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 zorost/AI-Engineering-Lab --skill prompt-suite-versioninggit clone --depth 1 https://github.com/zorost/AI-Engineering-LabWrote 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/zorost/ai-engineering-lab/prompt-suite-versioning)<a href="https://agentmods.dev/skills/zorost/ai-engineering-lab/prompt-suite-versioning"><img src="https://agentmods.dev/badge/skills/zorost/ai-engineering-lab/prompt-suite-versioning/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/zorost/ai-engineering-lab/prompt-suite-versioning"><img src="https://agentmods.dev/badge/skills/zorost/ai-engineering-lab/prompt-suite-versioning.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 | $0.00044 | $0.00954 |
| Opus 5 | $0.00022 | $0.00477 |
| Sonnet 5 | $0.00009 | $0.00191 |
| Haiku 4.5 | $0.00004 | $0.00095 |
Grade A, and why
prompt-suite-versioning 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 10d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prompt Suite Versioning
1 · Purpose
Turn prompt editing from taste into engineering: every version has a score, every change is one variable, every regression is reversible.
2 · When to use
- Improving, refactoring, or "cleaning up" any production prompt.
- Adding few-shot examples, schema constraints, or injection rules to a prompt.
Do not use to draft the very first prompt of a brand-new task, draft freely,
then enter this loop the moment a scorer exists (eval-first-development).
3 · Inputs
- A golden set and scorer (
eval-first-development) with a recorded baseline. - The current prompt, saved as a file (not living only in code or chat history).
- A versions log:
prompts/log.mdor the eval results directory.
4 · Procedure
- Save the current prompt as
v{N}.mdand record its baseline score. If it has no score, score it now, this is the number every later version must beat. - Pick one variable to change: one system-prompt line, one example, one schema field, one formatting rule. Write down the hypothesis: "adding the null rule will fix the missing-field hallucinations."
- Create
v{N+1}.mdwith exactly that change. No drive-by edits. - Run the scorer. Record version, score, and the one-line reason in the log.
- If the score dropped, revert without debate. If it rose, keep it. If it is flat, decide by cost: keep the cheaper or simpler version.
- Read the failures of the new version, not the score alone. Cluster them. If a failure class is new, add a case for it to the golden set with a commit message.
- Repeat from step 2 until the score clears the spec gate or three consecutive versions fail to move it. On the third flat version, STOP and report: the prompt axis is exhausted; the lever is elsewhere (retrieval, model, data).
- Keep every version file. Disk is cheap; archaeology is expensive.
5 · Anti-rationalization
| Excuse | Answer |
|---|---|
| "This change is obviously safe." | "Obviously safe" prompt edits are the leading cause of silent regressions. Score it. |
| "I'll batch these three tweaks to save time." | Then you will never know which one helped, and you will re-learn it next month. |
| "The eval is slow; I'll just eyeball this one." | Eyeballing is how v1 shipped with a carrier-suffix bug. Run the scorer. |
| "Deleting old versions keeps the repo clean." | The old version is the rollback. Keep it. |
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.
- 10d ago First seen · 85 lines · 44 tokens per session scan A b21a4695b06d
prompt-suite-versioning is a skill published in the GitHub repository zorost/AI-Engineering-Lab (302 stars, last pushed 23d ago), licensed MIT. It adds 44 tokens to every session and 954 once invoked, about $0.0002 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
ai-engineering-toolkit
6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.
llm-ops
LLM Operations -- RAG, embeddings, vector databases, fine-tuning, prompt engineering avancado, custos de LLM, evals de qualidade e arquiteturas de IA para producao.
llm-integration
LLM integration patterns for function calling, streaming responses, local inference with Ollama, and fine-tuning customization. Use when implementing tool use, SSE streaming, local model deployment, LoRA/QLoRA fine-tuning, or multi-provider LLM APIs.
ai-engineering-toolkit
AI Engineering Toolkit workflow skill. Use this skill when the user needs 6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching and the operator…
fine-tuning-os
Use this skill when conducting a professional LLM fine-tuning engagement in Zero-Data mode via the MCP server fine-tuning-os. Triggers on: prestation de fine-tuning LLM, livraison de modèle affiné, pipeline Zero-Data, orchestration des 64 outils fine-tuning-os, cycle de vie entraînement (préparation / données…
ai-engineering-toolkit
6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.