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.
git clone --depth 1 https://github.com/marcoemrich/mad-tdd-mob-ai-drivenWrote 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.
[](https://agentmods.dev/rules/marcoemrich/mad-tdd-mob-ai-driven/human-in-the-loop)<a href="https://agentmods.dev/rules/marcoemrich/mad-tdd-mob-ai-driven/human-in-the-loop"><img src="https://agentmods.dev/badge/rules/marcoemrich/mad-tdd-mob-ai-driven/human-in-the-loop/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.
<a href="https://agentmods.dev/rules/marcoemrich/mad-tdd-mob-ai-driven/human-in-the-loop"><img src="https://agentmods.dev/badge/rules/marcoemrich/mad-tdd-mob-ai-driven/human-in-the-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01445 |
| Opus 5 | $0.00000 | $0.00723 |
| Sonnet 5 | $0.00000 | $0.00289 |
| Haiku 4.5 | $0.00000 | $0.00145 |
Grade A, and why
human-in-the-loop 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.
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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Human-in-the-Loop TDD Rules
Description
These rules ensure the human stays engaged and can provide guidance at critical decision points during Test-Driven Development. The AI should pause and explicitly ask for user feedback in these specific situations.
Rule 1: End-of-Phase Confirmation
When to Apply
At the end of every TDD phase (Red, Green, or Refactor), before proceeding to the next phase or test.
What to Do
-
Stop after completing the current phase
-
Summarize what was just completed in this phase:
After Red Phase:
- Which test was activated
- Prediction made and whether it was correct
- Type of failure achieved (compilation/runtime error)
After Green Phase:
- Implementation approach taken (minimal code added)
- Confirmation that test now passes
- Any trade-offs or decisions made
After Refactor Phase:
- Refactorings attempted/completed:
- Naming changes made
- Mass calculations (before/after if applicable)
- Structural improvements
- Any refactoring opportunities that were rejected and why
-
Explicitly ask for permission to continue:
- After Red: "Red phase complete. Should I proceed to Green phase?"
- After Green: "Green phase complete. Should I proceed to Refactor phase?"
- After Refactor: "Refactor phase complete. Should I proceed to the next test?"
Why This Matters
- Human maintains full control - No phase proceeds without explicit approval
- Educational opportunity - Human can guide each individual step
- Prevents over-implementation - Each phase does only what's required
- Quality assurance - Human reviews every phase before proceeding
- Fine-grained control - Human can intervene at any point in the process
Examples
🔴 Red Phase Complete:
**Test Activated**: "should return sum for two numbers"
**Prediction**: Runtime assertion error (Expected: 3, Received: 1) ✅ Correct
**Result**: Test fails as expected with assertion error
Red phase complete. Should I proceed to Green phase?
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.
- 9d ago First seen · 160 lines · 0 tokens per session scan A 52b91f9c06e7
human-in-the-loop is a cursor rule published in the GitHub repository marcoemrich/mad-tdd-mob-ai-driven (33 stars, last pushed 10mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,445 tokens. 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.
Other cursor rules, from other repositories
manual-review.backend
A set of rules for verifying backend-only changes and changes that use different verification channels. It separates evidence the agent can collect from checks that require a person, a production authorization, or a business decision.
mcpnuke-tests
Test conventions for mcpnuke — enforces TDD workflow.
debug-issue
When the user reports a bug, an error, or unexpected behavior. Enforces four structured phases — reproduction, failing test, root cause isolation, fix and verify — to stop guess-and-check loops.
testing-discipline
TDD, BDD, and testing best practices — stack-agnostic.
common-testing
Testing requirements: 80% coverage, TDD workflow.
tdd
Test-driven development — red-green-refactor cycle.