What does the Jaroslav skill do?

Build coding agents that ship: repo context, sandboxed test loops, safe PR limits, and cost per resolved issue. Load it once into Claude Projects and you get a configured Coding Agent Engineer without re-explaining context at the start of every session. Measure the model on your own data before trusting any public benchmark, keep a human in the loop wherever a wrong answer is expensive, and state the failure modes as plainly as the wins.

How do I install this skill file?

Download the .skill package (it contains SKILL.md), paste the contents into Claude Projects Instructions or your AI's system prompt, add your own context and start your first session. Works with Claude, ChatGPT, or any AI chat that accepts system prompts.

Which AI tools does this skill work with?

Works with Claude (recommended), ChatGPT, Gemini, Perplexity and Copilot, and any AI chat that accepts system prompts. Claude Projects gives the best results.

What is included in this download?

One .skill package delivered instantly after purchase: the full SKILL.md role configuration plus a worked-example file with a real scenario so you see the quality before you rely on it. No subscription, yours permanently.

How is this different from using Claude without the Jaroslav skill?

Without a skill file your AI starts every session as a general assistant. With Jaroslav loaded it applies Coding Agent Engineer methodology from the first message, with consistent quality every time. Measure the model on your own data before trusting any public benchmark, keep a human in the loop wherever a wrong answer is expensive, and state the failure modes as plainly as the wins.

Coding Agent Engineer AI Skill for Claude & ChatGPT

Name: Jaroslav — Coding Agent Engineer AI Skill
Brand: KissMySkills
Price: 14.99 USD
Availability: InStock

Jaroslav — Coding Agent Engineer AI Skill

$14.99

Jaroslav — Coding Agent Engineer AI Skill

$14.99 this skill vs $160+/hr, hiring a coding agent engineer

⚡ Instant download ◆ Claude & ChatGPT ∞ Keep forever

Instant download · 30-day money-back guarantee. Pay once, keep forever — no subscription. Refund policy

Build coding agents that ship: repo context, sandboxed test loops, safe PR limits, and cost per resolved issue.

Repo context strategy and the real limits of a big window
Sandboxed execution with a real shell and a real test run
Patch generation, multi-file edits and agent-authored PRs
Cost per resolved issue, blast-radius limits, loop and fake-fix detection

Teams deploying coding agents on a private codebase who need honest numbers instead of benchmark marketing.A coding agent engineer commands $160+/hr, this is one file, yours forever.

// what's inside

Drop Jaroslav into Claude and get a coding agent engineer who measures cost per resolved issue, because a benchmark score on a public repo tells you nothing about your codebase.

Jaroslav builds agents that write and ship real code: repository context strategy across retrieval, file-tree summarisation and the limits of stuffing a monorepo into a window; sandboxed execution because an agent needs a real shell and a real test run; test-driven agent loops where the suite is the reward signal; patch and diff generation versus whole-file rewrite; multi-file coordinated edits; build and dependency resolution inside the sandbox; agent-authored pull requests and the review contract with humans; CI integration and blast-radius limits on what an agent may merge; why SWE-bench scores do not transfer to a private codebase; and the failure taxonomy for an agent that loops or fakes a fix.

What you get

→Repo context strategy and the real limits of a big window
→Sandboxed execution with a real shell and a real test run
→Patch generation, multi-file edits and agent-authored PRs
→Cost per resolved issue, blast-radius limits, loop and fake-fix detection

📄 jaroslav-coding-agent-engineer.skill Under 2 min install Works with Claude, ChatGPT & any AI chat

How to install

Download the .skill package → open Claude → paste SKILL.md into your Project Instructions or system prompt → describe your requirement → Jaroslav builds the answer. Includes a full worked example so you see exactly what you get.

jaroslav-coding-agent-engineer.skill

# Jaroslav - Coding Agent Engineer

You are Jaroslav, a coding agent engineer. Cost per resolved issue is the metric; a public benchmark score is marketing.

## How you work
1. Design repository context before choosing a model
2. Give the agent a real sandbox, a real build and a real test run
3. Use the test suite as the reward signal; reject unverified patches
4. Measure resolution rate and cost on the client's own issues

Never let an agent merge outside a defined blast radius, and never accept a fix the test suite did not prove.

Excerpt from the actual file you'll download.

// try it

prompt

$We want an agent that closes our backlog tickets. Design it and tell me what it will really cost.

Jaroslav returns a repository context design with the retrieval strategy justified, a sandbox and test-loop specification, a patch generation approach with the review contract, CI blast-radius limits and permission scoping, a measured resolution rate on a sample of the client's own issues rather than a public benchmark, cost per resolved issue, and the failure taxonomy with detection for loops and faked fixes.

// how to install Under 2 minutes

Four steps. Any AI chat.

01
Download the file
After checkout, the download link lands in your inbox. Save the file anywhere on your device.
02
Open your AI chat
Claude, ChatGPT, Gemini, Grok, or Copilot — whichever one you already use.
03
Paste the file contents
Drop it into the system prompt, Project instructions, or custom instructions field.
04
Start working
Your AI is now configured as a specialist. Ask it anything inside its domain.