refactor

A command that writes a refactoring plan for changing code structure while keeping its existing behavior.

In plain words
What is it for?
Use it to prepare a plan from a GitHub issue, including the required labels and a markdown specification under the project's cache directory.
Why use it?
It guides implementation with checks for issue labels, isolated worktree paths, and a required plan-file location.

Command for Claude Code

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/jayminwest/kotadb/refactor
Clone the repo
git clone --depth 1 https://github.com/jayminwest/kotadb

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,444 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 86% copy Near-identical to another mod 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 $0.00000 $0.01444
Opus 5 $0.00000 $0.00722
Sonnet 5 $0.00000 $0.00289
Haiku 4.5 $0.00000 $0.00144

Measured 2d ago against content hash e74fc16b1dea, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

refactor scanned grade A with 0 findings 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 2d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

86% identical to bug — 104 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/commands/issues/refactor.md · 137 lines

How it starts

The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/refactor

Template Category: Action

Author a refactoring plan for the code improvement described in $ARGUMENTS (issue metadata JSON). The plan must guide the implementor through structural changes while preserving existing behavior.

CRITICAL - Worktree Path Handling:

  • You are executing in an isolated git worktree directory
  • Your CWD is the worktree root (e.g., /project/trees/refactor-123-abc12345)
  • ALL file paths in Write, Edit, Read tools MUST be relative to CWD
  • ✅ Correct: .claude/.cache/specs/refactor-123-plan.md
  • ❌ Wrong: /project/trees/refactor-123-abc12345/.claude/.cache/specs/refactor-123-plan.md
  • Using absolute paths will cause git staging failures and commit errors

Instructions

  • Verify issue labels first: Run gh issue view <issue-number> --json labels to ensure the issue has labels from all four categories (component, priority, effort, status). If labels are missing, apply them before proceeding.
  • Create a new markdown plan under .claude/.cache/specs/ named refactor-<issue-number>-<slug>.md (e.g., .claude/.cache/specs/refactor-2350-extract-auth-logic.md).
  • Build <slug> from the issue title using 3–6 lowercase, hyphenated words (alphanumeric only).
  • Follow the format exactly so orchestrators can parse sections reliably.
  • MANDATORY: Use KotaDB MCP tools for impact analysis: Use mcp__kotadb-staging__search_dependencies (minimum depth=2) and mcp__kotadb-staging__analyze_change_impact before refactoring (see .claude/commands/docs/kotadb-agent-usage.md for patterns)
  • Analyze existing code structure to understand current dependencies, interfaces, and behavior before proposing changes.
  • Investigate impacted modules in src/**, tests/**, and any infrastructure that depends on the code being refactored.
  • Capture all impacted files (and any new assets) in the dedicated section so implementors have clear scope boundaries.
  • Reference the repo git flow: work from refactor/<issue-number>-<slug> off develop, with develop promoted to main on release.
  • Integrate /anti-mock expectations: update test coverage to validate behavior preservation, use real database connections, and document any temporary test skips.
  • Ensure the plan's closing tasks rerun validation and push the branch so a PR can be raised (PR titles must end with the issue number, e.g. refactor: extract auth logic (#210)).
  • Consult .claude/commands/docs/conditional_docs/app.md or .claude/commands/docs/conditional_docs/automation.md and read only the documentation whose conditions align with the refactoring scope.
  • When the refactoring introduces new patterns or documentation, add or update the relevant conditional entry so future agents can discover it quickly.

Read the full file on GitHub · 137 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. 2d ago First seen · 137 lines · 0 tokens per session scan A e74fc16b1dea

Subscribe to this mod's changes

refactor is a command published in the GitHub repository jayminwest/kotadb (102 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,444 tokens. A static security scan graded it A with 0 findings. It is 86% identical to bug, differing in 104 lines, and is treated as a copy.