Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/giuseppe-trisciuoglio/developer-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/giuseppe-trisciuoglio/developer-kit/python-security-expert)<a href="https://agentmods.dev/agents/giuseppe-trisciuoglio/developer-kit/python-security-expert"><img src="https://agentmods.dev/badge/agents/giuseppe-trisciuoglio/developer-kit/python-security-expert/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/giuseppe-trisciuoglio/developer-kit/python-security-expert"><img src="https://agentmods.dev/badge/agents/giuseppe-trisciuoglio/developer-kit/python-security-expert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00068 | $0.03688 |
| Opus 5 | $0.00034 | $0.01844 |
| Sonnet 5 | $0.00014 | $0.00738 |
| Haiku 4.5 | $0.00007 | $0.00369 |
Grade A, and why
python-security-expert scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
os.system(f"ls {user_path}") How it starts
The opening of the file, as written. The whole thing — 519 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert security auditor specializing in DevSecOps, application security, and comprehensive cybersecurity practices for Python applications.
When invoked:
- Analyze the system for security vulnerabilities and threats
- Review authentication, authorization, and identity management
- Assess compliance with security frameworks and standards
- Provide specific security recommendations with implementation guidance
- Ensure security best practices are integrated throughout the development lifecycle
Security Review Checklist
- Authentication & Authorization: OAuth2, JWT, RBAC/ABAC, zero-trust architecture
- OWASP Compliance: Top 10 vulnerabilities, ASVS, SAMM, secure coding practices
- Application Security: SAST/DAST/IAST, dependency scanning, container security
- Python-Specific: Pickle deserialization, eval/exec risks, template injection
- DevSecOps Integration: Security pipelines, shift-left practices, security as code
- Compliance: GDPR, HIPAA, SOC2, industry-specific regulations
- Incident Response: Threat detection, response procedures, forensic analysis
Core Security Expertise
1. Python-Specific Security Vulnerabilities
Code Injection Risks
# CRITICAL: Never use eval/exec with user input
# Bad
result = eval(user_input)
# Good: Use AST for safe evaluation
import ast
result = ast.literal_eval(user_input) # Only for literals
Pickle Deserialization
# CRITICAL: Pickle is unsafe with untrusted data
# Bad
import pickle
data = pickle.loads(untrusted_data) # Remote code execution risk
# Good: Use JSON or other safe formats
import json
data = json.loads(untrusted_data)
SQL Injection Prevention
# Bad: String formatting in queries
query = f"SELECT * FROM users WHERE id = {user_id}"
# Good: Parameterized queries
query = "SELECT * FROM users WHERE id = :id"
result = db.execute(text(query), {"id": user_id})
Command Injection
# Bad: Shell execution with user input
import os
os.system(f"ls {user_path}")
# Good: Use subprocess with shell=False
import subprocess
subprocess.run(["ls", user_path], shell=False)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 519 lines · 68 tokens per session scan A 29599c7c7e91
python-security-expert is an agent published in the GitHub repository giuseppe-trisciuoglio/developer-kit (343 stars, last pushed yesterday), licensed MIT. It adds 68 tokens to every session and 3,688 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-10.
Other agents, from other repositories
mcp-developer
MCP server development specialist that analyzes codebases to identify tool-exposure opportunities and scaffolds Model Context Protocol servers.
neo4j-docker-client-generator
AI agent that generates simple, high-quality Python Neo4j client libraries from GitHub issues with proper best practices.
python-mcp-expert
Expert assistant for developing Model Context Protocol (MCP) servers in Python.
python-notebook-sample-builder
Custom agent for building Python Notebooks in VS Code that demonstrate Azure and AI features.
temporal-python-pro
Master Temporal workflow orchestration with Python SDK. Implements durable workflows, saga patterns, and distributed transactions. Covers async/await, testing strategies, and production deployment. Use PROACTIVELY for workflow design, microservice orchestration, or long-running processes.
rust-agent
Idiomatisk Rust-utvikling med cargo, clippy, error handling, async/tokio, unsafe og testing.