tcr

tcr is a skill for Claude Code, Codex from xpepper/tcr-skill. It costs 164 tokens per session (2,428 once invoked), scanned A, original, MIT.

A guide to TCR, TDD, and TCRDD development workflows. TDD means writing a failing test first; TCR means committing passing work and reverting failing work.

In plain words
What is it for?
Use it to follow the Red, Green, Refactor cycle with git-gamble, the tool described for automating test-and-commit-or-revert workflows.
Why use it?
It enforces small development steps while keeping each change tied to a test and quickly removing work that fails verification.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

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 skills/xpepper/tcr-skill/tcr
Any agent
npx skills add xpepper/tcr-skill --skill tcr
Clone the repo
git clone --depth 1 https://github.com/xpepper/tcr-skill

Made for: Claude Code, Codex.

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 tcr

README.md
[![agentmods](https://agentmods.dev/badge/skills/xpepper/tcr-skill/tcr.svg)](https://agentmods.dev/skills/xpepper/tcr-skill/tcr)
Your own site
<a href="https://agentmods.dev/skills/xpepper/tcr-skill/tcr"><img src="https://agentmods.dev/badge/skills/xpepper/tcr-skill/tcr.svg" alt="Measured on agentmods" height="20"></a>
Per session 164 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,428 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 5.1 $0.00164 $0.02428
Opus 5 $0.00082 $0.01214
Sonnet 5 $0.00033 $0.00486
Haiku 4.5 $0.00016 $0.00243

Measured 6d ago against content hash 90cb309af69a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

tcr 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 6d 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.

tcr/SKILL.md · 227 lines

How it starts

The opening of the file, as written. The whole thing — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TCRDD — TCR + TDD

TCRDD = TCR (test && commit || revert) + TDD (Test Driven Development).

It blends two disciplines so that:

  • You always develop the right thing (TDD's guarantee: one failing test drives each step)
  • You're forced to take baby steps, because reverting wipes out wrong work fast (TCR's guarantee)

The tool that automates this workflow is git-gamble. Run git-gamble --help for usage details.

TDD Foundation (the part TCR is built on)

TCRDD assumes you follow the Red / Green / Refactor micro-cycle from TDD:

  1. Red — write exactly one failing test (just enough to fail — compilation failures count)
  2. Green — write the minimum production code to make it pass, nothing more
  3. Refactor — clean up structure while keeping all tests green

Each phase maps directly to a git gamble flag. The discipline of one test at a time and minimum code to pass is what makes the gamble meaningful — if you write too much, a revert erases more than intended.

Strict TCR Rule

In strict TCR, git gamble is the test run for the current cycle.

  • Do not run cargo test, npm test, or any other standalone test command before git gamble inside a Red, Green, or Refactor cycle
  • If you already know whether the tests will pass or fail, there is no gamble
  • The only acceptable test execution inside a cycle is git gamble --red|--green|--refactor -- <test command>
  • Standalone test runs are allowed only outside the cycle, such as end-of-session verification or after a revert when preparing the next attempt
  • git gamble commits or reverts the entire dirty worktree, not just the file you meant to touch
  • Start each TCR cycle from a clean worktree, or isolate unrelated edits before gambling

This is non-negotiable when the user asks for strict TCR/TCRDD.


The Three Phases

TCRDD cycles through three phases. Each phase ends in either a commit (success) or a revert (failure → retry).

Read the full file on GitHub · 227 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. 6d ago First seen · 227 lines · 164 tokens per session scan A 90cb309af69a

Subscribe to this mod's changes

tcr is a skill published in the GitHub repository xpepper/tcr-skill (2 stars, last pushed 5mo ago), licensed MIT. It adds 164 tokens to every session and 2,428 once invoked, about $0.0008 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 2026-08-31.