Works with the AI Agents you already use
Cloud Storage Exposure Risk Investigator
Investigate cloud storage policies to identify unintended public exposure, cross-account risks, and access gaps.
$29.99
Cloud Storage Exposure Risk Investigator
Example session with this skill installed
Review this S3 bucket policy for a bucket named 'prod-backups-789'. It contains database snapshots. Principal is a wildcard but there is a Condition for a specific VPC.
- Read your context and instructions
- Compiled the cloud storage exposure
I have reviewed the provided S3 bucket policy for prod-backups-789. While the policy uses a wildcard principal, the VPC condition provides significant network-level isolation, though several configuration gaps remain.
Exposure Risk Investigation: prod-backups-789
| Finding ID | Severity | Confidence | Status | Summary |
|---|---|---|---|---|
| EXP-001 | Medium | High | Observed | Conditional Public Access: Wildcard principal allows access from any identity within the specified VPC. |
| EXP-002 | Low | Medium | Potential | Missing Public Access Block: Account-level 'Block Public Access' status is unverified. |
| EXP-003 | Informational | High | Observed | Encryption Enforcement: Policy correctly denies non-encrypted uploads (s3:x-amz-server-side-encryption). |
Finding Details: EXP-001
- Access Path: Resource-based policy.
- Principal:
*(Wildcard). - Condition:
StringEqualsonaws:SourceVpcforvpc-12345678.
Reasoning: While restricted to a VPC, any compromised instance or over-privileged user within that VPC can access these snapshots regardless of their specific IAM permissions. This is "network public" within your internal perimeter.
- Remediation: Replace the wildcard principal with specific IAM Roles used by the backup/restore services.
Positive Controls Observed
- VPC Restriction: Effective at preventing internet-wide anonymous access.
- Encryption Headers: Ensures all data at rest is encrypted via mandatory headers.
Coverage Gaps
- IAM Policies: Identity-based permissions for users in
vpc-12345678were not provided. - KMS Policy: Access to the encryption keys used for these snapshots was not reviewed.
Next steps
- Provide the IAM policies for the roles operating within
vpc-12345678to confirm the least-privilege model. - Verify if
BlockPublicAccessis enabled at the S3 Account level. - Update the bucket policy to restrict the
Principalto the specific Backup Service Role ARN.
Connects securely to your tools. The creator never sees your data.
About this skill
Cloud storage misconfigurations are a leading cause of data breaches. This skill provides a defensive framework for auditing storage policies and identifying unintended exposure without risking the data itself. It moves beyond simple "public bucket" checks to analyze the complex interaction between resource policies, IAM roles, and organizational guardrails.
What it does
- Analyze access paths by evaluating the intersection of bucket policies, ACLs, IAM roles, and organization-level public access blocks.
- Identify exposure patterns including anonymous read/write, unintended cross-account access, and overly broad workload roles.
- Prioritize data risks by classifying assets based on metadata and business purpose to determine the severity of a finding.
- Validate findings using policy reasoning and metadata-level evidence instead of invasive object enumeration.
- Generate remediation plans that provide specific, least-privilege policy adjustments and rollback guidance.
How it works
- Define scope by providing the cloud provider, service type, and redacted configuration files or policy exports.
- Classify assets based on your descriptions of the data types stored, from public assets to regulated PII.
- Map permissions as the skill cross-references identity-based and resource-based policies to find the effective access level.
- Review report containing prioritized findings, evidence-led reasoning, and safe remediation steps.
Frameworks & tools
The skill is designed for AWS S3, Azure Blob Storage, and Google Cloud Storage (GCS). It works with infrastructure-as-code snippets, IAM policy JSON, and sanitized cloud provider audit logs.
Why this beats prompting it yourself
Generic prompts often miss the nuances of policy evaluation, such as how an explicit 'Deny' at the organization level overrides a 'Allow' at the bucket level. This skill uses a structured investigation playbook to ensure cross-account trust and encryption key permissions are factored into the risk score.
Use cases
- Audit new infrastructure before deploying infrastructure-as-code to production environments.
- Investigate audit findings from automated scanners that lack business context or data classification.
- Review cross-account access to ensure third-party integrations haven't created persistent backdoors.
- Validate public-access blocks across entire projects or subscriptions to enforce hardening standards.
Known limitations
Does not perform live data exfiltration or active penetration testing. Requires the user to provide policy exports as it does not have direct read access to your cloud console.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Requires manual provision of redacted policy/config exports. Cannot directly query cloud APIs. Does not validate object-level ACLs unless explicitly provided in the evidence.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 10 days ago