Writ

Claude Code plugin that blocks risky code writes until a human approves the plan and tests.

Pluginfreeglobal

Writ is a Claude Code plugin that acts as a governance runtime, enforcing engineering controls outside the model rather than trusting it to remember them from context. It can stop a coding agent until a human approves its plan and tests, guards credential writes and other protected actions, surfaces only the rules relevant to the file and phase in front of the agent from a set of 287 security, architecture, and testing rules, and records approved plans and governing rules as provenance on pull requests.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of Writ - the signals our matcher actually reads to decide when to surface it. It stays private; its maker sees and controls it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Something wrong with this listing — dead link, not a real product, wrong info?

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether Writ (or something else) actually fits.

Try a problem