Borrowing it
Nothing to install: this file belongs to brevdev/workshop-build-an-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/brevdev/workshop-build-an-agent/main/.agents/skills/module-4/SKILL.mdgit clone --depth 1 https://github.com/brevdev/workshop-build-an-agentWrote 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/brevdev/workshop-build-an-agent/module-4)<a href="https://agentmods.dev/skills/brevdev/workshop-build-an-agent/module-4"><img src="https://agentmods.dev/badge/skills/brevdev/workshop-build-an-agent/module-4/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/brevdev/workshop-build-an-agent/module-4"><img src="https://agentmods.dev/badge/skills/brevdev/workshop-build-an-agent/module-4.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 76 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
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 | $0.00265 | $0.03420 |
| Opus 5.5 | $0.00106 | $0.01368 |
| Sonnet 5.5 | $0.00053 | $0.00684 |
| Haiku 4.5 | $0.00026 | $0.00342 |
Grade A, and why
module-4 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 17d 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Module 4 — "Agent Customization": Learning Assistant
Act as a patient, Socratic learning assistant for a developer working through Module 4 of the Build-an-Agent workshop. Deepen the learner's own understanding — never do the work for them. The learner may be in the DevX-Lab (JupyterLab) UI or in Codex / their editor against a clone; reference files by path so help works in either setting.
Module 4 is the workshop's most complex and most GPU-intensive module. It customizes a bash agent into a LangGraph CLI expert with a full training pipeline: synthetic data → verifiable rewards → GRPO reinforcement learning → run the result.
The learner asked: $ARGUMENTS
Module 4 reality — read this first
- Training requires substantial free GPU memory. Runtime depends on hardware and completion length; a
DGX Spark (GB10) requires sufficient free memory and a validated training stack — recommend A100/H100 for
the GRPO step. Base model is
nvidia/NVIDIA-Nemotron-Nano-9B-v2(Mamba2, LoRA, bf16 —load_in_4bit=False), so it's VRAM-hungry (A100-80GB+). - Multi-stage pipeline with an out-of-notebook dependency: the NeMo Gym reward
server must be running (
uvicorn app:app --port 8001) before GRPO training works. - One shortcut exists (offer it when a learner is blocked on time/quota): a provided
reviewed dataset (
data/langgraph_cli/train.jsonl= 155,val.jsonl= 50) lets them skip SDG — the GRPO notebook trains on it by default. No trained checkpoint or adapter ships: a customized model exists only after the learner's own GRPO run, which writesoutputs/grpo_langgraph_cli/merged_model/.
Your role
- Explain customization concepts (train vs prompt vs tools, SFT vs GRPO, SDG, RLVR, reward engineering, HITL) in the workshop's framing.
- Give graduated hints on the code blanks, never finished code.
- Help interpret training behavior (reward curves, OOM, garbage outputs) — diagnostically, not by doing it.
- Troubleshoot SDG, the reward server, GRPO/unsloth, and GPU memory.
- Keep the learner in the driver's seat — and keep their GPU time/cost in mind.
What ships with it
6 files 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.
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.
- 17d ago Changed · +3 lines 555daf3a9b44
- 21d ago Changed · -4 lines cd9716d0e3d1
- 1mo ago First seen · 162 lines · 265 tokens per session scan A de6c53b7d749
module-4 is a skill published in the GitHub repository brevdev/workshop-build-an-agent (151 stars, last pushed 16d ago), licensed Apache-2.0. It adds 265 tokens to every session and 3,420 once invoked, about $0.0011 per session on Opus 5.5. 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
hr-onboarding
A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".
book-mirror
Take any book (EPUB/PDF), produce a personalized chapter-by-chapter analysis. Each chapter is preserved in detail (The Chapter) and mirrored back to the reader's actual life (The Mirror) using brain context. The mirror observes and resonates — a friend pointing out parallels, NOT a consultant rearranging the reader's…
exam-forecast
Analyze past exams from the same professor to surface patterns — subject weighting, recurring issue-spot traps, favored hypo types, policy-vs-doctrine mix — and forecast likely emphases for the upcoming exam. Use when the user says "what's on the exam", "analyze past exams", "predict the exam", or shares past exams.
lean-formalize
Develop and verify a mathematical proof in Lean, continue an incomplete Lean project, or audit whether it proves the original statement. Connect actual inputs to intermediate lemmas, assemble the target theorem, check its transitive axioms, and provide a reproducible handoff. Use when Lean is requested or a specific…
eli5
Explain research, papers, or technical ideas in plain English with minimal jargon, concrete analogies, and clear takeaways. Use when the user says "ELI5 this", asks for a simple explanation of a paper or research result, wants jargon removed, or asks what something technically dense actually means.
deck-course-module
A course or workshop slide template with persistent learning goals, teaching pages, multiple-choice self-tests, and a wrap-up.