karpathy-loop skillA
karpathy-loop is agent-read markdown (skill) from andersonlimahw/lemon-ai-hub: Executes autonomous Loop Engineering cycles for continuous software engineering optimization — performance, build time, bundle size, test coverage, correctness — against any measurable metric..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Karpathy Loop (Loop Engineering) Implements **Loop Engineering** — the practice of delegating experimentation loops to an AI agent that iteratively modifies code, runs a timed experiment, evaluates a metric, and decides to keep or revert the change. Generalized from Andrej Karpathy's Autoresearch method to general-purpose software engineering: performance tuning, build/CI optimization, bundle-size reduction, test-coverage loops, flaky-test fixes, refactors — anything with a measurable metric and a revert path. No ML/GPU/dataset assumptions required. ## Core Architecture Three-part contract: | Part | Role | Agent Editable? | Required? | |------|------|----------------|-----------| | Setup script (wired via the `evaluation_script` field; e.g. `prepare.sh`, fixtures, seed data) | Deterministic baseline for the experiment | No (immutable baseline) | Optional — see `evaluation_script` under `/runLoop` below | | Target file(s) | The code under optimization — any language, any layer | Yes (agent's playground) | Required | | `program.md` | Human-written instructions, constraints, metric definition, and acceptance criteria | No (human edits only) | Required | …
Read the whole file at its exact version.
How to install
mdr add andersonlimahw/lemon-ai-hub/karpathy-loop@v2.1.0mdr add andersonlimahw/lemon-ai-hub/karpathy-loop@sha256:91479c74b3a6c01fPin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_bkc4qnvsbfv4a6hf)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v2.1.0 latest | 2026-09-07 | 68f584a | 5,597 B | A | view · diff |
| v2.0.0 | 2026-07-10 | 9c72b5d | 4,797 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (5597 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
andersonlimahw/lemon-ai-hub · 14 stars · license none · pushed 2026-09-20 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_bkc4qnvsbfv4a6hf GET https://markdownregistry.com/api/v1/resolve?ref=andersonlimahw/lemon-ai-hub/karpathy-loop GET https://markdownregistry.com/api/v1/blob/91479c74b3a6c01f22311146a2c40d4ba2787ac30742fe97dfa045e052f3c7f9
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.