bkit

bkit is a command for coding agents from ww-w-ai/bkit-claude-code. It costs 112 tokens per session (4,889 once invoked), scanned A, original, Apache-2.0.

A command toolkit for Claude Code that guides feature work through planning, design, implementation, testing, analysis, reporting and archiving. It uses PDCA, a repeatable plan-do-check-act workflow, and provides 14 user-invoked skills plus 9 automatic phase skills.

In plain words
What is it for?
Use it to plan and design features, approve implementation scope, run gap analysis, test changes at five levels, improve results iteratively, and produce or archive project reports.
Why use it?
It gives development work a defined sequence and checkpoints, helping reduce missed requirements and unreviewed scope changes.

Command

Part of the bkit plugin — 19 skills, 2 commands, 35 agents, 21 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/ww-w-ai/bkit-claude-code/bkit
Clone the repo
git clone --depth 1 https://github.com/ww-w-ai/bkit-claude-code

Or install bkit, the plugin that ships this one along with the rest of its 19 skills, 2 commands, 35 agents, 21 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for bkit

README.md
[![agentmods](https://agentmods.dev/badge/commands/ww-w-ai/bkit-claude-code/bkit.svg)](https://agentmods.dev/commands/ww-w-ai/bkit-claude-code/bkit)
Your own site
<a href="https://agentmods.dev/commands/ww-w-ai/bkit-claude-code/bkit"><img src="https://agentmods.dev/badge/commands/ww-w-ai/bkit-claude-code/bkit.svg" alt="Measured on agentmods" height="20"></a>
Per session 112 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,889 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00112 $0.04889
Opus 5 $0.00056 $0.02445
Sonnet 5 $0.00022 $0.00978
Haiku 4.5 $0.00011 $0.00489

Measured 4d ago against content hash 977d2f666a56, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bkit scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

| Clean Architecture 4-Layer | Automatic (CI-enforced) | `lib/domain/` 11 modules (6 ports + 4 guards + 1 rules) / `lib/infra/` adapters / `lib/cc-regression/` application / hooks+scripts presentation. `scripts/check-dom
commands/bkit.md · 312 lines

How it starts

The opening of the file, as written. The whole thing — 312 lines — stays where its author put it; the contents beside it link to each section on GitHub.

bkit Functions

Show all available bkit functions (Skills autocomplete workaround)

Display the following help message:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  bkit - AI Native Development Toolkit (Claude Code Edition)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

PDCA (Document-Driven Development)
  /pdca pm <feature>         PM Agent Team analysis (43 frameworks, v2.0.3)
  /pdca plan <feature>       Plan with Checkpoints 1-2 (requirements + questions)
  /pdca design <feature>     Design with Checkpoint 3 (3 architecture options)
  /pdca do <feature>         Implement with Checkpoint 4 (scope approval)
  /pdca do <feature> --scope module-N   Multi-session incremental (v2.0.5)
  /pdca analyze <feature>    Gap analysis with Checkpoint 5 (fix strategy)
  /pdca qa <feature>         QA phase testing (L1-L5 test levels, v2.1.1)
  /pdca iterate <feature>    Auto-improvement iteration
  /pdca report <feature>     Generate completion report
  /pdca archive <feature>    Archive completed PDCA documents
  /pdca cleanup              Cleanup archived features from status
  /pdca team <feature>       Start PDCA Team Mode (Agent Teams)
  /pdca team status          Show Team status
  /pdca team cleanup         Cleanup Team resources
  /pdca status               Show current PDCA status
  /pdca next                 Guide to next step

Sprint Management (v2.1.13 — meta-container for 1+ features)
  /sprint init <id> [--name <name>] [--features a,b] [--trust L0-L4]
                             Create sprint (8-phase: prd→plan→design→do→iterate→qa→report→archived)
  /sprint start <id> [--trust L0-L4]   Auto-run up to Trust Level scope (L4=full-auto)
  /sprint status <id>        Current phase + status + auto-pause triggers
  /sprint list               All sprints overview
  /sprint watch <id>         Live snapshot + 4 triggers + 3 matrices (dataFlow/cumulative/feature-phase)
  /sprint phase <id> --to <phase>      Manual phase transition
  /sprint iterate <id>       matchRate 100% loop (max 5 iterations)
  /sprint qa <id> [--feature <name>]   7-Layer S1 dataFlow integrity check
  /sprint report <id>        Generate sprint completion report
  /sprint archive <id>       Move to archived (terminal state)
  /sprint pause <id>         Stop auto-run (4 triggers also auto-pause)
  /sprint resume <id>        Resume from paused state
  /sprint fork <id> --new <newId>      Carry incomplete features → new sprint
  /sprint feature <id> --action list|add|remove --feature <name>
  /sprint trust <id> --to <L0-L4> [--reason "..."] [--force]
                             Mutate sprint trust level (persistent, audit-logged) — v2.1.18 (Issue #101)
  /sprint help               Sprint Management help
  Korean guide: docs/06-guide/sprint-management.guide.md
  Migration:    docs/06-guide/sprint-migration.guide.md (PDCA↔Sprint, orthogonal coexistence)

Project Initialization
  /starter init <name>       Static web project (HTML/CSS/Next.js)
  /dynamic init <name>       Fullstack app (bkend.ai BaaS)
  /enterprise init <name>    Enterprise system (K8s/Terraform)

Development Pipeline
  /development-pipeline start    Start pipeline
  /development-pipeline next     Proceed to next phase
  /development-pipeline status   Check current phase

Quality Management
  /code-review <path>        Code review
  /qa-phase <feature>        QA phase testing (L1-L5 test levels)
  /zero-script-qa            Start Zero Script QA

Learning
  /claude-code-learning          Learn Claude Code
  /claude-code-learning setup    Analyze current project setup

Code Quality (v1.6.0)
  /simplify                  Review changed code for reuse, quality, efficiency
  /batch                     Process multiple features in batch mode

Memory & Clipboard (v1.5.6)
  /memory                    Manage Claude auto-memory (view/edit entries)
  /copy                      Copy code blocks to clipboard (interactive picker)

Output Styles (v1.5.3)
  /output-style              Select response style
  /output-style-setup        Install bkit styles to .claude/
  Available: bkit-learning, bkit-pdca-guide, bkit-enterprise, bkit-pdca-enterprise

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  Note: These functions don't have autocomplete in CLI.
  Type the command directly (e.g., /pdca plan login)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Read the full file on GitHub · 312 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 312 lines · 112 tokens per session scan A 977d2f666a56

Subscribe to this mod's changes

bkit is a command published in the GitHub repository ww-w-ai/bkit-claude-code (595 stars, last pushed 17d ago), licensed Apache-2.0. It adds 112 tokens to every session and 4,889 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.