Sourcery MCP
Let your coding agent request Sourcery reviews and fix the comments and security findings itself
HOL Guard
Open-source local firewall that checks what your coding agent is about to do and blocks the dangerous stuff
About
Stop your AI agent from deleting files, leaking secrets or installing shady packages
Install with pipx install hol-guard, then run hol-guard init: the setup wizard finds your agents and asks before each change. To set up one agent explicitly run hol-guard install codex (or claude, cursor, opencode), and try hol-guard run codex --dry-run first. Check it with hol-guard status. Things to know: it needs Python 3.10+, enforcement depth depends on the hooks each agent exposes (the coverage matrix in the repo lists what is blocked versus only observed), and Guard Cloud is optional, only for synced history and team policy.
Features
- Allow, ask or block decisions before risky agent actions run
- Blocks destructive commands, secret reads and outbound data leaks
- Reviews package installs against supply-chain intelligence
- Inspects MCP servers, plugins, skills and agent config for poisoning and prompt injection
- Local dashboard, approval center and decision history
- Works offline with no account, open source under Apache-2.0
- plugin-scanner CLI and GitHub Action to gate plugins and skills in CI
Install
Related skills
All skillsRelated articles
All articlesGrok Bot vs Claude Code and Codex: What Each One Is For
How Grok Bot, a persistent personal agent, differs from coding agents like Claude Code and Codex, and when it makes sense to use each one or both together.
7 Tools to Cut Your Coding Agent's Token Bill (2026)
Seven tools that cut Claude Code, Codex and Cursor token usage by compressing shell output, searching code and fetching current docs, with honest caveats.
Best Memory Tools for Claude Code (2026)
Five tools that give Claude Code memory across sessions, agents and machines, with install commands, honest caveats and when CLAUDE.md is enough.
