenerArns.member.N: port, protocol, TLS policy (SslPolicy), default certificate, default actions, mutual TLS. Needs elasticloadbalancing:DescribeListeners.
- **list_all_amazon_web_services_elbv_2_tags** — ELBv2 DescribeTags for up to 20 resources per call: pass ResourceArns.member.1 ... ResourceArns.member.20. Needs elasticloadbalancing:DescribeTags.
- **list_all_amazon_web_services_elbv_2_ssl_policies** — ELBv2 DescribeSSLPolicies: every predefined TLS security policy with its protocols (SslProtocols) and ciphers, or only Names.member.N. LoadBalancerType narrows to application, network or gateway. Needs elasticloadbalancing:DescribeSSLPolicies.
- **list_all_amazon_web_services_cloudfront_distributions** — CloudFront ListDistributions (global; signed for us-east-1): distribution summaries with aliases, origins, viewer protocol policy per behavior, viewer certificate and minimum TLS policy, WAF web ACL, geo restriction. Logging and DefaultRootObject are only in GetDistribution. Needs cloudfront:ListDistributions.
- **get_single_amazon_web_services_cloudfront_distribution_by_id** — CloudFront GetDistribution (id = distribution id, e.g. E1U5RQF7T870K0): the full DistributionConfig, including Logging and DefaultRootObject. 404 aws_no_such_entity. Needs cloudfront:GetDistribution.
- **list_all_amazon_web_services_cloudfront_tags** — CloudFront ListTagsForResource: pass Resource=<distribution ARN>. Needs cloudfront:ListTagsForResource.
- **list_all_amazon_web_services_acm_certificates** — ACM ListCertificates (regional; certificates used by CloudFront live in us-east-1). Every key type and key-pair origin is requested: by default ACM lists only RSA_2048 certificates and leaves out ACME ones. Filter with CertificateStatuses[0]=ISSUED etc. Needs acm:ListCertificates.
- **get_single_amazon_web_services_acm_certificate_by_id** — ACM DescribeCertificate (id = certificate ARN): InUseBy, issuer, signature algorithm, renewal summary, validation, key usages. 404 aws_no_such_entity. Needs acm:DescribeCertificate.
- **get_single_amazon_web_services_acm_certificate_tag_by_id** — ACM ListTagsForCertificate (id = certificate ARN). Needs acm:ListTagsForCertificate.
- **list_all_amazon_web_services_codecommit_repositories** — CodeCommit ListRepositories: repository names and ids, 1,000 per page (fixed by AWS; ListRepositories takes no page size). sortBy repositoryName|lastModifiedDate, order ascending|descending. Needs codecommit:ListRepositories.
- **list_all_amazon_web_services_codecommit_repository** — CodeCommit BatchGetRepositories: pass repositoryNames (an array of up to 100 names). Repositories it could not read (errors[]) are returned as name-only rows with _truto_read_error, and ones that no longer exist as name-only rows with _truto_not_found. Needs codecommit:BatchGetRepositories.
- **get_single_amazon_web_services_codecommit_repository_by_id** — CodeCommit GetRepository (id = repository name): default branch, KMS key, clone URLs, dates. 404 aws_no_such_entity. Needs codecommit:GetRepository.
- **get_single_amazon_web_services_codecommit_tag_by_id** — CodeCommit ListTagsForResource (id = repository ARN): the tag map. Needs codecommit:ListTagsForResource.
- **get_single_amazon_web_services_dynamodb_table_tag_by_id** — DynamoDB ListTagsOfResource for one table ARN (the id): the whole response, {Tags: [{Key, Value}], NextToken}, so a NextToken stays visible. AWS limits this call to 10 per second per account. Needs dynamodb:ListTagsOfResource.
- **list_all_amazon_web_services_guardduty_list_findings_next** — GuardDuty ListFindings: finding ids on the detector in detector_id, up to 50 per page (nextToken). finding_criteria is GuardDuty's own FindingCriteria, sent as findingCriteria: an object (bracket query) or a JSON string, e.g. {"criterion":{"service.archived":{"equals":["false"]}}} for unarchived findings only. Without it every finding is listed, archived ones included. Needs guardduty:ListFindings.
- **list_all_amazon_web_services_iam_user_details** — IAM GetAccountAuthorizationDetails, users only, one row per user (UserDetailList). Account-wide and paged by IAM, which may return fewer than limit with a next_cursor. Needs iam:GetAccountAuthorizationDetails.
- **list_all_amazon_web_services_iam_group_details** — IAM GetAccountAuthorizationDetails, groups only, one row per group (GroupDetailList). Account-wide and paged by IAM, which may return fewer than limit with a next_cursor. Needs iam:GetAccountAuthorizationDetails.
- **list_all_amazon_web_services_sso_permission_sets** — IAM Identity Center ListPermissionSets: the instance's permission set ARNs. Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:ListPermissionSets.
- **get_single_amazon_web_services_sso_permission_set_by_id** — IAM Identity Center DescribePermissionSet: name, description, session duration and relay state of one permission set (id = permission set ARN). Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:DescribePermissionSet.
- **list_all_amazon_web_services_sso_account_assignments** — IAM Identity Center ListAccountAssignments: who (PrincipalType USER or GROUP, PrincipalId) holds permission_set_arn in target_account_id. All three of instance_arn, target_account_id and permission_set_arn are required by AWS; list the account's permission sets with sso-permission-sets-provisioned-to-account. Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:ListAccountAssignments.
- **list_all_sso_account_assignments_for_principal** — IAM Identity Center ListAccountAssignmentsForPrincipal: every account and permission set assigned to one user or group (principal_id = Identity Store UserId or GroupId, principal_type = USER or GROUP). Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:ListAccountAssignmentsForPrincipal.
- **list_all_sso_permission_sets_provisioned_to_account** — IAM Identity Center ListPermissionSetsProvisionedToAccount: the permission set ARNs provisioned to target_account_id. Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:ListPermissionSetsProvisionedToAccount.
- **list_all_sso_managed_policies_in_permission_set** — IAM Identity Center ListManagedPoliciesInPermissionSet: the AWS-managed policies (Name, Arn) attached to permission_set_arn. Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:ListManagedPoliciesInPermissionSet.
- **list_all_sso_customer_managed_policy_references** — IAM Identity Center ListCustomerManagedPolicyReferencesInPermissionSet: customer-managed policy names and paths referenced by permission_set_arn (resolved in each target account). Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:ListCustomerManagedPolicyReferencesInPermissionSet.
- **get_sso_inline_policy_for_permission_set_by_id** — IAM Identity Center GetInlinePolicyForPermissionSet: the inline policy document (InlinePolicy, a JSON string; empty when none) of one permission set (id = permission set ARN). Call it in the Identity Center home region (region=), with instance_arn from sso-instances. Needs sso:GetInlinePolicyForPermissionSet.
- **list_all_identitystore_group_memberships_for_member** — Identity Store ListGroupMembershipsForMember: the groups (GroupId) one user (user_id = UserId) belongs to, in identity_store_id. Home region only. Needs identitystore:ListGroupMembershipsForMember.
- **list_all_amazon_web_services_identitystore_groups** — Identity Store ListGroups: the groups (GroupId, DisplayName) in identity_store_id. Home region only. Needs identitystore:ListGroups.
- **get_single_amazon_web_services_iam_account_summary_by_id** — IAM GetAccountSummary: IAM entity counts and quotas for the calling account (the account whose role signs the call; it takes no account parameter), as { entry: [ { key, value } ] }. AccountMFAEnabled is 1 when the root user has an MFA device, AccountAccessKeysPresent is non-zero when the root user has access keys. Needs iam:GetAccountSummary.
- **get_iam_account_root_user_summary_by_id** — Internal helper for GET accounts/{id} (root_user_mfa_registered, root_user_has_static_credentials); direct callers should use iam-account-summary. It answers an access denial with 200 and a marker, { truto_read_error: { reason: "permission_denied", provider_error, provider_status } } in place of the summary map, instead of 403, so the account read can report it. Otherwise the same IAM GetAccountSummary call for the calling account. Needs iam:GetAccountSummary.
- **list_all_amazon_web_services_ec_2_boundary_flow_logs** — EC2 DescribeFlowLogs for the region, as ec2-flow-logs. Used by network_boundaries list for flow_logs_enabled and the flow-log destinations. Unlike ec2-flow-logs, an access denial answers 200 with one item { truto_read_error: { reason: "permission_denied", provider_error, provider_status } } instead of 403, so the list can report it; 429 and 5xx stay errors. Needs ec2:DescribeFlowLogs.

## How it works

1. **Link your customer's Amazon Web Services account.** Use Truto's frontend SDK; we handle every OAuth and API key flow so you don't need to create the OAuth app.
2. **Authentication is automatic.** Truto refreshes tokens, stores credentials securely, and injects them into every API request.
3. **Call Truto's API to reach Amazon Web Services.** The Proxy API is a 1-to-1 mapping of the Amazon Web Services API.
4. **Get a unified response format.** Every response uses a single shape, with cursor-based pagination and data in the `result` field.

## Use cases

- **Power a CSPM product with AWS configuration data** — SaaS companies building Cloud Security Posture Management tools can ingest S3 bucket encryption, public access block, IAM policies, Security Hub findings, and Config rule evaluations to score customer environments against CIS and PCI-DSS without writing per-service AWS SDK code.
- **Offer CIEM and identity hygiene insights** — Identity security products can map the full IAM graph — users, roles, groups, access keys, MFA devices, and SSO identities — to surface stale credentials, over-privileged principals, and shadow admin paths across every account in a customer's AWS Organization.
- **Centralize threat and vulnerability findings in a SIEM or ticketing workflow** — Security operations platforms can sweep GuardDuty, Inspector2, Security Hub, and Access Analyzer findings across every region and member account, then normalize them into unified Threat and Vulnerability Findings schemas for triage in Jira, ServiceNow, or an in-app dashboard.
- **Build a cloud asset inventory and network topology map** — DSPM, CMDB, and network visualization tools can enumerate VPCs, subnets, transit gateways, route tables, security groups, RDS instances, EC2 volumes, Lambda functions, EKS clusters, and S3 buckets to render a live topology of a customer's AWS footprint.
- **Enable multi-account enterprise onboarding** — Platforms serving enterprises can use AWS Organizations data to auto-discover every account and OU under a management account, so customers onboard their entire cloud estate in one connection instead of linking accounts one by one.

## What you can build

- **Stale credential detector** — Generate IAM credential reports and correlate access key last-used timestamps to flag long-lived keys and dormant admin users across every linked AWS account.
- **S3 public exposure scanner** — Evaluate bucket policies, ACLs, public access block settings, encryption, and ownership controls to produce a single effective-exposure verdict for every S3 bucket a customer owns.
- **Cross-region threat feed** — Iterate EC2 regions and pull GuardDuty, Inspector2, and Security Hub findings into a normalized Threat Findings stream that updates on a schedule or via Truto webhooks.
- **Snapshot leak monitor** — Inspect RDS snapshot, DB cluster snapshot, and EBS snapshot attributes to detect snapshots shared with external AWS account IDs or made public.
- **Transit Gateway blast-radius visualizer** — Pull TGW attachments, route tables, associations, and propagations to render lateral-movement paths and highlight flat-mesh routing between VPCs and accounts.
- **Organization-wide compliance dashboard** — Use AWS Organizations to enumerate accounts and OUs, then aggregate Config rule compliance and Security Hub standards results into a single posture score per account, OU, or framework.

## FAQs

### How do end users authenticate their AWS account?

AWS integrations are typically authorized via IAM role assumption using an external ID and a cross-account trust policy. Truto handles the STS assume-role flow and credential refresh so your application only sees a connected account ID.

### Does Truto handle multi-account AWS Organizations setups?

Yes. You can list all accounts, roots, OUs, and policy targets under a management account using the Organizations tools, then query member accounts individually. This is the standard pattern for enterprise customers with dozens or hundreds of accounts.

### How are regional services like GuardDuty and Inspector handled?

GuardDuty, Inspector2, Security Hub, and most EC2 resources are region-scoped. You can enumerate active regions with the EC2 regions tool and iterate the per-region list endpoints — Truto exposes region as a parameter on these calls.

### What AWS data maps into Truto's Unified Cloud Infrastructure API?

AWS resources normalize into unified models including Object Stores (S3), Relational Databases (RDS), Workload Identities (IAM roles), Human Identities (IAM users, Identity Store), Static Credentials (access keys), Firewall Rule Sets (security groups), Network Boundaries (VPCs/subnets), Threat Findings (GuardDuty), Vulnerability Findings (Inspector2), Posture Findings (Security Hub), and more.

### Are AWS API rate limits abstracted?

Yes. AWS enforces per-service throttling (e.g., IAM, EC2, Config each have their own limits). Truto applies backoff and retry logic transparently, and pagination across large result sets like S3 buckets, IAM entities, and findings is handled by the list endpoints.

### Can I get fresh data on demand or is it synced?

Both. You can call any list or get tool on demand for real-time reads, or configure Truto to sync resources into the Unified Cloud Infrastructure models on a schedule with change events delivered via webhooks.

### Is write access to AWS supported?

The current AWS tool set is read-only and optimized for discovery, inventory, and posture use cases (CSPM, CIEM, DSPM, SIEM). If you need write actions for a specific workflow, Truto can add them on request.
