No entries match these filters. Clear filters
safe-chain adds a Containers tab to Device Protection that lists every protected Docker container and WSL2 WSL2 A Microsoft-maintained compatibility layer in Windows 10/11 that runs a real Linux kernel inside a lightweight VM, letting cyber tools built for Linux execute natively on Windows without dual-booting or a full VM. distro linked to a host device︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
safe-chain adds Azure DevOps support for release gating via scan-release, with --fail-on-sast-scan, --fail-on-iac-scan, and --fail-on-secrets-scan flags.
New integration Run a release-gating scan in Azure DevOps pipelines using aikido-api-client scan-release with the commit SHA and repository name as arguments, authenticated via --apikey and the AIKIDO_CLIENT_API_KEY secret variable. Not checked yet Fail a pipeline selectively by scan category using --fail-on-sast-scan, --fail-on-iac-scan, and --fail-on-secrets-scan flags on the scan-release command. Not checked yet Install the Aikido CI client in any pipeline with npm install --global @aikidosec/ci-api-client using Node.js 22.x. Not checked yet
3 details in Everything
safe-chain adds aikidoctl request to request admin approval for blocked package installs from the command line︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
On a CI runner or headless server where blocked installs need admin approval, send a request for specific packages without any interactive prompt.
copy
$ aikidoctl request --package lodash,chalk --yes
Inspect all packages currently waiting for approval in a script-friendly format before deciding which to request.
copy
$ aikidoctl request --json
Send an approval request for a single known block by its event ID, non-interactively, useful when only one specific block should be escalated.
copy
$ aikidoctl request --event <event-id> --yes
Request admin approval for blocked packages on headless machines using aikidoctl request, with --yes or -y to approve every waiting package non-interactively.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Filter requests to specific packages using --package with one name, repeated, or comma-separated (e.g., lodash,chalk). Not checked yet Target a specific block by its ID or unique prefix using --event <event-id> on aikidoctl request. Not checked yet Inspect the full waiting list as machine-readable output using aikidoctl request --json, which prints one JSON object per line with fields product, package, versions, status, blocked_at, and event_id, without sending any requests. Not checked yet
4 details in Everything
safe-chain adds the aikidoctl CLI for inspecting, controlling, and troubleshooting Aikido Device Protection agents from the command line︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Reset device identity after cloning a VM so the new machine registers as its own device rather than collapsing into the clone source's entry.
copy
$ aikidoctl.exe device-identity reset
Emit only the installed version number in a script that checks the agent version before automated deployment.
copy
$ aikidoctl --simple
Adds certs to inspect and manage CA certificates on macOS (not available on Windows or Linux), and certconfig to manage per-tool certificate trust configuration for node, git, java, and others.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Adds pause/unpause to temporarily disable and resume all device protection, restart-daemon to restart the Device Protection daemon on Linux, version to show the installed Device Protection version, build details, and service state, and the --simple global flag to print only the installed version for use in scripts. Not checked yet Inspect on-device secret scanning secret scanning Checking files, code or other data for exposed credentials such as passwords, API keys and tokens. Findings help teams remove the secret and revoke or rotate it. on macOS with the secretscanning subcommand (not available on Windows or Linux). Not checked yet Show or reset the device's registered identity with the device-identity subcommand; run device-identity reset after deploying a VM clone or snapshot so each machine registers independently. Not checked yet Collect and submit diagnostics to Aikido — the command-line equivalent of the in-app Report A Problem flow — with the diagnostics subcommand. Not checked yet
+1 more
show less
List packages blocked pending admin approval and send an approval request with the request subcommand. Not checked yet
6 details in Everything
safe-chain adds a Containers tab to Device Protection, listing linked Docker containers and WSL2 WSL2 A Microsoft-maintained compatibility layer in Windows 10/11 that runs a real Linux kernel inside a lightweight VM, letting cyber tools built for Linux execute natively on Windows without dual-booting or a full VM. distros per host with last-seen timestamps.
New supported platform A new Containers tab in Device Protection now lists every Docker container and WSL2 distro protected on a host device, including stopped containers with their last-seen timestamp — visible once Protect Containers or Protect WSL2 is enabled Not checked yet
1 detail in Everything
safe-chain: Authenticate the Aikido CLI in Azure DevOps by storing the API token as AIKIDO_CLIENT_API_KEY in pipeline secret variables.
└──▷ TRY IT
Gate an Azure DevOps release pipeline on SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. , IaC IaC Infrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. , and secrets findings — the pipeline fails if Aikido detects new issues on push to main.
copy
$ npm install --global @aikidosec/ci-api-client
aikido-api-client scan-release \
"$(Build.Repository.Name)" \
"$(Build.SourceVersion)" \
--apikey "$AIKIDO_CLIENT_API_KEY" \
--fail-on-sast-scan \
--fail-on-iac-scan \
--fail-on-secrets-scan
Full Azure DevOps pipeline YAML that triggers Aikido release gating on every push to main, using Node 22 and a secret AIKIDO_CLIENT_API_KEY variable.
azure-pipelines.yml
copy
trigger:
include:
- main
pool:
vmImage: 'ubuntu-latest'
steps:
- task: UseNode@1
displayName: 'Set up Node.js'
inputs:
version: '22.x'
checkLatest: true
- bash: |
npm install --global @aikidosec/ci-api-client
aikido-api-client scan-release \
"$(Build.Repository.Name)" \
"$(Build.SourceVersion)" \
--apikey "$AIKIDO_CLIENT_API_KEY" \
--fail-on-sast-scan \
--fail-on-iac-scan \
--fail-on-secrets-scan
workingDirectory: '$(Build.SourcesDirectory)'
displayName: 'Run Aikido Scan'
env:
AIKIDO_CLIENT_API_KEY: $(AIKIDO_CLIENT_API_KEY)
safe-chain adds auto-approval of pull requests when a gate passes, alongside the existing green check︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Automatically approve pull requests when a gate passes, in addition to adding a green check — requires AutoFix to be enabled.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet
1 detail in Everything
safe-chain adds admin-initiated remote log collection for devices from the Aikido dashboard︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Admins can now request diagnostic logs from any user device directly from the Device Protection device list — no action required from the end user — with requests queued when the device is offline and executed on reconnect.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet
1 detail in Everything
safe-chain adds remote log collection for managed devices, with offline queuing until the device reconnects︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ HOW TO FIND IT
When a device is behaving unexpectedly, remotely trigger log collection instead of asking the end user to file a report themselves.
📍 In the device management console, click the target device › open its details › select Collect Logs . If the device is offline, the request queues and runs automatically when the device next comes online. Then contact Aikido Support and mention which device you collected logs from.
Collect diagnostics logs from a managed device remotely using the Collect Logs action, which sends the same data as the in-app Report A Problem flow without requiring user interaction.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet
1 detail in Everything
safe-chain adds a Spending Limit cap to control the maximum credits a single continuous pentest run can spend︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Set a Spending Limit on a continuous pentest run to cap the number of agents launched and the credits spent; leave it empty for no per-run cap.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet
1 detail in Everything
safe-chain adds an Aikido Broker to scan internal applications on private networks via Docker or Kubernetes︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Start the broker on a host with access to your internal network, restricting it to RFC-1918 ranges and pointing it at the US Aikido region.
copy
$ docker run -d \
--name aikido-broker \
--network host \
-e CLIENT_SECRET="AIK_BROKER_XXX_YYY_ZZZZ" \
-e ALLOWED_INTERNAL_SUBNETS="192.168.0.0/16,10.0.0.0/8,172.16.0.0/12" \
-e BROKER_TARGET_URL="https://broker.us.aikidobroker.com" \
aikidosec/broker
Deploy the broker on Kubernetes with a private CA and internal DNS resolver so it can reach services that use internal PKI PKI A framework of policies, roles, and cryptographic standards for issuing and managing digital certificates that bind public keys to identities, letting tools verify authenticity and encrypt communications without pre-shared secrets. and private DNS zones.
copy
$ helm install broker-client aikido/broker-client \
--set config.clientSecret="AIK_BROKER_XXX_YYY_ZZZZ" \
--namespace aikido \
-f values.yaml
Run the built-in diagnostics to quickly surface DNS or connectivity failures in a running broker container.
copy
$ docker exec -it aikido-broker node /app/diagnostics.cjs
Deploy the broker as a Docker container, using CLIENT_SECRET and ALLOWED_INTERNAL_SUBNETS environment variables to control access to internal CIDR CIDR A notation for a block of IP addresses, such as 10.0.0.0/24. The number after the slash says how many leading address bits identify the network. ranges, or on Kubernetes via the Helm chart at aikido/broker-client, configured through config.clientSecret or a values.yaml file.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Configure the broker's internal network and TLS handling via environment variables: DNS_SERVERS to point it at internal DNS resolvers for zones such as *.corp.local; NODE_EXTRA_CA_CERTS with a mounted private CA file to validate TLS for services signed by an internal PKI or self-signed certificates; MTLS_PEM_PATH and MTLS_CA_PATH to enable mTLS mTLS Mutual TLS, an encrypted connection where both client and server present certificates to prove their identity. It helps services verify who they are communicating with. for internal services; HTTP_PROXY, HTTPS_PROXY, or ALL_PROXY with per-host exclusions via NO_PROXY to route traffic through a corporate proxy; FORCE_WEBSOCKET or FORCE_POLLING to force a specific transport when the default WSS-with-polling-fallback behaviour is restricted; and NODE_TLS_REJECT_UNAUTHORIZED set to 0 to disable TLS verification in self-signed certificate environments. Not checked yet Run a built-in health check covering DNS and connectivity with docker exec -it aikido-broker node /app/diagnostics.cjs or kubectl exec <pod> -c broker-client -- node /app/diagnostics.cjs. Not checked yet Scan internal code repositories, container registries, internal domains, APIs, and AI pentest targets by configuring their URLs as broker resources and using the generated *.aikidobroker.com Broker URLs in Aikido scans. Not checked yet
4 details in Everything
safe-chain adds an official Terraform/OpenTofu OpenTofu An open-source fork of Terraform, maintained by the Linux Foundation, that provisions and manages infrastructure as code; tools use it to automate cloud or on-premises environment setup without vendor lock-in. provider to manage repositories, AutoFix settings, and PR checks as code via the Aikido management API︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Declare the Aikido provider using environment variables for credentials so secrets never appear in source control, then activate a repository with sensitivity and labels.
main.tf
copy
# terraform.tfvars or environment
export AIKIDO_CLIENT_ID=your-client-id
export AIKIDO_CLIENT_SECRET=your-client-secret
# main.tf
terraform {
required_providers {
aikido = {
source = "AikidoTerraform/aikido"
version = "~> 1.1"
}
}
}
provider "aikido" {}
resource "aikido_repository" "app" {
id = "12345"
active = true
sensitivity = "sensitive"
connectivity = "connected"
labels = ["payments", "production"]
}
Configure workspace AutoFix to auto-remediate only critical and high dependency findings across all repos, using Aikido's curated library for major-version upgrades.
main.tf
copy
resource "aikido_autofix_dependency_settings" "workspace" {
enabled = true
severity_filter = "critical_and_high_only"
repos_scope = "all"
use_aikido_library_for_major = true
}
Block PRs that introduce high-severity findings across all major scan types, and post inline code comments for critical issues, targeting a US-region workspace.
main.tf
copy
provider "aikido" {
client_id = var.aikido_client_id
client_secret = var.aikido_client_secret
base_url = "https://app.us.aikido.dev/api"
}
resource "aikido_repo_pr_checks_settings" "app" {
code_repo_id = "12345"
minimum_severity = "high"
fail_on_dependency_scan = true
fail_on_sast_scan = true
fail_on_iac_scan = true
fail_on_secrets_scan = true
fail_on_malware_scan = true
post_inline_comments_min_severity = "critical"
minimum_license_severity = "high"
fail_on_code_quality_scan = true
enable_code_quality_scan = true
post_code_quality_inline_comments_min_severity = "medium"
}
Configure the Aikido Terraform provider with client_id and client_secret provider block arguments, or supply credentials via the AIKIDO_CLIENT_ID and AIKIDO_CLIENT_SECRET environment variables.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Target a non-default Aikido region by setting base_url in the provider block or via the AIKIDO_BASE_URL environment variable — supported values include https://app.us.aikido.dev/api, https://app.au.aikido.dev/api, and https://app.me.aikido.dev/api. Not checked yet Manage code repository activation, sensitivity, connectivity, and labels with the aikido_repository resource. Not checked yet Configure workspace-level AutoFix for dependencies, SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. , and Aikido Pentest findings with the aikido_autofix_dependency_settings resource, including severity_filter, repos_scope, and use_aikido_library_for_major arguments. Not checked yet Gate pull requests per repository — controlling severity thresholds, scan types (dependency, SAST, IaC IaC Infrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. , secrets, malware, code quality), inline comments, and Deep Review — with the aikido_repo_pr_checks_settings resource. Not checked yet
+2 more
show less
Adds an official Terraform/OpenTofu provider for Aikido, published on both the Terraform Registry and the OpenTofu Registry, requiring Terraform 1.0 or later or OpenTofu. Not checked yet Map repositories and containers to teams, and set up teams and user permissions, through the provider. Not checked yet
7 details in Everything
Aikido Webhooks adds HMAC-SHA256 HMAC-SHA256 A keyed hash that proves a message came from someone holding a shared secret and was not altered. Webhooks and API request signing commonly use it. signature verification and replay-attack protection via X-Aikido-Webhook-Signature and dispatched_at︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Verify an incoming Aikido webhook in an Express middleware — check the HMAC HMAC A keyed hash that proves a message came from someone holding a shared secret and was not altered. Webhooks and API request signing commonly use it. -SHA256 SHA256 A cryptographic hash function that turns data into a fixed 256-bit fingerprint. It is used to check integrity; hashing does not encrypt the original data. signature and reject payloads older than 30 seconds.
javascript
copy
const crypto = require('crypto');
app.use('/webhook', express.json(), (req, res, next) => {
const signature = req.headers['x-aikido-webhook-signature'];
const body = JSON.stringify(req.body);
const digest = crypto.createHmac('sha256', process.env.AIKIDO_WEBHOOK_SECRET)
.update(body)
.digest('hex');
if (digest !== signature) {
return res.status(401).send('Invalid signature');
}
const age = Math.floor(Date.now() / 1000) - req.body.dispatched_at;
if (age > 30) {
return res.status(401).send('Webhook too old — possible replay attack');
}
next();
});
Verify incoming webhooks are genuine by checking the HMAC-SHA256 digest sent in the X-Aikido-Webhook-Signature request header against your shared webhook secret.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Protect against replay attacks by validating the dispatched_at epoch timestamp in the webhook payload is no older than 30 seconds. Not checked yet Generate a webhook signing secret from the Aikido webhooks integration page by clicking Add secret . Not checked yet
3 details in Everything
Aikido adds Code Context to teach the scanner your team's coding conventions and acceptable exceptions, scoped per repo or org-wide︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Apply org-wide coding conventions — such as acceptable wildcard imports from an internal library — so Aikido stops flagging them in every PR.
copy
$ # In your repository root
cat > AGENTS.md << 'EOF'
Wildcard imports (import * as) are acceptable when importing from our @company/ui-components library as we intentionally export all components as a single namespace.
EOF
git add AGENTS.md && git commit -m 'Add Aikido Code Context'
Commit AGENTS.md at the repository root to apply Code Context guidance to every pull-request review automatically.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Add Code Context through the Code Quality Checks page to teach Aikido your team's conventions and acceptable exceptions, scoped to All repositories or Selected repositories . Not checked yet Post a comment to @AikidoSec in a pull request to generate Code Context directly from inline feedback. Not checked yet
3 details in Everything
Aikido adds Custom Code Rules so teams can define AI-powered checks for org-specific coding standards and patterns, with optional AutoFix prompts︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ HOW TO FIND IT
Walk through creating and saving a Custom Code Rule that enforces Alpine base images in Dockerfiles, with an AutoFix prompt.
📍 1. Go to Code Quality › Checks
2. Click Add Custom Code Rule
3. Enter description: 'Allow only Alpine base images in Dockerfiles as base images'
4. Select target language: Dockerfile
5. Click Generate Examples , then review/edit compliant (FROM alpine:3.18) and non-compliant (FROM ubuntu:latest) samples
6. Click Validate Rule
7. Add AutoFix prompt: Replace the base image with alpine , then click Validate Prompt
8. Fill in Title, TL;DR, and How-to-fix fields, then click Save Rule
Adds Custom Code Rules in Code Quality > Checks to let teams define AI-powered checks for patterns and standards not covered by default checks, including optional AutoFix prompts that auto-remediate violations.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Adds AutoFix prompt authoring and validation to Custom Code Rules, letting teams specify how Aikido should correct a violation; the prompt is validated against the non-compliant example before it can be saved. Not checked yet
2 details in Everything
safe-chain adds Azure DevOps PR gating via Microsoft Entra service principal, replacing personal access tokens personal access tokens A personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. with short-lived federated credentials︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Grant the Aikido PR Checks service principal permission to view and manage service hooks on a specific Azure DevOps project, without adding it to any administrator group.
copy
$ az devops permission update --org https://dev.azure.com/<your-organization> \
--namespace-id cb594ebe-87dd-4fc9-ac2c-6a10a4c92046 \
--subject <application-descriptor> \
--token PublisherSecurity/<project-id> \
--allow-bit \
-o table
Authenticate Azure DevOps PR checks as a Microsoft Entra service principal instead of a personal access token personal access token A personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. — no secret is created, stored, or rotated; Aikido uses short-lived federated credentials issued per call.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Use az devops permission update with namespace ID cb594ebe-87dd-4fc9-ac2c-6a10a4c92046 and --allow-bit to grant the service principal ViewSubscriptions and EditSubscriptions on Azure DevOps service hooks without adding it to any administrator group. Not checked yet Configure the Entra service principal in Aikido under General Settings > PR Checks > Use An Entra Service Principal Instead , supplying the Directory (tenant) ID and Application (client) ID from your Entra app registration. Not checked yet Switch back to a personal access token access token A credential that authorizes a client to use a service or API. It usually has limited permissions and a lifetime; anyone holding it may be able to use those permissions. at any time via Use a personal access token instead on the PR Checks tab; existing repository configurations are preserved. Not checked yet Revoke Aikido's access immediately by deleting the federated credential in Microsoft Entra or removing the application from your Azure DevOps organization. Not checked yet
5 details in Everything
safe-chain adds OpenShift OpenShift Red Hat's Kubernetes distribution that adds developer workflows, built-in CI/CD, and stricter security defaults on top of upstream Kubernetes, giving cyber tools a managed container platform with enterprise auth and RBAC already wired in. support, FIPS FIPS A set of US federal standards published by NIST that define approved cryptographic algorithms and modules; cyber tools use it to meet compliance requirements in government and regulated environments. -compliant agent images, runtime threat detection, image mirror resolution, custom CA cert mounting, and namespace exclusion for Kubernetes clusters.
New detection Enable runtime threat detection on a connected cluster with the runtimeDetection.enabled=true Helm flag, then toggle threat detection in the cluster's edit settings. Not checked yet Toggle in-cluster image scanning on an existing cluster with sbomCollector.enabled=true (or false) via helm upgrade aikido -n aikido --set sbomCollector.enabled=true. Not checked yet Mount custom CA certificates for OpenShift clusters by adding extra volumes and pointing the agent to the trust bundle via SSL_CERT_FILE or SSL_CERT_DIR environment variables. Not checked yet New supported platform Use FIPS-compliant Kubernetes agent and SBOM SBOM A software bill of materials, an inventory of the components and versions inside an application. It helps teams find affected software when a dependency has a vulnerability or licensing issue. collector images (public.ecr.aws/aikido-cloud/kubernetes-agent-fips:v1.19.0 and public.ecr.aws/aikido-cloud/kubernetes-sbom-collector-fips:v1.9.0) by appending -fips to repo values in your Helm chart. Not checked yet New supported platform safe-chain now supports OpenShift clusters, including on-prem and managed OpenShift Container Platform environments. Not checked yet
+4 more
show less
Automatically resolve OpenShift image mirror mappings from ImageContentSourcePolicies, ImageDigestMirrorSets, and ImageTagMirrorSets so images are correlated by their original source reference. Not checked yet Mount image pull secrets directly into the agent as an alternative to RBAC RBAC Role-based access control, where permissions are assigned to roles such as reader or administrator and users receive those roles. It makes access easier to manage than individual permission lists. -based registry access for restricted OpenShift environments. Not checked yet Exclude specific Kubernetes namespaces from resource collection when connecting a cluster, with quick-add options for commonly excluded namespaces such as kube-system and kube-public. Not checked yet Scan container images deployed in a cluster by enabling the in-cluster SBOM collector component, which generates SBOMs SBOMs A software bill of materials, an inventory of the components and versions inside an application. It helps teams find affected software when a dependency has a vulnerability or licensing issue. that safe-chain then scans for vulnerabilities. Not checked yet
9 details in Everything
safe-chain adds a Secret Keyword Filter so workspace admins can skip lines containing known-safe strings before gitleaks evaluates them︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Enable the Secret Keyword Filter for workspace-wide secrets-scan noise reduction by going to Settings > Advanced > Secret Keyword filter and clicking Manage; keywords must be at least 5 characters, are case-sensitive, and use plain string matching — the entire line containing a match is excluded from detection.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Use .aikido file exclusions to exclude specific files or paths from scanning as an alternative suppression method alongside the keyword filter and per-line code comments. Not checked yet
2 details in Everything
safe-chain adds .aikido config file support to exclude paths and ignore CVEs per repository︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Exclude test fixtures and a generated directory from all scans — secrets, SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. , lockfiles, and Deep Review — so noise from non-production paths does not block CI.
.aikido
copy
exclude:
paths:
- tests/fixtures
- generated/
Suppress a known, accepted CVE across all scans for this repository and surface the business reason in the Aikido UI.
.aikido
copy
ignore:
CVE-2023-12345:
reason: We do not care about this CVE
CVE-2024-67890:
reason: We handle this
Exclude files and directories from all code scanning (secrets, SAST, lockfiles, code quality, and Deep Review) by listing paths under the exclude.paths key in a .aikido file at the repository root.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Ignore specific CVEs with a documented reason by adding them under the ignore key in .aikido, with the Aikido UI surfacing those suppressions alongside a reference to the file. Not checked yet Configure AutoFix PR behavior per repository — such as a custom branch prefix or a single PR for the whole repo — via the autofix key in .aikido; see AutoFix PR Configuration for supported keys. Not checked yet
3 details in Everything
Aikido adds encrypted environment variable injection for private package registry authentication in AutoFix workflows︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Tell NPM where to find a private registry and authenticate without storing a token in the repo — Aikido injects NPM_TOKEN at AutoFix runtime.
.npmrc
copy
@my-org:registry=https://npm.pkg.github.com/
//npm.pkg.github.com/:_authToken=${NPM_TOKEN}
Authenticate AutoFix with private package registries by supplying encrypted environment variables via the AutoFix settings page, supporting package managers including NPM (via .npmrc), Poetry, and Bundle.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet
1 detail in Everything
safe-chain adds private PIP registry and private Git repository authentication for autofix dependency updates︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Connect a private PyPI registry for autofix dependency updates by providing an index URL with credentials via the Connect Registry modal in account autofixer settings.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Authenticate with private Git repositories in a different SCM SCM Source Code Management: version-control systems (Git, SVN, etc.) that track changes to code over time, giving cyber tools reproducible builds, audit trails, and collaborative development workflows. organisation by setting environment variables — PIP_GITHUB__COM, PIP_GITLAB__COM, PIP_BITBUCKET__ORG, or PIP_DEV__AZURE__COM — via the Connect Registry > Environment Variables modal, with the value being an access token access token A credential that authorizes a client to use a service or API. It usually has limited permissions and a lifetime; anyone holding it may be able to use those permissions. with repository access. Not checked yet
2 details in Everything
safe-chain adds Aikido Images, a registry of zero-CVE drop-in base image replacements with AutoFix and Local Scanner support︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Manually switch a Dockerfile from a vulnerable Debian Bookworm base image to its Aikido Images equivalent when AutoFix cannot open a PR (e.g. Local Scanner workflow).
Dockerfile
copy
# In your Dockerfile, replace the existing FROM line:
# FROM debian:bookworm
# with the Aikido Images variant (token visible in Aikido UI):
FROM docker.aikido.io/<token>/debian:bookworm
Browse available Aikido Images — zero-CVE drop-in base image replacements for CRITICAL and HIGH severity issues — at AutoFix > Cloud > Containers > Aikido Images in the Aikido UI.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Aikido Images are hosted on docker.aikido.io and support both amd64 and arm64 architectures. Not checked yet When using the Aikido Local Scanner without source-code access, Aikido displays the suggested Aikido Image in the UI for manual Dockerfile edits using a read-only account-specific token. Not checked yet
3 details in Everything
safe-chain adds per-repo AutoFix PR configuration via a .aikido file, with controls for branch names, commit templates, conditional auto-merge, and task creation︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Auto-merge only low-risk SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. fixes without manual review, while keeping higher-severity issues in draft for human sign-off.
.aikido
copy
autofix:
pr_config:
draft_pr: true
auto_merge:
sast:
enabled: true
severity: low
Enforce a Jira-linked branch naming convention and auto-label PRs by fix type and severity for triage dashboards.
.aikido
copy
autofix:
pr_config:
branch_prefix: "fix/$TASK_REF-"
title_template: "Aikido: $MSG [$AUTOFIX_TYPE]"
labels: "aikido,$AUTOFIX_TYPE,$SEVERITY"
create_task_for_pr: true
task_title_prefix: "[AIK] "
Isolate a legacy service directory so AutoFix never opens PRs against it, while grouping all other lockfiles into one PR per service directory.
.aikido
copy
autofix:
dependency:
update_policy: minor_and_patch_versions_only
lockfile_groups:
- group_name: service1
lockfiles:
- 'service1/*'
- group_name: legacy
disabled: true
lockfiles:
- 'legacy/*'
automated:
disabled: false
Adds per-repository AutoFix PR customization via autofix.pr_config keys in a .aikido file at the repo root, overriding global UI settings, supporting branch_prefix, title_prefix, commit_prefix, labels, draft_pr, auto_merge, create_task_for_pr, task_title_prefix, disable_codeowners_as_reviewers, and self_approve_pr; define a full branch-name format using $TASK_REF, $TASK_REF_LOWER, $TASK_REF_0, and $TASK_REF_0_LOWER variables; set a full PR title format with title_template using $MSG as a placeholder, taking precedence over title_prefix; set a full commit-message format with commit_template using $MSG as a placeholder, taking precedence over commit_prefix; enable conditional auto-merge via a nested auto_merge block with enabled and severity keys for types dependency, sast, iac, container, code-quality, pentest, and code-audit at severity levels critical, high, medium, and low, or via a free-text instruction evaluated against AutoFix type, severity, confidence, PR title and description, changed files, and diffs; automatically create an issue-tracker task per AutoFix PR with create_task_for_pr: true and optional task_title_prefix; disable CODEOWNERS as GitLab merge-request reviewers with disable_codeowners_as_reviewers: true; and enable Bitbucket self-approval of AutoFix PRs with self_approve_pr: true.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Opt a specific repository out of globally enabled Automated AutoFixes by setting autofix.automated.disabled: true in the .aikido file. Not checked yet Group lockfiles into a single AutoFix PR per group using autofix.dependency.lockfile_groups in the .aikido file, with group_name, lockfiles (glob patterns), and optional disabled: true to suppress PRs for a group. Not checked yet Target a custom base branch for AutoFix PRs by setting base_branch in the .aikido file. Not checked yet
4 details in Everything
safe-chain adds custom context instructions for AutoFix SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. rules, configurable per repository or per language via .aikido files︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Supply repo-wide AutoFix instructions — custom error handling, logging patterns, or required unit tests — so every SAST autofix in this repository follows your standards.
.aikido
copy
# .aikido (repo root)
- path: aikido-autofix-all.md
Restrict custom AutoFix context to Python files only, keeping other language fixes unaffected.
.aikido
copy
# .aikido (repo root)
- path: aikido-autofix-python.md
language: python
Add a path: aikido-autofix-all.md entry to the .aikido file in a repository to supply custom instructions — such as error-handling patterns, comment styles, or unit test requirements — to all AutoFix SAST rules for that repo.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Scope AutoFix custom context to a specific language by setting the language key in the .aikido file, with supported values including go, python, javascript, typescript, java, rust, terraform, kotlin, swift, php, ruby, scala, c, dart, elixir, clojure, .net, android, ios, vb, and yaml. Not checked yet
2 details in Everything
Aikido MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. adds token-based authentication via AIKIDO_API_KEY for headless and CI/CD environments across Claude Code, Cursor, JetBrains, Copilot, and more︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Run Aikido MCP in a CI/CD pipeline or headless Docker environment where no browser is available, using a Personal Access Token.
copy
$ export AIKIDO_API_KEY="your-token-here"
claude
Register Aikido MCP with GitHub Copilot Cloud Agents using a repository secret prefixed with COPILOT_MCP_ so Copilot exposes it to the MCP config.
json
copy
{
"aikido": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@aikidosec/mcp@latest"],
"tools": ["aikido_scan_paths", "aikido_full_scan"],
"env": {
"AIKIDO_API_KEY": "COPILOT_MCP_AIKIDO_API_KEY"
}
}
}
Authenticate Aikido MCP in OpenCode by storing the token in the environment block of the OpenCode config file.
~/.config/opencode/opencode.json
copy
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"aikido": {
"type": "local",
"command": ["npx", "-y", "@aikidosec/mcp@latest"],
"enabled": true,
"environment": {
"AIKIDO_API_KEY": "your-token-here"
}
}
}
}
safe-chain adds an Aikido MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. plugin for GitHub Copilot to scan AI-generated code for vulnerabilities and hardcoded secrets in real time︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Wire the Aikido MCP server into GitHub Copilot IDE so every AI-generated snippet is scanned automatically.
mcp.json
copy
{
"servers": {
"aikido": {
"command": "npx",
"args": ["-y", "@aikidosec/mcp@latest"]
}
}
}
Connect Aikido MCP to a GitHub Copilot Cloud Agent (headless CI) using a repository-scoped secret for token authentication.
mcp.json
copy
{
"servers": {
"aikido": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@aikidosec/mcp@latest"],
"tools": ["aikido_scan_paths", "aikido_full_scan"],
"env": {
"AIKIDO_API_KEY": "COPILOT_MCP_AIKIDO_API_KEY"
}
}
}
}
Configure the Aikido MCP server in GitHub Copilot's mcp.json by adding the @aikidosec/mcp@latest package via npx under the aikido server key to scan AI-generated snippets automatically.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Authenticate the Aikido MCP plugin in VS Code by asking Copilot to run aikido_login, which returns a browser link and stores the token on completion. Not checked yet Configure Aikido MCP for GitHub Copilot Cloud Agents (headless) by adding a stdio-type MCP config to repository Settings → Copilot → Coding agent, using aikido_scan_paths and aikido_full_scan tools and injecting credentials via the COPILOT_MCP_AIKIDO_API_KEY secret — secrets must be prefixed COPILOT_MCP_ to be exposed. Not checked yet Add prompt instructions to AGENTS.md or Copilot custom instructions to direct the agent to use Aikido MCP tooling on every generated snippet. Not checked yet
4 details in Everything
Aikido MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. adds eleven agent-callable tools covering SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. , secrets, IaC IaC Infrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. , OSS OSS Open-source software: code whose licence permits anyone to inspect, modify, and redistribute it. Cyber tools build on it to reuse auditable libraries without licensing fees. dependencies, Deep Review, and issue triage︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ HOW TO FIND IT
Run a Deep Review over only your uncommitted edits before opening a PR, so the agent surfaces auth and data-flow issues the diff alone would miss.
📍 Ask your agent: 'Run an Aikido Deep Review on my uncommitted changes only' — the agent calls aikido_deep_review with directory set to your repo root and scope set to uncommitted , then polls aikido_deep_review_status until status is completed .
Snooze a known issue for the sprint duration so it stops appearing in triage without dismissing it permanently.
📍 Ask your agent: 'Snooze Aikido issue <issue_id> for 1 month, we are fixing it in the next release' — the agent calls aikido_snooze_issue with issue_id, reason, and duration set to 1_month.
Persist a 'scan only changed files at commit time' policy so the agent follows it every session without a reminder.
CLAUDE.md
copy
# Add to CLAUDE.md, .cursor/rules, or AGENTS.md:
Don't run any Aikido scans after each edit. Run one aikido_scan_paths scan on staged files right before I commit.
Adds aikido_scan_paths to scan files on disk for SAST findings, hardcoded secrets, and IaC misconfigurations using paths and root (with optional repository_name to link results to a repo), aikido_full_scan to scan in-memory or unsaved code by passing relativeFilePath and content entries in files_to_scan, and aikido_oss_dependency_scan to scan dependency manifests and lockfiles such as package-lock.json, requirements.txt, and go.sum via paths and root for vulnerable, outdated, or malicious packages.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Adds aikido_deep_review to start an AI security review tracing data flows, auth boundaries, and business logic across the repository (directory, optional base_ref, scope of all, committed, or uncommitted, and include_untracked), and aikido_deep_review_status to poll an in-progress review via agent_id, returning completed, skipped, failed, or running. Not checked yet Adds aikido_issues_list to fetch security issues from the Aikido feed, filterable by scope (repo_name, cloud_name, vm_name, domain_name, container_name, workspace_name, team_name), issue type (sast, leaked_secret, iac, open_source, cloud, cloud_instance, docker_container, malware, eol, mobile, surface_monitoring, scm_security, ai_pentest, ai_code_analysis), severity (low, medium, high, critical), labels, SLA SLA A formal agreement between a service provider and customer that defines expected uptime, response times, and remedies for failures, giving cyber tools a contractual baseline for availability and support commitments. filters (out_of_sla, sla_due_soon), and page, or by passing issue_id to fetch full details of a single issue — including score_reasons, pentest attack analysis, and licenses. Not checked yet Adds issue management tools: aikido_ignore_issue and aikido_unignore_issue (requiring issue_id and a reason), aikido_snooze_issue (requiring issue_id, a reason, and either a duration of 1_day, 1_week, 1_month, 3_months, 6_months, or 1_year, or a snooze_until_date in ISO 8601 ISO 8601 An international standard for representing dates and times as text, using formats like 2025-04-23T14:00:00Z. Tools adopt it to ensure timestamps are unambiguous and sortable across systems and locales. format) and aikido_unsnooze_issue (requiring issue_id), and aikido_adjust_issue_severity (requiring issue_id, adjusted_severity of critical, high, medium, or low, and a reason). Not checked yet Start the Aikido sign-in flow with aikido_login; optionally set region (eu, us, me, au) or use force_reauth to switch accounts. Not checked yet
+1 more
show less
Persist scanning instructions across sessions by adding them to an agent rules file such as CLAUDE.md, .cursor/rules, or AGENTS.md. Not checked yet
6 details in Everything
safe-chain adds MDM MDM Mobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. deployment support for the Aikido IDE plugin across JetBrains IDEs and VS Code-based editors on macOS and Windows︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Pre-install the Aikido JetBrains plugin at scale on macOS developer machines via Jamf or Kandji, closing the IDE first to satisfy the installPlugins requirement.
copy
$ PLUGIN_ID="dev.aikido.aikido-intellij-plugin"
osascript -e 'quit app "PhpStorm"'
sleep 5
open -na "PhpStorm.app" --args installPlugins "$PLUGIN_ID"
Pre-install the Aikido extension across VS Code, Cursor, and Windsurf on macOS developer machines via a Jamf or Kandji custom script.
copy
$ code --install-extension AikidoSecurity.aikido
cursor --install-extension AikidoSecurity.aikido
windsurf --install-extension AikidoSecurity.aikido
Deploy the Aikido JetBrains plugin (ID dev.aikido.aikido-intellij-plugin) via MDM using the installPlugins CLI flag, compatible with all JetBrains products (idea, webstorm, phpstorm, rider) and deployable as a Jamf/Kandji script on macOS or a PowerShell/Win32 package on Windows.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Deploy the Aikido VS Code extension (ID AikidoSecurity.aikido) via MDM using the --install-extension CLI flag across VS Code, Cursor, and Windsurf on macOS (Jamf/Kandji) and Windows (Intune PowerShell remediation). Not checked yet
2 details in Everything
safe-chain adds a secrets pre-commit pre-commit An open-source framework for managing and running git hooks, maintained by the pre-commit project, that lets cyber tools enforce checks automatically on every commit before code reaches a repository. hook installable globally or via the pre-commit framework, with per-line suppression via // gitleaks:allow and single-commit bypass via AIKIDO_SKIP_PRE_COMMIT.
Skip a specific secret detection inline by adding a // gitleaks:allow comment on the flagged line. Not checked yet Bypass the Aikido pre-commit hook for a single commit using the AIKIDO_SKIP_PRE_COMMIT environment variable. Not checked yet Bypass all pre-commit hooks for a single commit using git commit --no-verify. Not checked yet New detection Add Aikido secrets scanning to the pre-commit framework by adding the aikido-local-scanner hook ID to .pre-commit-config.yaml. Not checked yet Install the aikido-local-scanner binary to ~/.local/bin/aikido-local-scanner globally for all git repositories on macOS/Linux via a curl script, or on Windows via a PowerShell iwr script. Not checked yet
+1 more
show less
Install or remove the Aikido pre-commit hook with one click via the Aikido IDE plugin's Expansion Packs. Not checked yet
6 details in Everything
Aikido adds a self-hosted Broker to scan internal applications unreachable from the internet, deployable via Docker or Helm︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Start the Broker on a host with access to your internal network, restricting it to RFC-1918 ranges and pointing it at the US Aikido region.
copy
$ docker run -d \
--name aikido-broker \
--network host \
-e CLIENT_SECRET="AIK_BROKER_XXX_YYY_ZZZZ" \
-e ALLOWED_INTERNAL_SUBNETS="192.168.0.0/16,10.0.0.0/8,172.16.0.0/12" \
-e BROKER_TARGET_URL="https://broker.us.aikidobroker.com" \
-e DNS_SERVERS="10.0.0.53" \
aikidosec/broker
Deploy the Broker to Kubernetes using Helm, supplying your client secret and namespace.
copy
$ helm repo add aikido https://aikidosec.github.io/helm-charts
helm repo update
helm install broker-client aikido/broker-client \
--set config.clientSecret="AIK_BROKER_XXX_YYY_ZZZZ" \
--namespace aikido
Run the built-in diagnostics check to surface DNS or connectivity problems without digging through raw logs.
copy
$ docker exec -it aikido-broker node /app/diagnostics.cjs
Run the Broker as a Docker container using the CLIENT_SECRET, ALLOWED_INTERNAL_SUBNETS, DNS_SERVERS, BROKER_TARGET_URL, HTTP_PROXY, HTTPS_PROXY, ALL_PROXY, NO_PROXY, NODE_EXTRA_CA_CERTS, MTLS_PEM_PATH, MTLS_CA_PATH, NODE_TLS_REJECT_UNAUTHORIZED, FORCE_WEBSOCKET, and FORCE_POLLING environment variables to control network access, TLS, mTLS mTLS Mutual TLS, an encrypted connection where both client and server present certificates to prove their identity. It helps services verify who they are communicating with. , DNS, proxy, and transport behaviour.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Deploy the Broker on Kubernetes 1.19+ using the official Helm chart at https://aikidosec.github.io/helm-charts (chart name aikido/broker-client), configuring it via config.clientSecret or a values.yaml file. Not checked yet Set BROKER_TARGET_URL to route the Broker to your Aikido region: https://broker.aikidobroker.com (EU, default), https://broker.us.aikidobroker.com (US), https://broker.me.aikidobroker.com (ME), or https://broker.aikidogov.us (Aikido for Government). Not checked yet Mount a private CA file into the broker container and point to it with NODE_EXTRA_CA_CERTS to validate TLS on internal services that use self-signed or private-PKI PKI A framework of policies, roles, and cryptographic standards for issuing and managing digital certificates that bind public keys to identities, letting tools verify authenticity and encrypt communications without pre-shared secrets. certificates. Not checked yet Mount mTLS client credentials into the broker container and reference them with MTLS_PEM_PATH (required) and MTLS_CA_PATH (required only when the mTLS certificate is self-signed). Not checked yet
+3 more
show less
Run docker exec -it aikido-broker node /app/diagnostics.cjs (or kubectl exec <pod> -c broker-client -- node /app/diagnostics.cjs) to run a built-in health check covering DNS, connectivity, and other common failure points. Not checked yet Define allowed CIDR CIDR A notation for a block of IP addresses, such as 10.0.0.0/24. The number after the slash says how many leading address bits identify the network. ranges in ALLOWED_INTERNAL_SUBNETS to restrict which internal networks the Broker can call, preventing accidental access to unrelated infrastructure. Not checked yet Use the Broker to extend Aikido scans — code scanning on on-prem GitLab, container scanning on private registries, AI pentesting, and front-end/API scans — to services inside your company network. Not checked yet
8 details in Everything
safe-chain adds an official Terraform/OpenTofu OpenTofu An open-source fork of Terraform, maintained by the Linux Foundation, that provisions and manages infrastructure as code; tools use it to automate cloud or on-premises environment setup without vendor lock-in. provider to manage repositories, AutoFix settings, and PR checks as code︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Bootstrap the Aikido provider for a US-region workspace using environment variables so credentials never appear in source control.
main.tf
copy
terraform {
required_providers {
aikido = {
source = "AikidoTerraform/aikido"
version = "~> 1.1"
}
}
}
provider "aikido" {
base_url = "https://app.us.aikido.dev/api"
}
# Set before running terraform init / plan / apply:
# export AIKIDO_CLIENT_ID=<your-client-id>
# export AIKIDO_CLIENT_SECRET=<your-client-secret>
Activate a production repository and enforce high-severity PR gating with inline comments on critical findings — keeping security policy in version control.
main.tf
copy
resource "aikido_repository" "app" {
id = "12345"
active = true
sensitivity = "sensitive"
connectivity = "connected"
labels = ["payments", "production"]
}
resource "aikido_repo_pr_checks_settings" "app" {
code_repo_id = "12345"
minimum_severity = "high"
fail_on_dependency_scan = true
fail_on_sast_scan = true
fail_on_iac_scan = true
fail_on_secrets_scan = true
fail_on_malware_scan = true
post_inline_comments_min_severity = "critical"
}
Roll out AutoFix for critical and high dependency findings across all repositories in the workspace.
main.tf
copy
resource "aikido_autofix_dependency_settings" "workspace" {
enabled = true
severity_filter = "critical_and_high_only"
repos_scope = "all"
use_aikido_library_for_major = true
}
Declare the Aikido provider in Terraform or OpenTofu using source AikidoTerraform/aikido (version ~> 1.1), authenticating via client_id and client_secret in the provider block or via AIKIDO_CLIENT_ID and AIKIDO_CLIENT_SECRET environment variables.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Target a non-EU Aikido region by setting base_url (or AIKIDO_BASE_URL) to https://app.us.aikido.dev/api, https://app.au.aikido.dev/api, or https://app.me.aikido.dev/api in the provider block; the default is https://app.aikido.dev/api. Not checked yet Manage code repository activation, sensitivity, connectivity, and labels with the aikido_repository Terraform resource. Not checked yet Configure workspace-level AutoFix for dependencies, SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. , and Pentest findings with the aikido_autofix_dependency_settings resource, including severity_filter, repos_scope, and use_aikido_library_for_major fields. Not checked yet Gate pull requests per repository with the aikido_repo_pr_checks_settings resource, controlling minimum_severity, per-scan-type fail flags (fail_on_dependency_scan, fail_on_sast_scan, fail_on_iac_scan, fail_on_secrets_scan, fail_on_malware_scan, fail_on_code_quality_scan), inline comment thresholds, and Deep Review. Not checked yet
+1 more
show less
Manage teams, user permissions, and mappings of repositories and containers to teams through the provider. Not checked yet
6 details in Everything
Aikido adds webhook signature verification with HMAC-SHA256 HMAC-SHA256 A keyed hash that proves a message came from someone holding a shared secret and was not altered. Webhooks and API request signing commonly use it. and replay-attack protection via a 30-second timestamp window︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Verify incoming webhooks are genuine by comparing the HMAC HMAC A keyed hash that proves a message came from someone holding a shared secret and was not altered. Webhooks and API request signing commonly use it. -SHA256 SHA256 A cryptographic hash function that turns data into a fixed 256-bit fingerprint. It is used to check integrity; hashing does not encrypt the original data. digest of the request body against the value in the X-Aikido-Webhook-Signature request header.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Protect against replay attacks by rejecting webhooks where the dispatched_at epoch timestamp in the payload is older than 30 seconds. Not checked yet Generate a shared webhook secret from the webhooks integration page by clicking Add secret — the secret is shown once in a modal for secure copying. Not checked yet
3 details in Everything
Aikido adds a Code Context feature to teach the scanner your team's coding conventions and acceptable exceptions, reducing noise in pull request reviews︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Adds Code Context, a way to define team-specific coding conventions, architectural decisions, and acceptable exceptions so Code Quality comments on pull requests are relevant rather than generic.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Scopes Code Context rules to either all repositories in an organization or to selected repositories. Not checked yet Supports adding Code Context directly from a pull request by commenting to @AikidoSec, which Aikido processes automatically to improve future reviews. Not checked yet Commit AGENTS.md at the repository root to apply Code Context guidance to every review for that repository. Not checked yet
4 details in Everything
Aikido adds Custom Code Rules so teams can define AI-powered checks for org-specific patterns and auto-fix violations in pull requests︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Aikido adds Custom Code Rules, letting teams write plain-language descriptions of patterns to detect — such as enforcing Alpine base images in Dockerfiles — and have AI generate, validate, and scan pull requests against those rules.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Custom Code Rules now support an optional AutoFix prompt: write a correction instruction (e.g. Replace the base image with alpine ), validate it against non-compliant examples, and Aikido will offer automated fixes for matching findings when AutoFix is enabled for the workspace. Not checked yet Custom rules are configured under Code Quality > Checks > Add Custom Code Rule , with per-rule fields for Title, TL;DR, and How to fix guidance shown to developers on each finding. Not checked yet
3 details in Everything
safe-chain adds Azure DevOps PR gating via a Microsoft Entra service principal, replacing personal access tokens personal access tokens A personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. with short-lived federated credentials︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Grant the Entra service principal permission to create and delete service hooks in a specific Azure DevOps project without adding it to any administrator group.
copy
$ az devops permission update \
--org https://dev.azure.com/<your-organization> \
--namespace-id cb594ebe-87dd-4fc9-ac2c-6a10a4c92046 \
--subject <application-descriptor> \
--token PublisherSecurity/<project-id> \
--allow-bit \
-o table
Look up the service principal descriptor (prefix aadsp.) needed for the permission grant command above.
copy
$ az devops invoke \
--org https://dev.azure.com/<your-organization> \
--area graph \
--resource serviceprincipals \
--api-version 7.1-preview \
--query "value[?displayName=='Aikido PR Checks'].descriptor" \
-o tsv
Gate Azure DevOps pull requests using a Microsoft Entra service principal instead of a personal access token personal access token A personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. (PAT PAT A personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. ), so no token or secret is stored and nothing needs rotating.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Use az devops permission update with namespace ID cb594ebe-87dd-4fc9-ac2c-6a10a4c92046 and --allow-bit to grant the service principal ViewSubscriptions and EditSubscriptions rights without adding it to administrator groups. Not checked yet Configure the integration through the 'Manage Azure DevOps PR Checks service principal' page, supplying the Directory (tenant) ID and Application (client) ID from Entra to let Aikido verify repository access and service-hook management before saving. Not checked yet Switch back to a PAT at any time via Use a personal access token instead on the PR Checks tab, preserving existing repository configurations. Not checked yet Supports Azure Repos Git repositories only; Team Foundation Version Control (TFVC) repositories are not supported for PR gating or Code Quality gates. Not checked yet
5 details in Everything
safe-chain adds OpenShift OpenShift Red Hat's Kubernetes distribution that adds developer workflows, built-in CI/CD, and stricter security defaults on top of upstream Kubernetes, giving cyber tools a managed container platform with enterprise auth and RBAC already wired in. support, FIPS FIPS A set of US federal standards published by NIST that define approved cryptographic algorithms and modules; cyber tools use it to meet compliance requirements in government and regulated environments. -compliant agent images, and runtime threat detection for connected Kubernetes clusters︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Enable runtime threat detection on a cluster that is already connected, without rerunning full onboarding.
copy
$ helm upgrade aikido -n aikido \
--set runtimeDetection.enabled=true
Enable runtime threat detection on a connected cluster by toggling threat detection in the cluster edit settings and running a Helm upgrade with runtimeDetection.enabled=true.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Use FIPS-compliant agent images for FedRAMP FedRAMP A US government program that standardizes cloud security authorization, letting vendors get a single assessment accepted across federal agencies instead of repeating the process for each one. environments by appending -fips to repo values in the Helm chart; images are published at public.ecr.aws/aikido-cloud/kubernetes-agent-fips and public.ecr.aws/aikido-cloud/kubernetes-sbom-collector-fips. Not checked yet safe-chain now supports OpenShift clusters (on-prem and managed), with automatic resolution of image mirrors from ImageContentSourcePolicies, ImageDigestMirrorSets, and ImageTagMirrorSets so images are correlated against their original source, not the mirrored reference. Not checked yet Mount custom CA certificates in OpenShift environments using extra volumes and pointing the agent to the trust bundle via SSL_CERT_FILE or SSL_CERT_DIR environment variables. Not checked yet Authenticate to registries in restricted OpenShift setups by mounting image pull secrets directly instead of relying on RBAC RBAC Role-based access control, where permissions are assigned to roles such as reader or administrator and users receive those roles. It makes access easier to manage than individual permission lists. -based image access. Not checked yet
+2 more
show less
Exclude specific Kubernetes namespaces from resource collection at onboarding time, with quick-add options for common system namespaces such as kube-system and kube-public. Not checked yet Toggle in-cluster image scanning on an already-connected cluster and apply the change with helm upgrade aikido -n aikido --set sbomCollector.enabled=true (or false). Not checked yet
7 details in Everything
└──▷ BREAKING ON UPGRADE
!
Changing the in-cluster image scanning setting on an existing cluster does NOT automatically update the Helm release — you must manually enable or disable `sbomCollector.enabled` in the Kubernetes agent Helm release, or the toggle and the deployed agent will be out of sync.
Our best guess
safe-chain adds a Secret Keyword Filter so workspace admins can skip false-positive lines before gitleaks evaluates them︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ HOW TO FIND IT
Suppress a recurring false positive tied to a stable config key across all repos without touching individual source files.
📍 Settings › Advanced › Secret Keyword filter › Manage
1. Click Manage
2. Enter keyword: EXAMPLE_SENDGRID_KEY (minimum 5 chars, case-sensitive)
3. Click Save
4. Trigger a new scan — any line containing EXAMPLE_SENDGRID_KEY is skipped by gitleaks entirely
Enable the Secret Keyword Filter workspace-wide by navigating to Settings > Advanced > Secret Keyword filter and clicking Manage, then adding keywords (minimum 5 characters, case-sensitive, plain string matching) to skip entire lines from secrets scanning before gitleaks runs.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Use .aikido file exclusions to exclude specific files or paths from scanning as an alternative suppression method alongside the Secret Keyword Filter. Not checked yet
2 details in Everything
safe-chain adds .aikido config file support to exclude paths, ignore CVEs, and tune AutoFix behavior per repository︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Exclude test fixtures and generated code directories from all Aikido scans — including secrets and SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. — without touching global settings.
.aikido
copy
exclude:
paths:
- test/fixtures
- src/generated
- vendor/
Suppress a known, accepted CVE for the repository so the Aikido UI shows it as intentionally ignored rather than unreviewed.
.aikido
copy
CVE-2023-12345:
reason: We do not care about this CVE
CVE-2021-99999:
reason: We handle this
Exclude specific paths and files from all code scanning (secrets, SAST, lockfiles, code quality, and Deep Review) by listing them under the exclude.paths key in a .aikido file at the repository root.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Ignore specific CVEs per repository by adding them under a CVE key in .aikido with a reason field; the Aikido UI will surface that the CVE is ignored with a reference to the .aikido file. Not checked yet Configure repository-specific AutoFix behavior — such as a custom branch prefix or a single AutoFix PR for the whole repository — via the autofix key in .aikido. Not checked yet
3 details in Everything
safe-chain adds pip private package registry and private Git repository configuration for autofix dependency updates︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ HOW TO FIND IT
Connect a private GitHub-hosted pip repository in a different org so the autofix bot can resolve and update those dependencies.
📍 In Aikido autofixer settings, click Connect Registry › select Environment Variables › set key: PIP_GITHUB__COM, value: <your-github-access-token> › click Connect Registry .
Connect a private PyPI registry for pip autofix updates by supplying an index URL with credentials via the Connect Registry modal under autofixer settings.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Authenticate autofix access to private pip Git repositories in a different SCM SCM Source Code Management: version-control systems (Git, SVN, etc.) that track changes to code over time, giving cyber tools reproducible builds, audit trails, and collaborative development workflows. organization by setting environment variables — PIP_GITHUB__COM, PIP_GITLAB__COM, PIP_BITBUCKET__ORG, or PIP_DEV__AZURE__COM — with an access token access token A credential that authorizes a client to use a service or API. It usually has limited permissions and a lifetime; anyone holding it may be able to use those permissions. via the Environment Variables option in the Connect Registry modal. Not checked yet
2 details in Everything
Aikido adds zero-CVE drop-in base image replacements via Aikido Images, hosted on docker.aikido.io, with AutoFix PR creation for containers︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Manually swap a vulnerable base image for its Aikido Images equivalent in a Dockerfile, using the read-only token from the Aikido UI.
Dockerfile
copy
# In your Dockerfile, replace the existing base image line, e.g.:
# FROM debian:bookworm
# with the Aikido Images path shown under Show Path:
FROM docker.aikido.io/<token>/debian:bookworm
Browse available zero-CVE base image replacements at AutoFix > Cloud > Containers > Aikido Images in the Aikido UI, covering both amd64 and arm64 architectures.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet When using the Aikido Local Scanner without source code access, the UI now shows the suggested Aikido Image and a Show Path button that reveals the full docker.aikido.io pull path with a read-only account token so you can apply the change manually. Not checked yet
2 details in Everything
safe-chain adds per-repo .aikido config file with branch templates, conditional auto-merge by type and severity, instruction-based auto-merge, and task tracker integration︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Enforce task-linked branch names, restrict updates to safe versions, and auto-merge only low-severity dependency PRs — all scoped to one repo without touching global UI settings.
.aikido
copy
autofix:
pr_config:
branch_prefix: "fix/"
title_prefix: "Aikido "
commit_prefix: "fix: "
labels: "aikido,$AUTOFIX_TYPE,$SEVERITY"
draft_pr: false
auto_merge:
dependency:
enabled: true
severity: low
create_task_for_pr: true
task_title_prefix: "[AIK] "
dependency:
update_policy: minor_and_patch_versions_only
base_branch: main
Group monorepo lockfiles so each service gets one consolidated AutoFix PR, while suppressing PRs entirely for a legacy service that is not actively maintained.
.aikido
copy
autofix:
dependency:
lockfile_groups:
- group_name: service1
lockfiles:
- 'service1/*'
- group_name: legacy
disabled: true
lockfiles:
- 'legacy/*'
Use natural language to auto-merge only high-confidence SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. fixes on non-critical paths, giving the AI context about diffs and changed files.
.aikido
copy
autofix:
pr_config:
auto_merge:
instructions: >
Auto-merge only if AutoFix type is sast, confidence is high,
and no changed file is under the src/payments directory.
Configure AutoFix PR behaviour per repository with a .aikido file at the repo root, overriding global Aikido UI settings for keys including branch_prefix, title_prefix, commit_prefix, labels, draft_pr, auto_merge, create_task_for_pr, task_title_prefix, disable_codeowners_as_reviewers, self_approve_pr, and automated.disabled.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet (new) Define full branch, PR title, and commit message formats in .aikido using branch_template (with $TASK_REF, $TASK_REF_LOWER, $TASK_REF_0, and $TASK_REF_0_LOWER variables for task-linked branch naming), title_template (with $MSG placeholder, taking precedence over title_prefix), and commit_template (with $MSG placeholder, taking precedence over commit_prefix); also use dynamic $AUTOFIX_TYPE (dependency, sast, iac, quality, container, pentest, audit, gradle, dotnet, zen) and $SEVERITY (critical, high, medium, low) variables in labels, branch_prefix, title_prefix, and commit_prefix fields. Not checked yet (new) Enable auto-merge controls in .aikido under autofix.pr_config.auto_merge, either conditionally by AutoFix type (dependency, sast, iac, container, code-quality, pentest, code-audit) and severity (critical, high, medium, low), or via natural-language autofix.pr_config.auto_merge.instructions that account for AutoFix type, severity, confidence, PR title and description, changed files, and diffs. Not checked yet (new) Add further per-repo .aikido settings: disable_codeowners_as_reviewers: true under autofix.pr_config to stop CODEOWNERS being added as reviewers on GitLab Merge Requests, self_approve_pr: true under autofix.pr_config to let Aikido AutoFix self-approve its own BitBucket PRs, create_task_for_pr: true with task_title_prefix under autofix.pr_config to automatically create a task in a connected issue tracker when an AutoFix PR is opened (supporting compliance remediation traceability), and autofix.automated.disabled: true to opt a repository out of globally enabled automated AutoFixes. Not checked yet (new) Add autofix.dependency settings in .aikido: lockfile_groups to group lockfiles into a single AutoFix PR per group (with per-group group_name, lockfiles glob patterns, and optional disabled: true to suppress PRs for a group), and base_branch to set the base branch for AutoFix PRs per repo. Not checked yet
5 details in Everything
Aikido adds custom context instructions for AutoFix SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. rules, configurable globally or per repository via the .aikido file︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Instruct AutoFix to always add structured logging and unit tests when remediating SAST findings across an entire repository.
.aikido
copy
# .aikido (repo root)
- path: aikido-autofix-all.md
Add custom instructions to SAST AutoFix rules — such as error handling logic, comment styles, or unit tests — by placing a markdown file reference under path: aikido-autofix-all.md in the .aikido config file at the root of a repository.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Scope custom AutoFix context to a specific programming language using a language key in the .aikido file, with supported values including python, go, javascript, typescript, java, rust, terraform, and 17 others. Not checked yet
2 details in Everything
Aikido MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. adds token-based authentication via AIKIDO_API_KEY for headless and CI/CD environments across Claude Code, Cursor, JetBrains, Copilot, and more︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Run Aikido MCP in a CI/CD pipeline or Docker container where no browser is available.
copy
$ export AIKIDO_API_KEY="your-token-here"
claude
Register Aikido MCP with OpenAI Codex CLI in a headless environment using a PAT PAT A personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. .
copy
$ codex mcp add \
--env AIKIDO_API_KEY=your-token-here \
-- npx -y @aikidosec/mcp@latest
Configure Aikido MCP token authentication for OpenCode so every session authenticates without a browser prompt.
~/.config/opencode/opencode.json
copy
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"aikido": {
"type": "local",
"command": ["npx", "-y", "@aikidosec/mcp@latest"],
"enabled": true,
"environment": {
"AIKIDO_API_KEY": "your-token-here"
}
}
}
}
Aikido adds an MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. plugin for GitHub Copilot that scans AI-generated code for vulnerabilities and hardcoded secrets in real time︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ USE IT
Wire Aikido MCP into GitHub Copilot IDE so every AI-generated snippet is scanned before you commit.
mcp.json
copy
{
"servers": {
"aikido": {
"command": "npx",
"args": ["-y", "@aikidosec/mcp@latest"]
}
}
}
Connect Aikido MCP to GitHub Copilot Cloud Agents for headless scanning in CI, using a repository secret for authentication.
json
copy
{
"servers": {
"aikido": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@aikidosec/mcp@latest"],
"tools": ["aikido_scan_paths", "aikido_full_scan"],
"env": {
"AIKIDO_API_KEY": "COPILOT_MCP_AIKIDO_API_KEY"
}
}
}
}
Add Aikido's MCP server to GitHub Copilot IDE settings via mcp.json, using the @aikidosec/mcp@latest package over npx, to scan AI-generated snippets before they are committed.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Authenticate the MCP server in Copilot IDE by running the aikido_login tool, which returns a browser link and stores your token on completion. Not checked yet Restrict which tools Copilot Cloud Agents can call by listing them explicitly under tools in the MCP JSON config, for example aikido_scan_paths and aikido_full_scan. Not checked yet Install Aikido MCP as a VS Code Expansion Pack within the Aikido VSCode IDE plugin, eliminating a separate setup process for Copilot IDE users. Not checked yet Add scanning instructions to AGENTS.md or Copilot custom instructions to direct the agent to invoke Aikido MCP tools automatically. Not checked yet
5 details in Everything
safe-chain adds an Aikido MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server with tools for scanning, deep AI review of git changes, issue feed filtering, lifecycle management, and severity adjustment.
New detection Scan files on disk for SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. findings, hardcoded secrets, and IaC IaC Infrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. misconfigurations by calling aikido_scan_paths with paths, root, and optional repository_name; paths outside root are skipped and listed in skipped_paths. Not checked yet Scan in-memory or generated code not yet saved to disk by calling aikido_full_scan with each file supplied as a relativeFilePath and content pair in files_to_scan. Not checked yet Start an AI-powered deep review of local git changes with aikido_deep_review, setting directory to the repository path; optionally scope to committed, uncommitted, or all changes via scope, diff against a custom ref with base_ref, and include untracked files with include_untracked. Not checked yet Poll the result of a running deep review by calling aikido_deep_review_status with the agent_id returned by aikido_deep_review; the tool returns completed, skipped, failed, or running and waits up to 50 seconds per call. Not checked yet Fetch security issues from your Aikido feed with aikido_issues_list, filtering by scope, issue type, severity, labels, repo_branch_name, SLA SLA A formal agreement between a service provider and customer that defines expected uptime, response times, and remedies for failures, giving cyber tools a contractual baseline for availability and support commitments. filters (out_of_sla or sla_due_soon), and zero-based page. Not checked yet
+4 more
show less
Manage issue lifecycle with aikido_ignore_issue, aikido_unignore_issue, aikido_snooze_issue (supporting duration or ISO 8601 ISO 8601 An international standard for representing dates and times as text, using formats like 2025-04-23T14:00:00Z. Tools adopt it to ensure timestamps are unambiguous and sortable across systems and locales. snooze_until_date), and aikido_unsnooze_issue. Not checked yet Adjust the severity of any issue with aikido_adjust_issue_severity, supplying issue_id, adjusted_severity, and a reason. Not checked yet Authenticate or switch accounts with aikido_login; optionally force re-authentication with force_reauth or override the auto-detected region with region (eu, us, me, or au). Not checked yet Pin default scan behaviour for every agent session by adding plain-language scan instructions to your agent's rules file (CLAUDE.md, .cursor/rules, AGENTS.md, etc.). Not checked yet
9 details in Everything
safe-chain adds MDM MDM Mobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. -based mass deployment of the Aikido IDE plugin for JetBrains and VS Code-family editors︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
└──▷ TRY IT
Pre-install the Aikido JetBrains plugin on all developer Macs via a Jamf or Kandji policy script, without requiring any manual action from developers.
copy
$ PLUGIN_ID="dev.aikido.aikido-intellij-plugin"
osascript -e 'quit app "PhpStorm"'
sleep 3
open -na "PhpStorm.app" --args installPlugins "$PLUGIN_ID"
Pre-install the Aikido extension across all VS Code, Cursor, and Windsurf installations on developer machines via a Jamf or Kandji script.
copy
$ code --install-extension AikidoSecurity.aikido
cursor --install-extension AikidoSecurity.aikido
windsurf --install-extension AikidoSecurity.aikido
Deploy the Aikido JetBrains plugin at scale using the installPlugins flag with plugin ID dev.aikido.aikido-intellij-plugin, via a Jamf policy script, Kandji custom script, or Windows PowerShell/Win32 package.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Deploy the Aikido VS Code extension at scale using --install-extension AikidoSecurity.aikido with code, cursor, or windsurf, via Jamf, Kandji, or Intune PowerShell remediation scripts. Not checked yet
2 details in Everything
Aikido's VS Code plugin adds support for Windsurf, Cursor, Kiro, and Google Antigravity IDEs alongside real-time SAST SAST Static application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. , secrets, IaC IaC Infrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. , and SCA SCA Software composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. scanning︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Run the Aikido: Log In command from the Command Palette to authenticate with the Aikido platform when the automated flow does not work.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Not checked yet Aikido's VS Code plugin now scans for hardcoded secrets (API keys, tokens) and insecure code patterns (SQL injection SQL injection An attack that makes an application treat untrusted input as part of a database command. It can expose or modify data by changing what the query does. , path traversal) automatically on file open or save, with issues highlighted inline and listed in the Aikido window. Not checked yet Aikido's VS Code plugin now supports Windsurf, Cursor, Kiro, and Google Antigravity IDEs in addition to VS Code, available from each editor's marketplace. Not checked yet Extends the plugin with Expansion Packs including MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. for AI agents, pre-commit pre-commit An open-source framework for managing and running git hooks, maintained by the pre-commit project, that lets cyber tools enforce checks automatically on every commit before code reaches a repository. hooks, and Safe Chain for additional scanning surfaces. Not checked yet Adds open-source dependency scanning (SCA) directly in the IDE alongside real-time SAST, secrets, and IaC scanning. Not checked yet
+1 more
show less
Adds a full workspace scan capability and an Aikido AI assistant inside the IDE. Not checked yet
6 details in Everything
safe-chain adds a secrets pre-commit pre-commit An open-source framework for managing and running git hooks, maintained by the pre-commit project, that lets cyber tools enforce checks automatically on every commit before code reaches a repository. hook for macOS, Linux, and Windows with inline suppression via gitleaks:allow and per-commit bypass via AIKIDO_SKIP_PRE_COMMIT.
New detection Install the secrets pre-commit hook globally on macOS/Linux by piping the install script through bash, which places aikido-local-scanner at ~/.local/bin/aikido-local-scanner. Not checked yet Install the secrets pre-commit hook globally on Windows by running the PowerShell install script via iwr and iex. Not checked yet Integrate with the pre-commit framework by adding the aikido-local-scanner hook ID to .pre-commit-config.yaml under the https://github.com/AikidoSec/pre-commit repo. Not checked yet Skip a specific secret inline by appending // gitleaks:allow as a comment on the flagged line. Not checked yet Bypass the Aikido pre-commit hook for a single commit by setting the AIKIDO_SKIP_PRE_COMMIT environment variable. Not checked yet
+2 more
show less
Bypass all pre-commit hooks for a single commit using git commit --no-verify. Not checked yet Uninstall the global hook on macOS/Linux or Windows using the provided uninstall scripts, or manually by removing ~/.git-hooks, running git config --global --unset core.hooksPath, and deleting ~/.local/bin/aikido-local-scanner. Not checked yet
7 details in Everything
Adds .aikido file for project-level configuration in safe-chain
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.5.15 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.5.15
Allows a project level config via .aikido file Not checked yet Add install script shasum verification in readme.md examples Not checked yet
2 details in Everything
Adds -is-mdm flag to the Windows endpoint installer.
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.5.8 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.5.8
└──▷ TRY IT
Install MDM MDM Mobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. -specific components during the silent setup of safe-chain on Windows.
copy
$ safe-chain install -is-mdm
1 detail in Everything
safe-chain v1.5.4 adds PDM package manager support and file-based logging
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.5.4 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.5.4
PDM added as a supported Python package manager alongside pip and poetry Not checked yet Adds file-based logging capability to persist logs to disk Not checked yet
2 details in Everything
safe-chain 1.5.3 adds Rush monorepo support via the new rush integration
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.5.3 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.5.3
└──▷ TRY IT
Run safe-chain checks within a Rush monorepo workspace using the new native integration.
copy
$ npx safe-chain rush
Adds rush subcommand and configuration to enable safe-chain in Rush-managed monorepos Not checked yet
1 detail in Everything
safe-chain 1.5.0 adds uvx package manager support and custom install directory configuration
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.5.0 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.5.0
└──▷ TRY IT
Configure safe-chain to install files into a specific location rather than the system default.
copy
$ safe-chain --install-dir /custom/path/to/install
[Python Improvements]: Add uvx support Not checked yet
1 detail in Everything
Adds manual setup/teardown on failure instructions and minimum package age enforcement for npm and PyPI sources.
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.4.6 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.4.6
└──▷ USE IT
Configure safe-chain to enforce a minimum age for npm packages in yarn lockfiles during dependency scanning.
shell
copy
safe-chain scan --npm-min-age=7d
Configure safe-chain to enforce a minimum age for PyPI packages during dependency scanning.
shell
copy
safe-chain scan --pypi-min-age=7d
Enforce minimum package age for direct npm downloads (e.g. yarn lockfiles) Not checked yet Add minimum package age checks for pypi Not checked yet
2 details in Everything
safe-chain 1.4.3 adds Windows install script support and fixes PowerShell execution policy checks for shell integration.
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.4.3 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.4.3
Supports Windows platforms via the install-safe-shain.sh script in environments like git bash or cygwin Not checked yet Adds pre-install execution policy verification in PowerShell to ensure safe-chain can run on Windows Not checked yet Propagates command-not-found errors from wrapped commands within the shell integration Not checked yet
3 details in Everything
safe-chain v1.4.2 adds min package age exclusion and configurable log level
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.4.2 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.4.2
SAFE_CHAIN_LOG_LEVEL env var allows configuring log level Not checked yet
1 detail in Everything
Adds verify command, Jenkins CI integration docs, and python custom registries configuration
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.3.4 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.3.4
Adds safe-chain verify command to check if safe-chain is actively intercepting package manager commands Not checked yet Allows configuring python custom registries through environment variable SAFE_CHAIN_PYTHON_REGISTRIES Not checked yet
2 details in Everything
safe-chain 1.3.3 adds Circle CI and PIPX support plus custom npm registry configuration
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.3.3 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.3.3
└──▷ TRY IT
Run safe-chain in a Circle CI pipeline by setting the provider environment variable.
copy
$ export CI_PROVIDER=circleci
safe-chain scan
Install safe-chain globally using PIPX for isolated, system-wide access without sudo.
copy
$ pipx install safe-chain
Configure safe-chain to resolve npm dependencies from a private registry by setting the NPM registry environment variable.
copy
$ export NPM_CONFIG_REGISTRY=https://private-registry.example.com/
safe-chain scan --source ./my-node-app
Adds circleci to the list of supported continuous integration platforms via new CI_PROVIDER=circleci environment variable Not checked yet Enables installation via PIPX using standard Python package indexing Not checked yet Supports configuring custom or private npm registries for dependency resolution Not checked yet
3 details in Everything
safe-chain 1.3.2 provides an install script artifact with version-skipping logic for improved setup security and idempotency.
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.3.2 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.3.2
The installer is now distributed as a release artifact to enable secure version pinning Not checked yet Added pre-flight check to skip installation if safe-chain is already present and the version matches Not checked yet
2 details in Everything
safe-chain v1.2.3 adds Poetry support and configurable package age setting via 0
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.2.3 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.2.3
└──▷ TRY IT
Analyze a project managed by Poetry without filtering packages by age
copy
$ safe-chain python --minimum-package-age 0 .
Remove safe-chain from the system using the new uninstall scripts
copy
$ ./scripts/uninstall.sh
Adds Poetry dependency management support to safe-chain python for lockfile-based analysis Not checked yet Allows minimum_package_age setting to accept 0, disabling the age filter Not checked yet Adds system-wide uninstall scripts to clean up safe-chain shims and related files Not checked yet
3 details in Everything
safe-chain v1.1.8 adds pip wrapping, malware list scanning, 24h package age suppression, and verbose logging
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.1.8 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.1.8
└──▷ USE IT
Enable Python package scanning in your environment setup.
shell
copy
safe-chain setup --include-python
Install a package immediately despite the 24-hour malware validation wait.
shell
copy
pip install <package-name> --safe-chain-skip-minimum-package-age
safe-chain setup --include-python enables pip/pip3 interception and PyPI malware list scanning Not checked yet New --safe-chain-skip-minimum-package-age flag to bypass the 24-hour validation delay for specific installs Not checked yet --safe-chain-logging=verbose option adds verbose output capability Not checked yet
3 details in Everything
safe-chain v1.1.6 adds silent mode for log suppression
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.1.6 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.1.6
└──▷ USE IT
Suppress noisy logging during unattended CI/CD runs
shell
copy
safe-chain --silent scan --target ./my-app
Adds --silent flag to disable all logging output Not checked yet
1 detail in Everything
└──▷ BREAKING ON UPGRADE
! The --safe-chain-malware-action flag has been removed; usage will result in an error.
Adds HTTP proxy mode, package install timeout reduction, and a version subcommand
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.1.5 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.1.5
└──▷ USE IT
Check which version of the proxy is currently installed in CI logs or debug output.
shell
copy
safe-chain --version
Adds --version flag (shorthand -v) to display the safe-chain version. Not checked yet Enables safe-chain to act as a regular HTTP proxy, allowing it to forward requests to local registries instead of only intercepting them. Not checked yet Reduces timeout limits during package installation to improve scanner performance on npm >= 10.4.0. Not checked yet
3 details in Everything
Safe Chain 1.1.0 adds a proxy-based blocker for all package managers and Bun support
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.1.0 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.1.0
Adds a lightweight proxy server that intercepts and blocks downloads of packages containing malware, enabling full dependency tree protection for all package managers. Not checked yet Adds bun install and bunx command protection via the same malware detection engine used for other package managers. Not checked yet
2 details in Everything
safe-chain adds setup-ci subcommand to wrap npm commands in CI/CD pipelines
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.0.24 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.0.24
Adds the safe-chain setup-ci command to configure GitHub Actions GitHub Actions GitHub's built-in CI/CD system that runs automated workflows triggered by repository events, letting cyber tools automate testing, building, and releasing without managing separate pipeline infrastructure. and Azure Pipelines for wrapping npm commands. Not checked yet Changes default behavior to block package installation and exit with code 1 when scanning fails (previously it ran the command). Not checked yet
2 details in Everything
└──▷ BREAKING ON UPGRADE
! safe-chain now exits with code 1 and blocks package installation when scanning fails, instead of continuing execution.
safe-chain v1.0.21 changes malware detection default from prompt to block
└──▷ GET THIS VERSION
copy
$ git clone --branch 1.0.21 https://github.com/AikidoSec/safe-chain.git
$ git checkout 1.0.21
└──▷ USE IT
Override the default block behavior to allow installation even if malware is detected.
shell
copy
safe-chain install --safe-chain-malware-action=prompt
Malware detection now blocks installation by default instead of prompting, enforcing security by default. Not checked yet Adds --safe-chain-malware-action=prompt flag to override the new blocking default behavior. Not checked yet
2 details in Everything
└──▷ BREAKING ON UPGRADE
! The default action when malware is detected has changed from prompt (allowing bypass by typing Y ) to block, which will cause safe-chain to exit with an error on malicious packages.