A single f-string in cursor.execute() is a SQL injection, and a single pickle.loads() on request data lets an attacker run code on your server. This skill finds those and eight other security mistakes in Python code (Flask, Django, FastAPI or plain scripts) before you merge or deploy, and tells you how to fix each one.

Your agent runs a scanner on your .py files. It reads them with Python's own parser, so it follows imports and aliases (import pickle as pk, from yaml import load) and follows a query built in a variable to the execute() call that runs it. Query text built only from ALL_CAPS constants is not flagged, and test files are read but not reported. It never imports or runs your code. 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)

• P01: SQL built with an f-string, %, + or .format() and passed to execute(), raw(), text() or read_sql()

• P02: pickle, marshal or dill loads, pandas.read_pickle, numpy.load(allow_pickle=True), or yaml.load without SafeLoader

• P03: eval() or exec() of anything but a fixed string

• P04: a password, token, secret or API key written as a string, or a private key in the code

REVIEW (check before release)

• P11: TLS verification turned off (verify=False, CERT_NONE, check_hostname=False)

• P12: a JWT decoded without checking its signature

• P13: MD5, SHA-1 or SHA-2 used to hash a password

• P14: Flask, Dash, FastAPI or Starlette debug mode on

• P15: tempfile.mktemp()

• P16: the random module used for a token, password, OTP, salt or key

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 app, and the real report the scanner produces on the risky one

Limits

It follows values within one function, so data that crosses functions or files is not traced. It does not check dependencies, subprocess calls or templates. NOT FLAGGED is not a guarantee.

Contents

One ZIP file with 8 files: SKILL.md, the scanner in scripts/ (3 Python files), references/rules.md (every rule with its reason and safer approach), and examples/ (2 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.