Stats
Actions
Tags
From frontend-skills
Enforces bun + tsgo toolchain by blocking npm, npx, tsc, and global installs via PreToolUse hooks. Also guards destructive commands and sets env vars on session start.
How this skill is triggered — by the user, by Claude, or both
Slash command
/frontend-skills:setup-toolchainThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- **PreToolUse hooks** block banned CLI commands, give actionable suggestions
rm -rf (except safe targets like node_modules/.next/dist/build), git push --force, git reset --hard, git checkout . / git restore ..claude/settings.json (project-level, committed to git)Copy scripts/enforce-toolchain.sh and scripts/session-env.sh to .claude/hooks/. Make executable: chmod +x .claude/hooks/*.sh
.claude/settings.jsonAdd to hooks config (merge existing):
Bash): .claude/hooks/enforce-toolchain.sh.claude/hooks/session-env.sh.claude/hooks/enforce-toolchain.sh exists, executable.claude/hooks/session-env.sh exists, executable.claude/settings.json has both hook entriesnpm install in Claude -- should blockbun add lodash in Claude -- should block (missing --yarn)Stage .claude/hooks/ and .claude/settings.json. Commit: Add toolchain enforcement hooks (bun + tsgo)
npx claudepluginhub redpanda-data/ui-harness --plugin frontend-skillsDevelops Claude Code hooks to run shell commands or LLM prompts on events like PreToolUse for automations, guardrails, quality checks, and pre-push tests.
Generates PermissionRequest hooks that auto-approve safe operations, auto-deny dangerous ones, and tailor rules to detected project stack. Safer alternative to --dangerouslySkipPermissions for manual permission mode.
Provides Claude Code bash patterns for PreToolUse hooks, command chaining, CLI orchestration, git workflows, custom commands, and handling bash permissions, failures, security guards.