Enforce an end-to-end development workflow from design validation through PR merge, with integrated codebase audits for DRY/YAGNI violations and cross-task review gates.
Read-only auditor that scans source code for DRY, YAGNI, simplicity, refactoring, and consistency issues with severity-tagged findings
Reviews a design doc before planning begins
Reviews an entire feature implementation for cross-task issues
Writes implementation plans from design docs with structured plan.json and task files
Reviews an implementation plan before execution begins
Use when the user wants to view, change, or reset their persistent preferences, or when triggered by "/caliper-settings", "my settings", "change defaults".
Use when asked to audit a codebase, find DRY/YAGNI/complexity issues repo-wide, or perform periodic code quality review
Use when a design doc has been written and before draft-plan is dispatched
Use when creating features, building components, adding functionality, or modifying behavior - before any creative or implementation work begins
Use when you have a spec or requirements for a multi-step task, before touching code
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
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.
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
Measure twice, cut once.
A Claude Code plugin that turns your goal into a PR with as little friction as possible. Every step is reviewed with a fresh context subagent. You get a design-reviewed, plan-validated, test-driven PR — with three human decisions.
Many claude workflows are either improperly context engineered, overly complicated, or don't understand how to effectively use AI agents. This tool tries to be different. We don't lock Claude in a box, but we have fresh agents check work as it goes to ensure perfection.
Install claude-caliper. Describe what you want to build. Walk away.
The plugin installs 11 skills that fire automatically at the right moment, enforcing a full development workflow: design before plan, plan before code, test before merge. You make three decisions — approve the design, review the PR, and confirm the merge — and everything between runs as a chain of fresh subagents with zero manual handoffs.
You say:
"Add rate limiting middleware with per-route config and 429 responses with retry-after headers"
Then the pipeline runs:
| Step | What happens | Who |
|---|---|---|
| 1 | Claude challenges your assumptions, proposes 2-3 approaches with trade-offs | You + Claude |
| 2 | You approve a design | You |
| 3 | Design review validates the doc against a 9-point checklist | Fresh subagent |
| 4 | Draft plan writes tasks with exact file paths, TDD steps, verification commands | Fresh subagent |
| 5 | Plan review catches vague steps, missing paths, design-plan drift | Fresh subagent |
| 6 | Orchestrator spawns one agent team teammate per task (parallel within phase), each running RED-GREEN-REFACTOR | Agent team teammates |
| 7 | Per-task reviewer checks each task; fixes sent back to the original implementer via messaging | Fresh subagents |
| 8 | Implementation review does a cross-task holistic pass | Fresh subagent |
| 9 | Create PR opens a PR | Automated |
| 10 | You review the PR and run /pr-review | You |
| 11 | PR reviewer reads the diff cold before any external feedback | Subagent |
| 12 | Fixes applied, feedback addressed | Automated |
| 13 | You run /pr-merge — squash merge, branch cleaned up | You |
Steps 3-9 run without any input from you.
%%{init: {'flowchart': {'nodeSpacing': 15, 'rankSpacing': 25}}}%%
flowchart TD
A([You: describe a feature]) --> B[design]
B --> BR[design-review]
BR --> C([You: approve design])
C --> D[draft-plan]
D --> DR[plan-review]
DR --> T1[Task 1]
DR --> T2[Task 2]
DR --> TN[Task N]
T1 --> R1[Review]
T2 --> R2[Review]
TN --> RN[Review]
R1 & R2 & RN --> IR[implementation-review]
IR --> S[pr-create]
S --> M([You: review PR])
M --> MR[pr-review]
MR --> MP([You: merge])
MP --> MG[pr-merge]
style A fill:#3b82f6,stroke:#2563eb,color:#fff
style B fill:#8b5cf6,stroke:#7c3aed,color:#fff
style BR fill:#eab308,stroke:#ca8a04,color:#000
style C fill:#3b82f6,stroke:#2563eb,color:#fff
style D fill:#8b5cf6,stroke:#7c3aed,color:#fff
style DR fill:#eab308,stroke:#ca8a04,color:#000
style T1 fill:#f97316,stroke:#ea580c,color:#fff
style T2 fill:#f97316,stroke:#ea580c,color:#fff
style TN fill:#f97316,stroke:#ea580c,color:#fff
style R1 fill:#eab308,stroke:#ca8a04,color:#000
style R2 fill:#eab308,stroke:#ca8a04,color:#000
style RN fill:#eab308,stroke:#ca8a04,color:#000
style IR fill:#eab308,stroke:#ca8a04,color:#000
style S fill:#22c55e,stroke:#16a34a,color:#fff
style M fill:#3b82f6,stroke:#2563eb,color:#fff
style MR fill:#eab308,stroke:#ca8a04,color:#000
style MP fill:#3b82f6,stroke:#2563eb,color:#fff
style MG fill:#22c55e,stroke:#16a34a,color:#fff
Blue = human decisions (3 total) · Purple = creative work · Orange = TDD implementation · Yellow = review gates · Green = shipping
/plugin marketplace add nikhilsitaram/claude-caliper
/plugin install claude-caliper@claude-caliper
Restart Claude Code.
Start a new session and describe something you want to build. The design skill fires automatically — no slash commands needed.
If Claude immediately starts discussing approaches and trade-offs instead of writing code, the plugin is working.
npx claudepluginhub nikhilsitaram/claude-caliper --plugin claude-caliper-toolingA curated set of skills for each stage of development — propose, spec, design, plan, implement, ship.
Persona-driven AI development team: orchestrator, team agents, review agents, skills, slash commands, and advisory hooks for Claude Code
Development workflow automation including feature development, code quality, and PR management
Focused agentic engineering workflow: design-doc, spec, plan, implement, tdd, refactor, review, address-pr-feedback, browser-verify, explain-visually, compress, branch, and commit.
AI-powered development workflow automation - Phase-based planning, implementation orchestration, preflight code quality checks with security scanning, ship-it workflow, and development principles generator for CLAUDE.md
Helder's personal SDLC toolbelt for AI coding agents — from PRD to ship. Bundles the tracer-bullet workflow alongside TDD, code review, audits, and shipping skills.