sota-llm-engineering SKILL.md skillA
sota-llm-engineering from martinholovsky/sota-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add martinholovsky/sota-skills/sota-llm-engineering@git:20261026.e511139Pins git:20261026.e511139 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- martinholovsky/sota-skills
- Path
skills/sota-llm-engineering/SKILL.md- Versions recorded
- 15
- Latest version
- git:20261026.e511139, committed Oct 26, 2026
- First version recorded
- Jun 17, 2026
- Size
- 15,006 bytes
- Repository stars
- 26
- License
- CC-BY-4.0
State-of-the-art LLM application engineering rules for BUILDING and AUDITING LLM-powered features. Claude should use this skill whenever it is building, modifying, or reviewing anything that calls a language model — chat features, RAG pipelines, agents and tool use, structured extraction, classification, summarization, embeddings/vector search, evals and regression gates, prompt or context engineering, model selection/routing, fine-tuning decisions, or LLM cost/latency/observability work.…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# SOTA LLM Engineering ## Purpose One skill, two modes. The `rules/` files define the baseline for engineering LLM-powered software that is **measured, grounded, bounded, and observable**. In **BUILD** mode you write LLM features that conform to the rules by default. In **AUDIT** mode you hunt for violations and report them as severity-rated findings. The rules are the single source of truth for both. Framing: an LLM call is a non-deterministic, expensive, latency-heavy RPC to a dependency that changes underneath you. Everything here follows from that — you don't ship logic you can't measure (evals), you don't trust output you didn't validate (structured output, grounding), you don't run loops you can't stop (budgets), and you don't deploy what you can't trace (observability). **Scope boundary:** this skill owns build quality. Prompt injection, the lethal trifecta, tool-call authorization, and model-output-as-untrusted-data live in [`sota-code-security` rules/08](../sota-code-security/rules/08-llm-ai-security.md); agent sandboxing/isolation lives in [`sota-sandboxing` rules/05](../sota-sandboxing/rules/05-ai-agent-sandboxing.md); PII/regulatory handling lives in …
Read the whole file at its exact version.
How to install
mdr add martinholovsky/sota-skills/sota-llm-engineering@sha256:1db4a0010bfb7209The line at the top of this page pins the label git:20261026.e511139, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_jgtyk5axlbysgjwk)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 15 versions of this file, between Jun 17, 2026 and Oct 26, 2026, about one every 9 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261026.e511139 latest | 2026-10-26 | e511139 | 15,006 B | A | view · diff |
| git:20261024.8dd2b0a | 2026-10-24 | 8dd2b0a | 14,984 B | A | view · diff |
| git:20261016.520862d | 2026-10-16 | 520862d | 14,743 B | A | view · diff |
| git:20261012.ae1b29b | 2026-10-12 | ae1b29b | 13,918 B | A | view · diff |
| git:20260926.3b81c65 | 2026-09-26 | 3b81c65 | 13,215 B | A | view · diff |
| git:20260926.1b7230a | 2026-09-26 | 1b7230a | 13,150 B | A | view · diff |
| git:20260926.e89beb1 | 2026-09-26 | e89beb1 | 13,159 B | A | view · diff |
| git:20260925.2b21d93 | 2026-09-25 | 2b21d93 | 13,179 B | A | view · diff |
| git:20260912.34a388b | 2026-09-12 | 34a388b | 12,884 B | A | view · diff |
| git:20260908.b22c911 | 2026-09-08 | b22c911 | 12,771 B | A | view · diff |
| git:20260903.36172e4 | 2026-09-03 | 36172e4 | 12,748 B | A | view · diff |
| git:20260903.02cb41e | 2026-09-03 | 02cb41e | 12,669 B | A | view · diff |
| git:20260827.8610b01 | 2026-08-27 | 8610b01 | 12,489 B | A | view · diff |
| git:20260724.d1f5ab0 | 2026-07-24 | d1f5ab0 | 12,461 B | A | view · diff |
| git:20260617.c5d2c7b | 2026-06-17 | c5d2c7b | 12,362 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 (15006 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
martinholovsky/sota-skills · 26 stars · license CC-BY-4.0 · pushed 2026-11-05 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_jgtyk5axlbysgjwk GET https://markdownregistry.com/api/v1/resolve?ref=martinholovsky/sota-skills/sota-llm-engineering GET https://markdownregistry.com/api/v1/blob/1db4a0010bfb72097dcec70b97bc9f4adc2c07955c22fc1128d2b01438b69982
Your AI assistant 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.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.