- Home
- Skills
- DevOps & Cloud
- AI Code Review Security Drift Sanitizer
Works with the AI Agents you already use
AI Code Review Security Drift Sanitizer
Identify security vulnerabilities and configuration drift in code and infrastructure with automated secret redaction.
$2.99
AI Code Review Security Drift Sanitizer
Example session with this skill installed
Review this Kubernetes deployment manifest against our baseline policy of no root containers and mandatory resource limits. Redact any secrets found in the env section.
- Read your context and instructions
- Compiled the ai code review
I have reviewed the provided Kubernetes deployment manifest. One secret was detected and redacted in the environment variables section. The analysis identified two instances of security drift relative to your baseline policy.
Security Review & Drift Report
Executive Summary
The manifest contains configuration that deviates from the established safety baseline regarding container privileges and resource constraints. One high-severity finding requires immediate remediation to prevent potential container breakout.
Findings
| ID | Finding | Severity | Confidence | Location |
|---|---|---|---|---|
| SEC-001 | Privileged Container Execution | High | High | spec.template.spec.containers[0].securityContext |
| SEC-002 | Missing Resource Limits | Medium | High | spec.template.spec.containers[0].resources |
Security Drift Analysis
| Policy Requirement | Observed Value | Status | Risk |
|---|---|---|---|
runAsNonRoot: true | privileged: true | DRIFT | Allows potential host-level access if container is compromised. |
resources.limits.cpu | Not defined | DRIFT | Risk of node resource exhaustion (DoS). |
Redacted Environment Snippet
env:
- name: DB_PASSWORD
value: "[REDACTED]"
- name: API_KEY
value: "[REDACTED]"
Remediation Plan
Apply the following patch to align with the security baseline:
securityContext:
allowPrivilegeEscalation: false
runAsNonRoot: true
runAsUser: 1000
resources:
limits:
cpu: "500m"
memory: "512Mi"
requests:
cpu: "250m"
memory: "256Mi"
Next steps
- Apply the suggested
securityContextandresourcesblocks to the deployment manifest. - Verify that the application user (UID 1000) has appropriate permissions for required local paths.
- Trigger a fresh deployment and monitor for OOMKill events due to new resource limits.
Connects securely to your tools. The creator never sees your data.
About this skill
Manual security reviews are slow and often overlook configuration drift that opens production environments to attack. This skill identifies vulnerabilities, misconfigurations, and sensitive data exposure in code and infrastructure files. It delivers evidence-based findings with redacted secrets and safe remediation snippets to harden your application without breaking functionality.
What it does
- Security Drift Analysis compares active Kubernetes, Terraform, or cloud manifests against baselines to flag unauthorized changes.
- Secret Sanitization automatically detects and redacts credentials, tokens, and private keys from logs and source code before analysis.
- Defensive Code Review audits logic for injection risks, broken authentication, and improper resource handling.
- Remediation Engineering generates minimal, reversible patches and validation checklists for identified weaknesses.
- CI/CD Auditing inspects GitHub Actions and pipeline workflows for excessive permissions or untrusted trigger risks.
How it works
- Scope Definition Provide your source code, diffs, configuration files, or logs along with any security baselines.
- Automated Sanitization The agent scans for sensitive strings, replacing them with
[REDACTED]to ensure no secrets are echoed in the output. - Deep Inspection The agent runs a heuristic scan and manual review of logic, permissions, and encryption settings.
- Report Generation Receive a structured report with findings categorized by severity and confidence.
Frameworks & tools
Works with Kubernetes manifests, Terraform HCL, GitHub Actions workflows, .env files, and various application codebases. Includes a standalone Python-based heuristic scanner for rule-based identification.
Why this beats prompting it yourself
Generic prompts often leak secrets back into the chat or hallucinate line numbers and CVEs. This skill enforces a strict redaction protocol and uses a formal schema to ensure every finding is backed by specific evidence rather than probabilistic guesses.
Use cases
- Auditing a pull request for security regressions before merging to main.
- Identifying configuration drift in Kubernetes clusters after an incident.
- Sanitizing production logs for debugging without exposing session cookies or PII.
- Hardening CI/CD pipelines against supply-chain attacks.
Known limitations
Does not perform dynamic analysis or runtime exploit verification. Requires an explicit baseline to confirm historical configuration drift.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.
No runtime or dynamic analysis. Requires explicit baselines for drift detection. Cannot rotate credentials or modify remote infrastructure. Redaction is heuristic-based.
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 3 days ago