chrisryugjchrisryugjCommunity·1 Agent Skills Included

fable-ish

Enforces verified completion and blocks risky coding actions

Classifies coding requests by risk and blocks destructive commands like rm -rf, git push, and secret file edits before they run. Records which files changed and which verification commands actually ran in a local evidence ledger. Stops unfinished work at completion time when code changed without passing proof, eliminating false done claims.
npx skills add chrisryugj/fable-ish --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install fable-ish?

Run `npx skills add chrisryugj/fable-ish --all -g -y` in your terminal to install the skill globally, then restart your coding environment.

What does fable-ish do?

It adds verification discipline to coding work by classifying task risk, blocking dangerous commands, and refusing to accept completion without observable proof like passing tests.

How does fable-ish block dangerous commands?

Its pre-execution hooks deny destructive actions such as rm -rf, git push, terraform destroy, and edits to secret files like .env or .pem before they run.

Does fable-ish work with Claude Code?

Yes. It is built as a Claude Code plugin with lifecycle hooks and a SKILL.md workflow, and it follows the universal skill standard.

Do I need to configure fable-ish after installing?

Minimal setup is needed. Review and trust the bundled hooks once, and task classification, evidence logging, and the completion gate then run automatically.

Related Repositories in Software Engineering

View All in Software Engineering