agentsclimarketplace

Claude code bash patterns

Skill secondsky/claude-skills/plugins/claude-code-bash-patterns/skills/claude-code-bash-patterns

Claude Code Bash tool patterns with hooks, automation, git workflows. Use for PreToolUse hooks, command chaining, CLI orchestration, custom commands, or encountering bash permissions, command failures, security guards, hook configurations.From its SKILL.md

Install
npx -y skills add secondsky/claude-skills --skill claude-code-bash-patterns

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

8.4 KB, ~2.0k tokens by cl100k_base, as published. Nobody here has run it

Claude Code Bash Patterns

Status: Production Ready ✅ | Last Verified: 2025-11-18


Quick Start

Basic Command

ls -la

Command Chaining

bun install && bun run build && bun test

Hooks

Create .claude-hook-pretooluse.sh:

#!/usr/bin/env bash
# PreToolUse hook - runs before every Bash command

echo "Running: $1"

Load references/hooks-examples.md for complete hook patterns.


The Five Core Patterns

1. Sequential Operations (&&)

Use when: Each command depends on previous success

git add . && git commit -m "message" && git push

Why: Stops chain if any command fails


2. Parallel Operations (Multiple tool calls)

Use when: Commands are independent

Message with multiple Bash tool calls in parallel

Load references/cli-tool-integration.md for parallel patterns.


3. Session Persistence

Use when: Need to maintain state across commands

# Set environment variable
export API_KEY="sk-..."

# Use in later commands (same session)
curl -H "Authorization: Bearer $API_KEY" api.example.com

4. Background Processes

Use when: Long-running tasks

npm run dev &
# Get PID with $!

5. Hooks for Automation

Use when: Need pre/post command logic

Load references/hooks-examples.md for all hook types.


Critical Rules

Always Do ✅

  1. Use && for sequential dependencies (not semicolons)
  2. Quote paths with spaces (cd "path with spaces")
  3. Check environment before destructive ops (rm, git push --force)
  4. Use specialized tools first (Read, Grep, Glob before Bash)
  5. Set timeouts for long operations (up to 10 minutes)
  6. Validate inputs before passing to shell commands
  7. Use hooks for repeated patterns (logging, validation)
  8. Maintain session state (export variables once)
  9. Handle errors explicitly (check exit codes)
  10. Document custom commands in .claude/commands/

Never Do ❌

  1. Never use ; for dependent commands (use &&)
  2. Never skip quoting paths with spaces
  3. Never run rm -rf without confirmation
  4. Never expose secrets in command output
  5. Never ignore timeout limits (max 10 min)
  6. Never use bash for file operations when specialized tools exist
  7. Never chain with newlines (use && or ; explicitly)
  8. Never force-push to main without explicit user request
  9. Never skip hooks (--no-verify) without user request
  10. Never use interactive commands (git rebase -i, git add -i)

Git Workflows

Basic Commit

git add . && git commit -m "feat: add feature"

Commit with Testing

npm test && git add . && git commit -m "fix: bug fix" && git push

Pull Request

git checkout -b feature/new && git add . && git commit -m "feat: new feature" && git push -u origin feature/new

Load references/git-workflows.md for complete workflows including:

  • Feature branch workflow
  • PR creation automation
  • Commit message conventions
  • Pre-commit validation

Hooks: Advanced Automation

PreToolUse Hook

.claude-hook-pretooluse.sh:

#!/usr/bin/env bash
COMMAND="$1"

# Log all commands
echo "[$(date)] Running: $COMMAND" >> ~/claude-commands.log

# Block dangerous patterns
if [[ "$COMMAND" =~ rm\ -rf\ / ]]; then
    echo "❌ Blocked dangerous command"
    exit 1
fi

Hook Types

  • pretooluse - Before every Bash command
  • stop - Before conversation ends
  • user-prompt-submit - After user submits message

Load references/hooks-examples.md for all hook types and examples.


CLI Tool Integration

npm/bun

bun install && bun run build

wrangler (Cloudflare)

bunx wrangler deploy

gh (GitHub CLI)

gh pr create --title "Fix bug" --body "Description"

Load references/cli-tool-integration.md for complete tool patterns.


Custom Commands

Create .claude/commands/deploy.md:

---
description: Deploy to production
---

Run these steps:
1. Run tests: `npm test`
2. Build: `npm run build`
3. Deploy: `wrangler deploy`

User can invoke with: /deploy

Load templates/custom-command-template.md for template.


Security

Allowlisting Tools

settings.json:

{
  "dangerousCommandsAllowList": [
    "git push --force"
  ]
}

Secrets Management

# ✅ Good: Use environment variables
export API_KEY="$SECURE_VALUE"

# ❌ Bad: Hardcode secrets
curl -H "Authorization: Bearer sk-abc123..."

Load references/security-best-practices.md for complete security guide.


Common Use Cases

Use Case 1: Test Before Commit

npm test && git add . && git commit -m "message"

Use Case 2: Deploy with Validation

npm run lint && npm test && npm run build && bunx wrangler deploy

Use Case 3: Multi-Repo Operations

cd repo1 && git pull && cd ../repo2 && git pull

Use Case 4: Background Process

npm run dev &

Load references/cli-tool-integration.md for more patterns.


Troubleshooting

Issue: Command times out

Solution: Increase timeout or use background mode

# Background mode
npm run dev &

Issue: Path with spaces fails

Solution: Quote the path

cd "path with spaces/file.txt"

Issue: Hook blocks command

Solution: Check hook logic in .claude-hook-pretooluse.sh

Load references/troubleshooting-guide.md for all issues.


When to Load References

Load references/git-workflows.md when:

  • Setting up git automation
  • Creating PRs programmatically
  • Need commit message conventions
  • Want pre-commit validation patterns

Load references/hooks-examples.md when:

  • Creating custom hooks
  • Need hook templates
  • Want validation patterns
  • Implementing logging/security

Load references/cli-tool-integration.md when:

  • Orchestrating multiple CLI tools
  • Need tool-specific patterns
  • Want parallel execution examples
  • Troubleshooting tool integration

Load references/security-best-practices.md when:

  • Configuring security guards
  • Setting up allowlisting
  • Managing secrets
  • Preventing dangerous operations

Load references/troubleshooting-guide.md when:

  • Debugging command failures
  • Encountering timeouts
  • Hooks behaving unexpectedly
  • Session state issues

Using Bundled Resources

References (references/)

  • git-workflows.md - Complete git automation patterns
  • hooks-examples.md - All hook types with examples
  • cli-tool-integration.md - Tool orchestration patterns
  • security-best-practices.md - Security configuration guide
  • troubleshooting-guide.md - Common issues and solutions

Templates (templates/)

  • custom-command-template.md - Custom command template
  • settings.json - Security settings example
  • .envrc.example - Environment variables example
  • github-workflow.yml - GitHub Actions integration
  • dangerous-commands.json - Dangerous patterns list

Examples

Feature Development Workflow

git checkout -b feature/oauth && \
  npm test && \
  git add . && \
  git commit -m "feat(auth): add OAuth support" && \
  git push -u origin feature/oauth

CI/CD Pipeline

npm run lint && \
  npm test && \
  npm run build && \
  bunx wrangler deploy

Multi-Project Update

cd project1 && bun install && cd ../project2 && bun install

Official Documentation


Questions? Issues?

  1. Check references/troubleshooting-guide.md for common issues
  2. Review references/git-workflows.md for git patterns
  3. See references/hooks-examples.md for automation
  4. Load references/security-best-practices.md for security

What ships with it: 13 files

66.8 KB alongside SKILL.md, 3 of them executable

Gives 0 of the 12 instructions most context ai engineering skills give in ~2.0k tokens

Counted across 1,193 of the 1,976 authors here whose files we hold, read 2026-08-07

  • Dispatch a fresh implementer subagent per taskin 48 of 1193, across 19 files
  • Dispatch a final code reviewer after all tasksin 33 of 1193, across 8 files
  • Provide full task text to the subagentin 30 of 1193, across 9 files
  • Review spec compliance before code qualityin 27 of 1193, across 10 files
  • Make the hook script executablein 26 of 1193, across 8 files
  • Re-snapshot after navigation or DOM changesin 25 of 1193, across 19 files
  • Read files before editing themin 22 of 1193, across 11 files
  • Answer subagent questions before proceedingin 22 of 1193, across 7 files
  • Mark task complete in TodoWrite after approvalin 22 of 1193, across 6 files
  • Merge hook into existing settingsin 21 of 1193, across 3 files
  • Ask if installation is global or projectin 20 of 1193, across 2 files
  • Copy the hook script to target locationin 20 of 1193, across 2 files

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,629. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.