Vuln Sandbox Labs: Secure-Coding Training
Ten deterministic secure-coding labs for practicing OWASP exploit payloads and fixes in a safe JavaScript sandbox.
$9
Works with the AI Agents you already use
Vuln Sandbox Labs: Secure-Coding Training
Example session with this skill installed
I want to practice SQL injection. Show me the scenario for kev-sqli and then let me try to bypass the login as admin.
- Read your context and instructions
- Compiled the vuln sandbox labs
I have loaded the SQL Injection lab. Here is your scenario and goal. Use the feedback from the simulated engine to refine your payload.
Lab: kev-sqli (SQL Injection)
| Attribute | Details |
|---|---|
| Weakness | SQL Injection in Login Form |
| Goal | Log in as the admin user without knowing the password. |
| Hint | The application uses a simple string concatenation for the query. Try to close the quote and comment out the rest of the statement. |
Simulation Output for your first attempt:
node scripts/lab.cjs try kev-sqli "admin' OR '1'='1"
{
"correct": true,
"output": "SELECT * FROM users WHERE username = 'admin' OR '1'='1' AND password = '...'",
"verdict": "Success. The engine returned the first row in the table (admin) because the OR condition evaluated to true for all rows."
}
Remediation: The Fix
The vulnerability existed because the input was treated as code. To fix this, you must use
Parameterized Queries (Prepared Statements).
// VULNERABLE
const query = "SELECT * FROM users WHERE user='" + req.body.user + "' AND pass='" + req.body.pass + "'";
// SECURE
const query = "SELECT * FROM users WHERE user = ? AND pass = ?";
db.execute(query, [req.body.user, req.body.pass]);
Next steps
- Try the
kev-pathlab to practice bypassing path traversal guards. - Review
reference/fixes.mdfor a deep dive into whymysql.escape()is less secure than preparation. - Apply the parameterized query pattern to your current project's database layer.
Connects securely to your tools. The creator never sees your data.
About this skill
Security education often relies on dangerous live environments or dry theory. This skill provides ten self-contained, deterministic labs for practicing exploit payloads and secure remediation without any real-world risk.
What it does
- Simulates vulnerabilities using toy engines for SQL, file systems, and shells that execute nothing on the host.
- Grades inputs by checking learner payloads against specific win conditions for OWASP-top-10 style weaknesses.
- Provides scenarios with goals and hints for weaknesses like SQL injection, SSRF, and buffer overflows.
- Teaches remediation with reference fixes showing why a filter failed and how to implement the secure version.
How it works
- List labs to see available modules like
kev-sqli,kev-xss, orkev-ssrf. - Display a scenario card for a specific lab to give the learner a goal and a hint.
- Grade a payload by passing the learner's input into the lab script to receive a verdict and simulation output.
- Review the fix from the reference library to explain the underlying secure coding principles.
Frameworks & tools
Plain JavaScript environment using node. Uses local scripts in scripts/ and reference documentation in reference/.
Why this beats prompting it yourself
Generic LLMs often refuse to generate exploit payloads due to safety filters or provide inconsistent "simulations" that hallucinate success. This skill uses deterministic logic to ensure a payload only works if it actually bypasses the simulated guardrail, providing a reliable grading gate.
Use cases
- Conduct hands-on secure coding training for engineering teams.
- Run offline capture-the-flag (CTF) style exercises during workshops.
- Verify a developer's understanding of a specific CVE before a code review.
- Provide a safe sandbox for students to test "illegal" syntax without network access.
Known limitations
Simulations are simplified JavaScript logic and do not represent the full complexity of production databases or kernels. Requires a Node.js environment to run the underlying scripts.
Requires Node.js to run scripts. Simulations are deterministic logic gates, not actual operating systems or databases. Only supports the ten included vulnerability types.
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 today
Requires