Intercept destructive commands and enforce directory locks during production operations.
npx skills add https://github.com/nextc/nextc-claude --skill safety-guard-nextc
Please help me install this Agent Skill. Skill: safety-guard Source: https://github.com/nextc/nextc-claude/tree/main/nextc-ecc/skills/safety-guard Command: npx skills add https://github.com/nextc/nextc-claude --skill safety-guard-nextc
Prevents destructive operations on production systems and during autonomous agent sessions by enforcing safety guardrails and limiting risky actions.
Enable safety-guard in an autonomous session and attempt a destructive command to see the protection in action.
High-intent search queries and answers about installing and using this skill.
Preventing destructive bash commands during autonomous agent runs requires intercepting commands before execution. Safety-guard intercepts risky bash operations and prompts for confirmation to halt inadvertent destructive actions.
Locking directory edits during production maintenance is achieved by defining a project subset and restricting modifications. Freeze mode locks edits to a defined subset of the project to prevent inadvertent changes to critical files.
Audit logging for blocked destructive commands records intercepted actions for later review. Safety-guard logs blocked actions during autonomous sessions and production maintenance to provide an audit trail of prevented destructive operations.
Safety guardrails can be used for fully automated deployments by combining protections. Guard mode combines careful command interception and freeze directory locks to maximize safety during fully autonomous runs and deployments.
The difference between careful and freeze mode lies in their protective scope. Careful mode intercepts destructive commands and prompts for confirmation, while freeze mode locks edits to a defined project subset to prevent inadvertent file changes.
Configuring directory locks before running autonomous agents on production systems is recommended to prevent inadvertent changes. Freeze mode enforces directory locks to guard critical files and configurations during automated agent sessions.
26mitch26Monitor and restrict destructive commands and file edits within specified directories.
richardnpaulBlocks destructive commands and enforces directory-specific write locks on production systems.
riftzen-bitAutomate continuous agent loops with quality gates, evals, and recovery controls.
VertaKhanWarn on destructive commands and restrict edits to a specified directory.
