By dokipen
Issue-driven, multi-agent development workflow with git worktrees, structured phases, and specialist delegation
Create a ticket in the project's issue tracker. Collects title, description, and acceptance criteria, then creates the ticket and stops — no implementation.
Coordinate implementation work through structured phases with specialist agents. All work is tracked via a ticket provider (GitHub Issues or issues-api).
When delegating to any agent, include all of the following:
> This section applies **only** when the ticket has the `human-activity` label. The standard implementation phases (2–7) are skipped entirely. No source code is changed — the lead presents a step-by-step walkthrough and guides the human through completing the required manual tasks.
> **Skip this phase entirely** if the ticket has the `human-activity` label. The human-activity workflow makes no code changes and requires no worktree. After completing step 7 (claim), proceed directly to the [Human Activity Workflow](human-activity-workflow.md). Do not set `WORKTREE_DIR` or `BRANCH` — leave them unset so Phase 4 cleanup is skipped automatically.
Expert in Claude Code configuration, agent prompts, and skills. Use when updating agents, creating new skills, or optimizing AI workflows.
Review code for quality, best practices, and maintainability. Use for PR reviews and code audits.
Performance optimization specialist. Use for build size, response time, memory usage, startup time, and performance profiling.
Security review specialist. Use for input validation, data handling, dependency audits, and security best practices.
Test engineer for running tests and analyzing results. Use for bug reproduction, test coverage analysis, and integration testing.
Evaluate, benchmark, and iteratively improve an existing Claude Code skill using Anthropic's skill-creator tooling. Use when a skill draft already exists and the user wants to measure trigger accuracy, run quantitative evals, review output quality, or optimize the skill description. Do NOT use this for creating a skill from scratch — use create-skill for that.
Ticket provider abstraction layer that reads CLAUDE.md configuration and dispatches to the correct backend (GitHub Issues or issues microservice). Use when performing ticket operations.
Understanding agentd's hub-based architecture for managing AI agent sessions. Use when working with agent sessions, profiles, or the agentd API through the agent-hub.
Create a pull request for the current branch. Use when code is ready for review and needs a PR created with pre-flight verification.
Bootstrap a new Claude Code skill or command for the project. Use when creating a new skill from scratch — setting up the directory, frontmatter, required sections, and initial SKILL.md content. For evaluating, benchmarking, or iteratively improving an existing skill, use the skill-iterate skill instead.
Uses power tools
Uses Bash, Write, or Edit tools
No model invocation
Executes directly as bash, bypassing the AI model
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
An issue-driven, multi-agent development workflow plugin for Claude Code.
Claude Cadence provides a structured development workflow with git worktrees, phased implementation, specialist agent delegation, and GitHub issue tracking — all stack-agnostic and customizable per project.
This repo also includes a thin Codex compatibility layer that reuses the existing Claude prompts, scripts, and CLAUDE.md as the source of truth. See docs/codex-compatibility.md.
Codex support is packaged as a compatibility layer rather than a second copy of the workflow docs:
.codex-plugin/plugin.json registers the plugin for Codexcodex/skills/ contains generated thin wrappers for commands, shared skills, and specialist rolesscripts/sync-codex.py regenerates those wrappers from the canonical Claude source filesWhen updating Cadence workflows, edit the existing Claude files first and then run:
python3 scripts/sync-codex.py
git clone https://github.com/dokipen/claude-cadence.git
claude --plugin-dir ./claude-cadence
To always load the plugin, add --plugin-dir to your shell alias or config:
claude() { command claude --plugin-dir /path/to/claude-cadence "$@"; }
# Add the marketplace (once)
/plugin marketplace add dokipen/claude-cadence
# Install the plugin
/plugin install claude-cadence
cd /path/to/claude-cadence && git pull
Run claude plugin update cadence@claude-cadence to update, then restart Claude Code.
| Command | Purpose |
|---|---|
/lead | Coordinate implementation through 8 structured phases (0–7) |
/refine or /refine 123 | Refine issues to quality standards |
| Skill | Purpose |
|---|---|
new-work | Create a git worktree for isolated development |
create-pr | Create a PR with pre-flight verification |
create-skill | Bootstrap new Claude Code skills |
skill-iterate | Evaluate, benchmark, and iteratively improve an existing skill using Anthropic's skill-creator |
project-ops | Shared worktree management utilities |
| Agent | Role |
|---|---|
code-reviewer | PR reviews, code quality, best practices |
tester | Test execution, bug reproduction, coverage |
security-engineer | Security audits, dependency checks |
performance-engineer | Performance profiling, optimization |
claude-specialist | Claude Code configuration, agent/skill design |
ticket-refiner | Issue quality assurance |
Claude Cadence agents read your project's CLAUDE.md to discover stack-specific commands. Add a ## Verification section:
## Verification
flutter analyze && flutter test
or:
## Verification
go vet ./... && go test ./...
The agents and scripts will use this command automatically.
~/.claude/cadence.json)You can set provider configuration in ~/.claude/cadence.json to avoid repeating it in every project's CLAUDE.md. Configuration precedence (highest to lowest):
.repos["owner/repo"]) — overrides CLAUDE.mdCLAUDE.md — project-level config checked into the repoCLAUDE.md has no provider configprovider: githubGlobal default — applies to any repo that has no ## Ticket Provider section in CLAUDE.md:
{
"provider": "issues-api",
"project_id": "my-default-project",
"api_url": "https://issues.example.com"
}
Per-repo override — overrides CLAUDE.md for a specific repo, identified by owner/repo slug derived from git remote get-url origin:
{
"repos": {
"owner/repo": {
"provider": "issues-api",
"project_id": "my-project"
}
}
}
Both can be combined in one file. Per-repo entries take precedence over CLAUDE.md; global defaults only apply when CLAUDE.md has no provider config. api_url is not overridable per-repo — it belongs in CLAUDE.md or the global default. Use ISSUES_API_URL for QA/local api_url overrides.
Add domain-specific agents to your project's .claude/agents/ directory. They layer on top of the plugin's core agents:
your-project/.claude/
├── agents/
│ ├── designer.md # Your custom agent
│ ├── game-mechanics-engineer.md # Your custom agent
│ └── tester.md # Overrides plugin's generic tester
Same-name agents in your project override the plugin version.
npx claudepluginhub dokipen/claude-cadence --plugin cadencePortable Development System — AI-assisted development methodology with skills for consistency and agents for scale.
Implementation planning, execution, and PR creation workflows with multi-agent collaboration
AI-powered agents for specialized development tasks
Complete project development toolkit: 23 agents, 23 slash commands, 29 lifecycle hooks, and 69 reusable skills for Claude Code workflows
AI-Driven Engineering workflow commands for managing issues, tasks, implementation, and PRs.
Software engineering workflows with skills for planning, implementation, quality review, and structured thinking, plus a suite of specialist agents