skill-refiner SKILL.md skillA

skill-refiner from iuliandita/skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add iuliandita/skills/skill-refiner@git:20261103.cd466ca

Pins git:20261103.cd466ca in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
iuliandita/skills
Path
skills/skill-refiner/SKILL.md
Versions recorded
24
Latest version
git:20261103.cd466ca, committed Nov 3, 2026
First version recorded
Jun 14, 2026
Size
22,460 bytes
Repository stars
11
License
MIT

Improve skills through repeated scoring, behavioral tests, and peer review toward a requested quality target.

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

# Skill Refiner: Iterative Self-Improvement Loop

Adaptive evaluation loop for AI skill collections, inspired by Karpathy's AutoResearch.
Orchestrates repeated score-improve-verify cycles using **skill-creator** as the engine
and mandatory peer review as an adversarial check (cross-model when distinct model identity
is verified, fresh-context self-review as the minimum fallback).

## When to use

- Batch-improving the entire skill collection after a period of manual edits
- Targeted improvement of one named skill when the user explicitly asks for skill-refiner,
  multiple iterations, a score target, or "no ceiling" polish
- Running quality sweeps before a release or publish
- Triggering a self-improvement cycle where skills bootstrap each other
- After adding several new skills that need polish and consistency alignment
- When cross-model perspective would catch single-model blind spots
- Periodic maintenance: scheduled improvement runs to keep skills current

## When NOT to use

- Creating a new skill from scratch - use **skill-creator** (Mode 1)
- Single-pass review of one skill without iteration, scoring, or peer review - use **skill-creator** (Mode 2)
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add iuliandita/skills/skill-refiner@sha256:deffba52f63ac58a

The line at the top of this page pins the label git:20261103.cd466ca, 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_vlszg5dyyr573fss.svg)](https://markdownregistry.com/a/art_vlszg5dyyr573fss)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 24 versions of this file, between Jun 14, 2026 and Nov 3, 2026, about one every 6 days.

versioncommittedcommitsizeaudit
git:20261103.cd466ca latest2026-11-03 cd466ca 22,460 BA view · diff
git:20261028.1c536c52026-10-28 1c536c5 22,323 BA view · diff
git:20261028.292c1f12026-10-28 292c1f1 33,608 BA view · diff
git:20261028.1f2283e2026-10-28 1f2283e 33,580 BA view · diff
git:20261002.a7f9a1c2026-10-02 a7f9a1c 33,426 BA view · diff
git:20260924.60a78262026-09-24 60a7826 36,022 BA view · diff
git:20260924.0a16b1f2026-09-24 0a16b1f 35,178 BA view · diff
git:20260920.ebf9a3d2026-09-20 ebf9a3d 35,163 BA view · diff
git:20260920.ac018532026-09-20 ac01853 34,404 BA view · diff
git:20260917.7c60b3a2026-09-17 7c60b3a 32,208 BA view · diff
git:20260917.d4d17e42026-09-17 d4d17e4 32,185 BA view · diff
git:20260917.4e375f02026-09-17 4e375f0 29,489 BA view · diff
git:20260917.b6a97a02026-09-17 b6a97a0 26,785 BA view · diff
git:20260916.ee8810d2026-09-16 ee8810d 25,785 BA view · diff
git:20260916.01118282026-09-16 0111828 25,046 BA view · diff
git:20260916.52ec7432026-09-16 52ec743 24,893 BA view · diff
git:20260910.4c98b5e2026-09-10 4c98b5e 22,352 BA view · diff
git:20260910.64816162026-09-10 6481616 21,736 BA view · diff
git:20260908.81be7e92026-09-08 81be7e9 21,844 BA view · diff
git:20260904.ccfce412026-09-04 ccfce41 20,629 BA view · diff
git:20260819.aae1ce62026-08-19 aae1ce6 19,573 BA view · diff
git:20260722.d313b9e2026-07-22 d313b9e 19,377 BA view · diff
git:20260722.99f7e842026-07-22 99f7e84 18,834 BA view · diff
git:20260614.4a94e9c2026-06-14 4a94e9c 18,808 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (22460 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

GitHub

iuliandita/skills · 11 stars · license MIT · pushed 2026-11-05 · branch develop

API

GET https://markdownregistry.com/api/v1/artifacts/art_vlszg5dyyr573fss
GET https://markdownregistry.com/api/v1/resolve?ref=iuliandita/skills/skill-refiner
GET https://markdownregistry.com/api/v1/blob/deffba52f63ac58a1d699cd7f6a85527ed59c574d096d63ac1e46fccae157340

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.

More from iuliandita/skills

AGENTS.md AGENTS.md file
iuliandita/skills · AGENTS.md
git:20261103.cd466ca · audit A · 11 stars
CLAUDE.md CLAUDE.md file
iuliandita/skills · CLAUDE.md
git:20261103.cd466ca · audit B · 11 stars
ansible SKILL.md skill
iuliandita/skills · skills/ansible/SKILL.md · Write, review, and debug Ansible playbooks, roles, inventories, Ansible Vault, Molecule tests, and AWX/AAP.
git:20261028.5d3d48b · audit A · 11 stars
anti-ai-prose SKILL.md skill
iuliandita/skills · skills/anti-ai-prose/SKILL.md · Strip AI-sounding filler, hedging, and canned phrasing from docs, READMEs, emails, and PR bodies. Use when text sounds…
git:20261105.aa96fb6 · audit A · 11 stars
arch-linux SKILL.md skill
iuliandita/skills · skills/arch-linux/SKILL.md · Administer Arch Linux, CachyOS, EndeavourOS, and Manjaro: pacman, AUR, upgrades, boot, GPU, and desktop issues.
git:20261028.a4222b4 · audit A · 11 stars
backend-api SKILL.md skill
iuliandita/skills · skills/backend-api/SKILL.md · Design and review REST/HTTP APIs, with bounded GraphQL and gRPC contract guidance: OpenAPI, auth, pagination…
git:20261028.b1e0514 · audit A · 11 stars
ci-cd SKILL.md skill
iuliandita/skills · skills/ci-cd/SKILL.md · Build, review, and debug CI/CD pipelines and runners: GitHub Actions, GitLab CI, Forgejo/Gitea, and Woodpecker.
git:20261028.050926d · audit A · 11 stars
code-review SKILL.md skill
iuliandita/skills · skills/code-review/SKILL.md · Review code and diffs for bugs, regressions, edge cases, races, and leaks. Use for "review this diff" or a pre-merge…
git:20261105.aa96fb6 · audit A · 11 stars
code-simplification SKILL.md skill
iuliandita/skills · skills/code-simplification/SKILL.md · Review code for dead code, duplication, overengineering, invented APIs, and weak tests. Report safe reductions; no…
git:20261028.baa0989 · audit A · 11 stars
databases SKILL.md skill
iuliandita/skills · skills/databases/SKILL.md · PostgreSQL, MySQL/MariaDB, MongoDB, MSSQL, Redis, Valkey: schemas, slow queries, EXPLAIN, indexes, migrations…
git:20261105.aa96fb6 · audit A · 11 stars
debian-ubuntu SKILL.md skill
iuliandita/skills · skills/debian-ubuntu/SKILL.md · Debian, Ubuntu, Mint, Pop!_OS, Kali admin: apt/dpkg errors, held packages, release upgrades, PPAs, kernels, drivers…
git:20261105.aa96fb6 · audit A · 11 stars
debug-triage SKILL.md skill
iuliandita/skills · skills/debug-triage/SKILL.md · Find the failing layer in an outage or post-deploy break, debug to root cause, profile slow code. Use for "it's down"…
git:20261105.aa96fb6 · audit A · 11 stars

Every file in iuliandita/skills

Other files named skill-refiner

skill-refiner SKILL.md skill
functoreality/agent-skills · skills/skill-refiner/SKILL.md · 通过复盘反思或专家输出对比,发现并提炼思维/写作/设计型 skill 的改进方向。 适用于分析问题、写高质量笔记等产出质量无客观指标,需要靠推演和专家判断来评估的 skill。 当用户要求审查、改进、更新或复盘某个…
v260804 · audit A · 11 stars

Browse by kind, by grade A, or by owner.