claude-shapeup: Skill for Claude Code

.claude/skills/behavior-tdd/SKILL.md

behavior-tdd is a skill for Claude Code from tiago-peixoto/claude-shapeup. It costs 92 tokens per session (1,559 once invoked), scanned A, original, MIT.

A test-first procedure for changing the written instructions that control an agent in this plugin. TDD means writing checks before making a change, then verifying that the change caused the intended behavior.

In plain words
What is it for?
Updating skill instructions, reference documents, behavioral rules, or shared scripts when a specific failing test, user complaint, or observed bad result justifies the change.
Why use it?
Editing instructions can change an agent's results even when ordinary automated tests still pass. This procedure requires evidence of the problem, a negative-control test, regression checks, and a clean versioned change.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: reads .claude/ paths; mentions CLAUDE.md.

This is tiago-peixoto/claude-shapeup's own configuration. It tells Claude Code how to work on claude-shapeup itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-shapeup configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is bash tests/behavioral/run-behavioral.sh <scenario>.

Part of the shapeup plugin — 4 skills, 3 hooks shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to tiago-peixoto/claude-shapeup. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/tiago-peixoto/claude-shapeup/main/.claude/skills/behavior-tdd/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/tiago-peixoto/claude-shapeup

Made for: Claude Code.

Or install shapeup, the plugin that ships this one along with the rest of its 4 skills, 3 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 behavior-tdd

README.md
[![agentmods](https://agentmods.dev/badge/skills/tiago-peixoto/claude-shapeup/behavior-tdd/github.svg)](https://agentmods.dev/skills/tiago-peixoto/claude-shapeup/behavior-tdd)
Your own site
<a href="https://agentmods.dev/skills/tiago-peixoto/claude-shapeup/behavior-tdd"><img src="https://agentmods.dev/badge/skills/tiago-peixoto/claude-shapeup/behavior-tdd/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for behavior-tdd

Your own site · 80×15
<a href="https://agentmods.dev/skills/tiago-peixoto/claude-shapeup/behavior-tdd"><img src="https://agentmods.dev/badge/skills/tiago-peixoto/claude-shapeup/behavior-tdd.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 92 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,559 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00092 $0.01559
Opus 5 $0.00046 $0.00779
Sonnet 5 $0.00018 $0.00312
Haiku 4.5 $0.00009 $0.00156

Measured 9d ago against content hash fea6c45647fc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

behavior-tdd 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 9d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (example.workflow.mjs), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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/skills/behavior-tdd/SKILL.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Behavior-change TDD (for this plugin)

This repo's agent behavior is defined by prompt text (skills/*/SKILL.md, references/*), not code. A prompt edit can silently change outcomes while every deterministic test still passes. This skill is the mandatory procedure for changing behavior here. The rule:

Deterministic checks for scaffolding. LLM-as-judge for outcomes. An ablation proves the prompt is what caused the change — not the model's defaults.

Read CLAUDE.md and CONTRIBUTING.md first; this skill operationalizes them.

When to use

  • Editing any skills/<phase>/SKILL.md, a references/*.md, or a shared script the skills call.
  • Adding or changing a behavioral invariant (a new commit rule, a renamed status, a reshaped artifact, a new "the agent must/never…").

Do NOT use for pure scaffolding (hook/script logic with exact-text assertions) — that's the unit layer only. And never edit speculatively: start from a concrete signal (a failing test, a user complaint, an observed bad run). No signal → no edit.

The loop (RED → edit → GREEN → regression → commit)

1. Establish the behavioral RED — and prove the test bites

  • Write or locate a criterion (tests/behavioral/criteria/<name>.md) and a scenario (tests/behavioral/scenarios/<name>.md). Criteria are binary pass/fail propositions that name the exact observable behavior; the judge must cite evidence (see judge-rubric.md).
  • Confirm it FAILs against the current prompt:
    bash tests/behavioral/run-behavioral.sh <scenario>
    
  • Ablation (negative control) — non-negotiable. Confirm the criterion needs the prompt:
    bash tests/behavioral/run-behavioral.sh --ablate <scenario>   # must print "OK: fails without the skill"
    
    If ablation prints CONTAMINATED, the criterion passes with a neutral prompt — it's measuring the model (or the scenario leaks the answer), not your SKILL. Tighten the criterion until it hinges on prompt-specific mechanism, then re-ablate. (This is the "confirm-the-red" / scaffold-removal discipline; a passing ablation is a failed test.)

Read the full file on GitHub · 117 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 9d ago First seen · 117 lines · 92 tokens per session scan A fea6c45647fc

Subscribe to this mod's changes

behavior-tdd is a skill published in the GitHub repository tiago-peixoto/claude-shapeup (22 stars, last pushed 2mo ago), licensed MIT. It adds 92 tokens to every session and 1,559 once invoked, about $0.0005 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.