More screenshots
Works with the AI Agents you already use
secure vibe coding
Security-first engineering agent for auditing, hardening, and verifying web apps, APIs, and AI agent integrations.
$15
secure vibe coding
Example session with this skill installed
Run an AUDIT on the current repository. Focus on authentication and database access in our Next.js and Supabase stack. Do not modify any files yet.
- Read your context and instructions
- Compiled the secure vibe coding
- Generated the UI component
I have completed the AUDIT of the repository. I inspected the Next.js frontend, Supabase schema, and environment configurations. I did not perform any active network probing or file modifications.
Security Audit Report
| ID | Title | Severity | Confidence | Status | Evidence |
|---|---|---|---|---|---|
| SEC-001 | Missing Row Level Security (RLS) | Critical | Confirmed | Confirmed | supabase/migrations/2023_init.sql:12-45 |
| SEC-002 | Client-side Authorization Logic | High | Likely | Needs validation | src/app/admin/page.tsx:8 |
| SEC-003 | Exposed Service Role Key | Critical | Confirmed | Confirmed | .env.example:4 |
| SEC-004 | Insecure CORS Configuration | Medium | Confirmed | Confirmed | next.config.js:15 |
Finding Details
SEC-001: Missing Row Level Security (RLS)
The profiles and orders tables are initialized without RLS enabled. This allows any authenticated user (or anonymous user, depending on API key) to read and write to any row via the Supabase client.
- Impact: Full data leak of user profiles and order history.
- Remediation: Enable RLS and implement policies using
auth.uid().
SEC-002: Client-side Authorization Logic
The admin dashboard checks user roles only via a client-side React component.
- Impact: Users can bypass UI restrictions by navigating directly to API routes.
- Remediation: Move authorization checks to Middleware or Server Actions.
SEC-003: Exposed Service Role Key
The .env.example file contains a placeholder that suggests developers should place the SERVICE_ROLE_KEY in environment variables accessible to the browser.
- Impact: Potential for accidental exposure of administrative keys in client bundles.
- Remediation: Use the
ANON_KEYfor client-side and keep theSERVICE_ROLE_KEYstrictly server-side.
Next steps
- Switch to
FIXmode to enable RLS on the identified Supabase tables. - Review the
src/middleware.tsto implement server-side path protection for/admin. - Rotate any existing service keys if they have been committed to version control history.
page.tsx
TSX · React component
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Writing code is only half the battle. Shipping insecure, non-compliant, or inaccessible software leads to data breaches and costly rework. This skill provides a security-first engineering partner to audit, harden, and verify your web applications and AI agent integrations before they hit production.
What it does
- Security Auditing Performs read-only discovery and assessment of repositories to identify vulnerabilities and privacy risks.
- Automated Hardening Implements approved remediations for authentication, authorization, secrets management, and database policies.
- Threat Modeling Maps assets, entry points, and trust boundaries to prioritize realistic threats over generic checklists.
- AI Agent Protection Reviews agent tools, prompt injection risks, and tool execution sandboxing to prevent exfiltration.
- Release Readiness Evaluates accessibility, legal headers, and payment security to ensure the product is actually ready for users.
How it works
- Scope definition The user confirms the repository, environment, and authorization boundaries for local or staging tests.
- Project discovery The agent identifies languages, frameworks, identity providers, and sensitive data flows.
- Assessment It applies controls from a master checklist, mapping findings to specific file paths and line numbers.
- Remediation and Verification It proposes minimal safe fixes, implements them upon approval, and runs verification tests to confirm the fix.
Frameworks & tools
Works with any web stack including React, Next.js, Node.js, Python, and SQL databases. Integrates with CI/CD workflows, containerized environments, and cloud infrastructure like AWS or Supabase.
Why this beats prompting it yourself
Generic prompts often miss the difference between authentication and authorization or suggest "sanitization" instead of context-aware encoding. This skill follows a mandatory workflow that enforces evidence collection, uses established cryptographic standards, and handles the specific security nuances of AI agents.
Use cases
- Pre-launch Audit Run a full security pass on a SaaS MVP to find blockers before the first customer signs up.
- Secrets Cleanup Identify and redact hardcoded credentials or service-role keys before a public repository push.
- AI Tool Hardening Review the permissions and schemas of an AI agent's toolset to prevent unauthorized shell or database access.
- Compliance Mapping Verify technical controls against privacy and data rights standards for regulated industries.
Known limitations
Does not provide legal advice or official certifications. Active testing against live production systems is restricted without explicit per-action authorization.
No legal certifications provided. Active probing of live targets requires explicit user approval. Cannot guarantee 100% security or compliance.
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 any payment.
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 7 days ago
Requires
Permissions
Why Secure Vibe Coding needs these permissions : Terminal / Shell: Required to run security audit scripts, execute tests, inspect project configurations, check dependencies, and verify remediation. Read Files: Allows the skill to inspect source code, configuration files, database rules, authentication logic, API endpoints, and project architecture to identify vulnerabilities. Write Files: Required to implement approved security fixes, update configurations, add security controls, create tests, and document remediation. Browser: Allows the skill to inspect web applications, review browser-based behaviour, test user flows, and identify UI-level security and usability issues. Network Access: Needed for authorized external security checks, dependency and vulnerability intelligence, documentation lookup, API integration testing, and verification of external service configurations. Environment Variables: Allows the skill to inspect environment variable names and configuration references to detect insecure setups, missing variables, and potential secret exposure without revealing or logging secret values.