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.
npx agentmods add commands/hculap/better-code/tdd-startgit clone --depth 1 https://github.com/hculap/better-codeWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00008 | $0.00829 |
| Opus 5 | $0.00004 | $0.00415 |
| Sonnet 5 | $0.00002 | $0.00166 |
| Haiku 4.5 | $0.00001 | $0.00083 |
Grade A, and why
tdd-start 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.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TDD Mode Activation
Activate TDD Dev Mode for this session. From now on, apply strict Test-Driven Development practices to all coding requests.
Step 1: Detect Configuration First
Before creating the flag file, detect the test command:
- Check for settings file: Look for
.claude/tdd-dev.local.mdin project root - Auto-detect test command:
- If
package.jsonexists: Usenpm testor check for vitest/jest in devDependencies - If
pyproject.tomlorpytest.iniexists: Usepytest - If
go.modexists: Usego test ./... - Otherwise: Ask user for test command
- If
- Read strictness setting: Default to
strictif not configured
Step 2: Create TDD Mode Flag File
CRITICAL: Create the flag file that enables hook enforcement.
- Ensure
.claude/directory exists in project root - Create the TDD mode state file at
.claude/.tdd-mode-activewith the following JSON content:
{
"active": true,
"activatedAt": "[current ISO timestamp]",
"strictness": "strict",
"testCommand": "[detected or configured test command]"
}
This file signals to hooks that TDD mode is active and stores the detected configuration for use by commands and hooks.
Step 2.5: Initialize TDD Cycle State
CRITICAL: Create the cycle state file that tracks TDD phases.
Create .claude/.tdd-cycle-state with the following JSON content:
{
"phase": "red",
"testFilesWritten": [],
"testsRan": false,
"testsFailed": false
}
This file tracks the current TDD phase:
- red: Waiting for failing test to be written and run
- green: Test failed, now implement to make it pass
- refactor: Tests passing, safe to refactor
The hooks use this file to enforce the TDD cycle:
- PreToolUse (Write|Edit) blocks source file edits unless phase is "green" or "refactor"
- PostToolUse (Bash) detects test runs and transitions phases automatically
Step 3: Session State
Set the following TDD mode state:
- TDD Mode: ACTIVE (flag file created)
- Strictness: [detected or default: strict]
- Test Command: [detected command]
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.
- yesterday First seen · 107 lines · 8 tokens per session scan A 0b12c984cac8
tdd-start is a command published in the GitHub repository hculap/better-code (2 stars, last pushed 7mo ago), licensed MIT. It adds 8 tokens to every session and 829 once invoked, about $0.0000 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.
Other commands, from other repositories
create-skill
Create an AI skill from any source (URL, repo, PDF, video, notebook, etc.).
sync-config
Sync a scraping config's URLs against the live documentation site.
create_plan
Create detailed implementation plans through interactive research and iteration.
review-pr
Multi-agent PR review with four modes (review, re-review, self-review, address-feedback) - spawns parallel subagents, saves diff to /tmp for context efficiency, supports file exclusion patterns.
research_codebase
Document codebase as-is with thoughts directory for historical context.
iterate_plan
Iterate on existing implementation plans with thorough research and updates.