PromptMarket — prompt packs for agent engineering
Expert-grade prompts for the agent engineering work you can't wing
Redline Prompts is 50 prompts for running LLM agents in production — system prompt architecture, eval design, failure diagnosis, injection hardening and more — built to pull senior-engineer-level output from whatever model you already use. Two companion packs cover the rest of the loop: Baseline Presets to start from, Press Check to grade with.
See the packs- Packs
- 3
- Items
- 80
- Price from
- $39
01Featured pack
02The set
Three packs, one loop: build the prompt, check it against reality, grade what comes out. Each stands alone, so start wherever your agent is.
- 01Start
Baseline Presets
Start from a solid prompt: 15 deployable system prompts, each with tool schemas, an output contract and 12 contract tests.
from $39 - 02Audit
Redline Prompts
Audit it against your traces: 50 prompts that turn your prompt, schema or incident log into an audit, a diagnosis or a plan.
from $39 - 03Grade
Press Check
Grade output at scale: 15 LLM judges with anchored labels, strict JSON output and a 10-item gold set each.
from $39
03How it works
Pick a pack, then Solo or Team
Solo covers one developer. Team covers up to 10 seats in one organisation on one flat price — no per-seat approval needed.
Get instant access
Checkout runs on Gumroad. Right after, the pack downloads from your Gumroad library.
Put it to work on your own data
Paste a prompt into the model you already use, deploy a preset as your agent's system prompt, or load a judge into your eval runner — then point it at your own traces.
04Frequently asked questions
01What exactly do I get after buying a pack?
An instant download from your Gumroad library. Redline Prompts is one Markdown file of prompts, each with the prompt text, when to use it, an example scenario, a sample strong response, notes on adapting it to your stack and which model to run it on. Baseline Presets and Press Check are .zip files: the whole pack as one readable Markdown file, plus the machine-readable pieces (system prompts, tool and output schemas, contract tests, judge prompts, gold sets) in folders you can diff, version and test.
02Which pack should I start with?
Building a new agent: Baseline Presets. Already running one that misbehaves, or about to change its model: Redline Prompts. Sitting on logged traffic or an eval set you grade by hand: Press Check. Each pack stands alone and is sold separately; where it helps, the items point to each other.
03What's the difference between Solo and Team?
Solo ($39) is licensed to one developer. Team ($149) covers up to 10 seats in one organisation at one flat price — no per-seat billing, easy to expense without a PO. Prices are in USD. Applicable VAT or sales tax is calculated and added by Gumroad at checkout.
04Can I use the packs in a commercial product?
Yes. Use the prompts, presets and judges in your own commercial projects, including deploying a Baseline preset as your product's system prompt. What you can't do is resell or republish the packs themselves. The details are in the Terms.
05Who makes these, and how?
The packs are drafted and reviewed with AI models and curated by the author. We don't publish benchmark scores: Baseline Presets ships contract tests and Press Check ships gold sets, so you can check them on your own model and data.
06What if a pack does not earn its keep?
14-day money-back guarantee. Email maksim344551@gmail.com within 14 days of purchase and we'll refund your full order through Gumroad, no questions asked.
07How do delivery and updates work?
Instant download from your Gumroad library right after purchase. Updates to a pack you own are free and arrive the same way: re-download it there for the latest version.