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 skills add SebastienDegodez/copilot-instructions --skill red-synthesize-greengit clone --depth 1 https://github.com/SebastienDegodez/copilot-instructionsWrote 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/skills/sebastiendegodez/copilot-instructions/red-synthesize-green)<a href="https://agentmods.dev/skills/sebastiendegodez/copilot-instructions/red-synthesize-green"><img src="https://agentmods.dev/badge/skills/sebastiendegodez/copilot-instructions/red-synthesize-green/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/skills/sebastiendegodez/copilot-instructions/red-synthesize-green"><img src="https://agentmods.dev/badge/skills/sebastiendegodez/copilot-instructions/red-synthesize-green.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00021 | $0.01223 |
| Opus 5 | $0.00010 | $0.00611 |
| Sonnet 5 | $0.00004 | $0.00245 |
| Haiku 4.5 | $0.00002 | $0.00122 |
Grade A, and why
red-synthesize-green 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 11d 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RED → SYNTHESIZE GREEN (AI TDD Cycle)
Overview
2-step cycle replacing traditional 3-step TDD. Optimized for AI synthesis.
- Traditional (3 steps): RED → green (dirty) → Refactor
- AI-Optimized (2 steps): RED (behavior failure) → SYNTHESIZE GREEN (clean synthesis)
Architectural guidance is mandatory between steps.
Hard rule: No implementation code before RED is a clean behavior failure.
Step 1: RED (Behavior Failure Only)
Write the failing test. Run it.
- Compilation errors = wishful thinking phase → implement stubs/empty returns to compile, rerun
- Assertion/behavior failure = RED ✓ → proceed to Step 2
- Never treat compilation errors as RED
Programming by Wishful Thinking: When your test won't compile, you're discovering the API you need. Stub just enough to compile, then confirm the test fails on behavior.
Between Steps: Architectural Guidance (MANDATORY)
Hard rule: This step is not skippable. Do not proceed to SYNTHESIZE GREEN without completing it.
Developer must review and explicitly validate the test before continuing. AI pauses here and waits for developer confirmation that the test correctly captures the intended behavior.
Orient design before synthesis:
- Which pattern? (specification, factory, builder)
- Which layer owns the logic?
- Immutability, return values vs mutations?
Step 2: SYNTHESIZE GREEN (Clean Synthesis)
Implement complete, clean, production-ready solution in one shot.
- Follows all architectural rules and coding standards
- No dirty-then-refactor — synthesize properly from the start
- Idiomatic code, domain semantics, SOLID principles
- If test was misunderstood → revise test, restart from RED
No iteration after SYNTHESIZE GREEN unless RED was wrong or architectural guidance changed.
Common Rationalizations
| Excuse | Reality |
|---|---|
| "Compilation error IS red" | No. Compilation = wishful thinking. RED = behavior failure. |
| "I'll write dirty code then refactor" | That's 3-step TDD. SYNTHESIZE GREEN produces clean code. |
| "I can skip RED, I know it'll fail" | Run it. RED proves your test catches real failures. |
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.
- 11d ago First seen · 126 lines · 21 tokens per session scan A 14752fcdabc7
red-synthesize-green is a skill published in the GitHub repository SebastienDegodez/copilot-instructions (193 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 21 tokens to every session and 1,223 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.
Other skills, from other repositories
testing
TDD/BDD testing principles. Use when writing tests, reviewing test coverage, setting up testing, or discussing test strategy and test architecture.
workflow-analysis
Workflow d'Analyse Obligatoire. Use when working with workflow analysis.
outside-in-tdd
Use when an approved scenario, Gherkin example, worked example, or expected result has to become working software through outside-in / double-loop TDD -- start from an acceptance or application-boundary test, get a trustworthy RED before implementation, let domain logic emerge only from failing behavior, and drive one…
quality-gates-dotnet
Use when the active repository is a .NET solution (.sln / .csproj present) and the software-engineer must produce falsifiable evidence for the quality gates. Most gates are captured at the end of the COMMIT phase; the G10 RED capture is taken at RED, before any production code, and cannot be reconstructed later.…
quality-gates-evidence-contract
Use when producing or verifying the structured evidence log that attests quality gates (tests, build, mutation, commits, RED/GREEN integrity). Tech-agnostic schema. Loaded by software-engineer (writer) at COMMIT phase and by quality-gates-lens (reader) during review.
tdd-workflow
Test-Driven Development workflow principles. RED-GREEN-REFACTOR cycle.