This skill should be used when the user asks about "using requirements framework", "how to configure requirements", "add requirement checklist", "customize requirements", "requirements not working", "bypass requirements", "satisfy requirements", or needs help with the requirements framework CLI (req command). Also triggers on questions about requirement scopes, session management, or troubleshooting hooks.
Configures, customizes, and troubleshoots the hook-based requirements framework for development workflows.
npx claudepluginhub harmaalbers/claude-requirements-frameworkThis skill inherits all available tools. When active, it can use any tool Claude has access to.
examples/bash-command-trigger.yamlexamples/custom-requirement.yamlexamples/dynamic-branch-size.yamlexamples/guard-protected-branch.yamlexamples/minimal-project-config.yamlexamples/project-requirements.yamlexamples/single-use-pre-commit.yamlexamples/team-strict-config.yamlreferences/advanced-features.mdreferences/cli-reference.mdreferences/configuration-patterns.mdreferences/troubleshooting.mdscripts/check-session.shscripts/validate-config.shHelp users configure, customize, and troubleshoot the Claude Code Requirements Framework - a hook-based system that enforces development workflow practices.
Repository: https://github.com/HarmAalbers/claude-requirements-framework
Documentation: ~/.claude/hooks/README-REQUIREMENTS-FRAMEWORK.md
req command and session managementreq status # Check requirement status
req satisfy commit_plan # Mark requirement satisfied
req clear commit_plan # Clear a requirement
req list # List all requirements
req sessions # View active sessions
req init # Interactive project setup
req config commit_plan # View/modify configuration
req doctor # Verify installation
→ Full CLI reference: See references/cli-reference.md
~/.claude/requirements.yaml) - Defaults for all projects.claude/requirements.yaml) - Shared team config (committed).claude/requirements.local.yaml) - Personal overrides (gitignored)| Scope | Lifetime | Use Case |
|---|---|---|
session | Until Claude session ends | Daily planning, ADR review |
branch | Persists across sessions | GitHub ticket linking |
permanent | Never auto-cleared | Project setup |
single_use | Cleared after action | Pre-commit review (each commit) |
# In .claude/requirements.yaml
requirements:
commit_plan:
enabled: true
scope: session
checklist:
- "Plan created via EnterPlanMode"
- "Atomic commits identified"
- "TDD approach documented"
requirements:
my_requirement:
enabled: true
scope: session
trigger_tools:
- Edit
- Write
message: |
🎯 **Custom Requirement**
Explain what needs to be done.
**To satisfy**: `req satisfy my_requirement`
checklist:
- "First step"
- "Second step"
→ More examples: See examples/custom-requirement.yaml
requirements:
pre_commit_review:
enabled: true
scope: single_use
trigger_tools:
- tool: Bash
command_pattern: "git\\s+commit"
message: "Review required before commit"
Pattern Tips:
\\s+ matches whitespace| for OR: git\\s+(commit|push)→ More patterns: See examples/bash-command-trigger.yaml
Option 1: Local override
# .claude/requirements.local.yaml
enabled: false
Option 2: Environment variable
export CLAUDE_SKIP_REQUIREMENTS=1
Option 3: Disable specific requirement
req config commit_plan --disable --local
Use req init for guided project setup:
req init # Interactive wizard
req init --preset strict # Use preset (non-interactive)
req init --yes # Non-interactive with defaults
Presets:
strict - All requirements, session scope (teams)relaxed - Basic requirements, branch scopeminimal - Only commit_plan (learning)advanced - All features + branch limits + guardsinherit - Inherit from global configView and modify settings without editing YAML:
# View
req config # All requirements
req config commit_plan # Specific requirement
# Modify
req config commit_plan --enable
req config commit_plan --disable
req config commit_plan --scope branch
req config adr_reviewed --set adr_path=/custom/path
→ Full config options: See references/cli-reference.md
Sessions are auto-detected. Manual override when needed:
req sessions # List active sessions
req satisfy commit_plan --session ID # Explicit session
req configreq doctorls -la ~/.claude/hooks/*.pyecho $CLAUDE_SKIP_REQUIREMENTSreq sessions # Find session ID
req satisfy NAME --session ID
req prune # Clean stale sessions
→ Full troubleshooting guide: See references/troubleshooting.md
Skills can automatically satisfy requirements:
1. git commit → Blocked by pre_commit_review
2. /requirements-framework:pre-commit runs
3. Auto-satisfies pre_commit_review
4. git commit → Success!
5. single_use clears → Next commit requires review again
Built-in mappings (review skills):
requirements-framework:pre-commit → pre_commit_reviewrequirements-framework:deep-review → pre_pr_review (recommended)requirements-framework:quality-check → pre_pr_review (lightweight alternative)requirements-framework:arch-review → commit_plan, adr_reviewed, tdd_planned, solid_reviewed (recommended)requirements-framework:plan-review → commit_plan, adr_reviewed, tdd_planned, solid_reviewed (lightweight alternative)requirements-framework:codex-review → codex_reviewerBuilt-in mappings (process skills):
requirements-framework:brainstorming → design_approvedrequirements-framework:writing-plans → plan_written, commit_planrequirements-framework:test-driven-development → tdd_plannedrequirements-framework:systematic-debugging → debugging_systematicrequirements-framework:verification-before-completion → verification_evidencerequirements-framework:requesting-code-review → pre_commit_reviewThe framework includes process skills that guide the full development lifecycle:
| Skill | Purpose |
|---|---|
brainstorming | Design-first development (explore → design → approve) |
writing-plans | Create bite-sized implementation plans |
executing-plans | Execute plans with batch checkpoints |
test-driven-development | RED-GREEN-REFACTOR cycle enforcement |
systematic-debugging | 4-phase root-cause investigation |
verification-before-completion | Fresh evidence before claiming done |
subagent-driven-development | Parallel task execution with review |
finishing-a-development-branch | Branch completion and merge options |
using-git-worktrees | Isolated workspace creation |
dispatching-parallel-agents | Concurrent problem solving |
receiving-code-review | Technical evaluation of feedback |
requesting-code-review | Dispatching review agents |
writing-skills | TDD-for-documentation (meta-skill) |
Use /brainstorm, /write-plan, /execute-plan commands to invoke process skills directly.
Requires satisfaction before EACH action:
pre_commit_review:
scope: single_use # Must satisfy before EVERY commit
Auto-calculated conditions (e.g., branch size):
branch_size_limit:
type: dynamic
threshold: 400
Condition checks (e.g., protected branches):
protected_branch:
type: guard
branches: [main, master]
→ Advanced features details: See references/advanced-features.md
Good:
checklist:
- "Plan created via EnterPlanMode"
- "Atomic commits identified"
- "Tests written (TDD)"
Bad:
checklist:
- "Think about what you're going to do and write it down"
- "Various commit-related activities"
# .claude/requirements.yaml (team - committed)
requirements:
commit_plan:
enabled: true
scope: session
# .claude/requirements.local.yaml (personal - gitignored)
requirements:
commit_plan:
enabled: false # I opt-out
# Project inherits and extends global
version: "1.0"
inherit: true
requirements:
commit_plan:
checklist:
- "Project-specific checklist item"
→ More patterns: See references/configuration-patterns.md
req doctor # Full installation check
req verify # Quick verification
req doctor checks:
~/.claude/hooks/README-REQUIREMENTS-FRAMEWORK.md~/Tools/claude-requirements-framework/DEVELOPMENT.md~/Tools/claude-requirements-framework/sync.sh~/.claude/hooks/test_requirements.pyreferences/cli-reference.md - Complete CLI command documentationreferences/configuration-patterns.md - Common configuration patternsreferences/advanced-features.md - Auto-satisfy, dynamic, guardsreferences/troubleshooting.md - Error messages, debuggingexamples/project-requirements.yaml - Full project configexamples/custom-requirement.yaml - Custom requirement templateexamples/bash-command-trigger.yaml - Bash command patternsActivates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
Search, retrieve, and install Agent Skills from the prompts.chat registry using MCP tools. Use when the user asks to find skills, browse skill catalogs, install a skill for Claude, or extend Claude's capabilities with reusable AI agent components.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.