refactor-safe

A command for restructuring internal code while keeping its public API—the functions, classes, signatures, and return formats used by other code—unchanged.

In plain words
What is it for?
Use it to split up complex functions, remove duplication, improve naming, separate concerns, and clean unused or unreachable code after the existing interface has been recorded.
Why use it?
It addresses long functions, repeated logic, confusing names, deep nesting, mixed responsibilities, and dead code without intentionally changing how callers use the code.

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/benkapner/claude-code-basecamp/refactor-safe
Clone the repo
git clone --depth 1 https://github.com/Benkapner/claude-code-basecamp

Made for: Claude Code.

Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 708 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00029 $0.00708
Opus 5 $0.00015 $0.00354
Sonnet 5 $0.00006 $0.00142
Haiku 4.5 $0.00003 $0.00071

Measured yesterday against content hash b7794daee76e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

refactor-safe 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 yesterday.

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.

.claude/commands/refactor-safe.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Safe Refactor

At this point, you've reviewed the code, tested it, and documented it. If it still has flagged structural problems (like a 65-line handler doing validation, inference, logging, and response formatting in one block), this is the command to fix it.

Refactoring with AI is risky because Claude tends to "improve" things you didn't ask it to touch. This command has an explicit constraint: the public API must not change. It focuses on internal structure only.

Instructions

1. Identify Scope

  • Read the target file(s) specified in $ARGUMENTS
  • Identify the public API: exported functions, class interfaces, function signatures, return types
  • Document the public API before making any changes

2. Identify Refactoring Opportunities

Look for these internal problems:

  • Long functions (50+ lines doing multiple things)
  • Duplicated logic (same pattern in multiple places)
  • Deep nesting (3+ levels of if/for/try)
  • Poor naming (single-letter variables, misleading names)
  • Mixed concerns (validation + business logic + formatting in one function)
  • Dead code (unused imports, unreachable branches)

3. Apply Refactoring

For each change:

  • Extract helper functions for distinct responsibilities
  • Simplify nested conditionals (early returns, guard clauses)
  • Rename variables/functions for clarity
  • Remove dead code
  • Consolidate duplicated logic

4. Verify

After refactoring:

  • Confirm the public API is unchanged — same function signatures, same return types, same behavior
  • Run tests if they exist (uv run python -m pytest)
  • If no tests exist, warn the user

5. Present Changes

Show what was refactored:

Safe Refactor Summary
=====================
File: scripts/summarize_people.py

PUBLIC API: Unchanged
  - main(argv) — same signature, same behavior
  - compute_person_stats(issues) — same signature, same return type

INTERNAL CHANGES:
  - Extracted _parse_date() from compute_person_stats (was inline, 8 lines)
  - Extracted _count_collaborators() from compute_person_stats (was inline, 12 lines)
  - Renamed 'ps' → 'person_summary' in main loop
  - Removed unused import: 'json'

Tests: 70 passed, 0 failed

Read the full file on GitHub · 80 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. yesterday First seen · 80 lines · 29 tokens per session scan A b7794daee76e

Subscribe to this mod's changes

refactor-safe is a command published in the GitHub repository Benkapner/claude-code-basecamp (16 stars, last pushed 12d ago), licensed MIT. It adds 29 tokens to every session and 708 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.