Warn on destructive commands and block edits outside a configured directory.
npx skills add https://github.com/florianhorner/mammamiradio --skill guard-florianhorner
Please help me install this Agent Skill. Skill: guard Source: https://github.com/florianhorner/mammamiradio/tree/main/.claude/skills/gstack/guard Command: npx skills add https://github.com/florianhorner/mammamiradio --skill guard-florianhorner
Provides a comprehensive safety layer that combines warnings for destructive commands with directory-based edit restrictions to prevent accidental damage in development or production.
Specify the directory to restrict edits to and enable destructive-command warnings to activate guard mode.
High-intent search queries and answers about installing and using this skill.
Preventing accidental destructive bash commands requires a safety layer that warns when dangerous shell commands are executed. This guard mode applies during risky shell sessions, ensuring critical systems remain shielded from unintended damage.
A directory-based edit boundary blocks file modifications outside a specifically configured directory. This prevents accidental edits to critical system files while still allowing safe changes inside the defined boundary during debugging.
Yes, you can restrict file edits to a specific directory by configuring a guard boundary. This blocks modifications outside the target folder, allowing teams to debug with confidence without risking unrelated files.
Configuring a guard boundary involves specifying a target directory to restrict edits to and enabling destructive-command warnings. This integrates with helper scripts to enforce the safety boundary across shell sessions.
A safety layer for destructive command warnings is needed when performing risky operations in development or production. It is essential during production maintenance or debugging tasks where accidental damage to critical systems is a concern.
Edit boundary enforcement limitations include its restriction to a single configured directory, meaning edits outside are completely blocked. It relies on integration with specific helper scripts to maintain the guard boundary during active shell sessions.
geoffreywooWarn on destructive commands and restrict edits to a specified directory.
GenEducationWarn on destructive shell commands and restrict file edits to an approved directory.
agwacomRestrict edits to a chosen directory and warn before destructive commands.
AskTinNguyenEnforce persona-level hard-deny policies on AI agent tool usage.
