Plugin Claude Code
The 10 Commandments of Claude Code — turned into skills, agents, hooks, and commands. 10 experts, 231 tips (209 from the experts, 22 from the docs), distilled into one plugin.
Agent
Use this agent to review code for security, quality, and convention adherence. Dispatched by /review command. Confidence-based filtering, stack-agnostic.
Agent
Use this agent to review implementation plans as a staff engineer. Dispatched by /plan-first command. Checks for missing edge cases, unclear scope, dependency risks, missing test strategy.
Agent
Use this agent when a raw, vague, or imperative development prompt needs to become declarative, exhaustive, and verification-driven before it gets executed. Dispatched by the /improve-prompt command. Context: User gives a short, imperative instruction that lacks scope and verification steps. user: "fix the login bug"…
Agent
Use this agent for deep codebase exploration without polluting main context. Dispatched by /research command. Returns structured summary.
Agent
Use this agent to enforce test-driven development with vertical slicing. Dispatched by /tdd command. RED-GREEN-REFACTOR cycle, one test at a time.
Command
Smart compaction. Summarize modified files, current task state, key decisions before compacting.
Command
Capture a correction or learning as a CLAUDE.md rule. Detects scope, checks duplicates, writes concise rule.
Command
Systematic 9-step debugging framework. Generate 20+ hypotheses, rank by likelihood, implement most probable fix.
Command
Auto-detect and run project verification. Find tests, linters, build scripts. Run all, iterate on failures up to 3x.
Command
Transform a raw prompt into an exhaustive, verification-driven development prompt. Use --preview to see before executing.
Command
Claude interviews you about a feature before building it. Asks questions one at a time, writes SPEC.md when done.
Command
Plan before writing code. Explore codebase, produce concise plan, list unresolved questions. Optionally dispatch plan-reviewer agent.
Command
Dispatch deep research subagent. Reads files, searches codebase, returns concise summary without polluting main context.
Command
Code review with subagent. Security, quality, and conventions. Confidence-based filtering — only reports findings >80% confidence.
Command
Bootstrap your project with CC Bible best practices. Generate CLAUDE.md rules, settings suggestions, and a guide to non-automatable commandments.
Command
Test-driven development with vertical slicing. ONE test, ONE implementation, repeat. Dispatch tdd-guide agent.
Command
Escalating strategy when Claude is stuck or looping. 4 techniques from 4 experts, applied in order.
Hook
Runs after a tool call finishes for Edit and Write tool calls, executing format-check.sh via bash. From 4riel/cc-bible.
Hook
Runs before the agent uses a tool for Bash tool calls, executing git-guardrail.sh via bash. From 4riel/cc-bible.
Skill Claude CodeCodex
This skill should be used when debugging a bug, error message, or test failure — tracing root cause, generating and ranking hypotheses, or when stuck/looping and needing an escalating strategy (step back, compare plans, naive-first, ultrathink). Contains Shumer's 9-step framework and 4 escalating techniques for…
Skill Claude CodeCodex
This skill should be used when planning implementation work before writing code — drafting a plan, reviewing one as a staff engineer, structuring unresolved questions and success criteria, comparing two competing approaches, or ordering phases by dependency and risk. Contains plan formats, evaluation criteria, and the…
Skill Claude CodeCodex
This skill should be used when rewriting a vague or imperative prompt into a declarative, exhaustive, verification-driven one — expanding scope, injecting test/lint/build verification commands, defining completion criteria, or auditing UI and cross-cutting changes for full coverage. Contains exhaustiveness rules…
Skill Claude CodeCodex
This skill should be used when practicing test-driven development — writing one failing test at a time through RED-GREEN-REFACTOR, designing testable interfaces, deciding what to mock versus what not to mock, or choosing between pass@k and pass^k test strategies. Contains vertical-slicing rules and testing…