latiotech

Harden AI ML — AI skill for Claude Code

AI community

Fix unsafe model deserialization and harden AI/ML model usage.

How to install Harden AI ML

Installs to ~/.claude/skills/latiotech-secure-supply-chain-skills-harden-ai-ml/SKILL.md

Terminal

mkdir -p ~/.claude/skills/latiotech-secure-supply-chain-skills-harden-ai-ml && curl -fsSL https://raw.githubusercontent.com/latiotech/secure-supply-chain-skills/HEAD/commands/harden-ai-ml.md -o ~/.claude/skills/latiotech-secure-supply-chain-skills-harden-ai-ml/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Harden AI ML does

description: Fix unsafe model deserialization and harden AI/ML model usage allowed-tools: Read, Write, Edit, Glob, Grep, Bash(pip:, python:, modelscan:, git:, curl:, which:) argument-hint: [directory]

Audit and harden AI/ML model usage for supply chain security. **This command takes action by default** - it fixes unsafe deserialization calls, flags dangerous model files, and adds hash verification. Changes are explained as they are made.

Read `${CLAUDE_PLUGIN_ROOT}/skills/supply-

Alternatives in AI

Codeburn — Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Code 9.7k ★
Ccstatusline — by sirmalloc - A highly customizable status line formatter for Claude Code CLI that displays model info, git b 5.5k ★
Manifest — Real-time cost observability for OpenClaw agents — track tokens, costs, messages, and model usage 4.1k ★