One jdbc.query("... WHERE email = '" + email + "'") or one new ObjectInputStream(request.getInputStream()) is enough to lose your data or your server. This skill finds those and eleven other Java security mistakes before you deploy, and tells you how to fix each one.

Your agent runs a scanner on your .java files. It skips comments, tests and generated code, understands strings, text blocks and concatenation, follows SQL built in a variable or a StringBuilder to the call that runs it, and knows which values come from the request (getParameter, @RequestParam, @PathVariable, JAX-RS parameters, upload file names). It knows the safe patterns: ? and :name placeholders, constants and numbers, nCopies placeholder lists, sort columns checked against an allowlist, ObjectInputFilter, disallow-doctype-decl, SecureRandom, and normalize() with startsWith. It never compiles or runs Java. It returns BLOCK, REVIEW, or NOT FLAGGED, with the file, line, reason, and a safer approach. Secrets are never printed.

What it checks

BLOCK (fix before shipping)

• J01: SQL or JPQL built with +, String.format or StringBuilder and run with JDBC, JdbcTemplate, JdbcClient, JPA or Hibernate, and ${} in MyBatis annotations

• J02: Runtime.exec or ProcessBuilder running sh -c with a changing command, one command string built from a value, or a program name from the request

• J03: ObjectInputStream without a filter, SerializationUtils.deserialize, XMLDecoder, XStream any-type permission, Jackson default typing or Id.CLASS on Object fields, SnakeYAML allowing every tag

• J04: hard-coded passwords, API keys, tokens, JWT and HMAC keys, @Value defaults and passwords in JDBC URLs

• J05: JWTs accepted without a signature (Algorithm.none(), parseClaimsJwt, unsecured()) or decoded and never verified

REVIEW (check before release)

• J11: XML parsers (JAXP, dom4j, JDOM) without DTDs or external entities turned off (XXE)

• J12: XStream without an allowlist and SnakeYAML new Yaml() without SafeConstructor

• J13: DES, triple DES, RC4, Blowfish, ECB mode and Cipher.getInstance("AES")

• J14: MD5 or SHA on passwords and NoOpPasswordEncoder or other weak Spring encoders

• J15: java.util.Random, Math.random(), ThreadLocalRandom or RandomStringUtils for tokens, OTPs and reset codes

• J16: trust-all TrustManagers and hostname verifiers, NoopHostnameVerifier, TrustSelfSignedStrategy

• J17: file paths from the request in new File, Paths.get, resolve or FileSystemResource

• J18: passwords, tokens and the Authorization header written to logs

What's included

• The scanner (Python 3.9+, standard library only)

• A rules guide with the reason and safer approach for each rule

• A risky and a safe example Spring Boot service, and the real report the scanner produces on the risky one

Limits

It reads code as text, not with the Java compiler: types are not known and values are followed within one file. application.properties and application.yml, dependencies and known CVEs, Kotlin and JSP files are not checked. NOT FLAGGED is not a guarantee.

Contents

One ZIP file with 20 files: SKILL.md, the scanner in scripts/ (5 Python files), references/rules.md (every rule with its reason and safer approach), and examples/ (12 example files plus report.md, the real scanner output on the risky example).

How to use: This is an agent skill: a folder with a SKILL.md file and Python scripts that an AI coding agent such as Claude Code, Cursor, or Codex CLI can run. Unzip it into your agent's skills folder (for Claude Code: ~/.claude/skills/), then ask your agent to check your project. The scripts also run on their own with Python 3.9 or newer, with nothing to install. They read your files only: no network access, and nothing is sent anywhere.