blind-eval SKILL.md skillA
blind-eval from tcuzzo/backs-aios-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add tcuzzo/backs-aios-skills/blind-eval@git:20261031.160d2a2Pins git:20261031.160d2a2 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- tcuzzo/backs-aios-skills
- Path
i18n/es/skills/blind-eval/SKILL.md- Versions recorded
- 5
- Latest version
- git:20261031.160d2a2, committed Oct 31, 2026
- First version recorded
- Aug 27, 2026
- Size
- 7,518 bytes
- Repository stars
- 2
- License
- a license file GitHub does not identify
Úsala antes de aterrizar cualquier cambio donde la pregunta es de gusto o de calidad y un test no puede decidirla. Juzga el cambio por sus méritos con la autoría oculta, y luego lo conserva o lo revierte — un empate revierte; solo aterriza la mejora probada. Trigger words: blind eval, karpathy, keep or revert, quality gate, taste call, blind judge, A/B judge, prove uplift. Disparadores: evaluación ciega, conservar o revertir, puerta de calidad, juicio a ciegas, juez A/B, probar la mejora.
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
# Blind Eval **Effort:** light — una sola corrida de juez ciego: varias lecturas barajadas del par congelado en una sesión de revisión nueva que no escribió ninguno de los dos. Elimina: aterrizajes autocalificados de "quedó mejor" — regresiones de gusto que el autor dejaría pasar. Una puerta de calidad de conservar-o-revertir para decisiones que un test no puede tomar — la calidad de una prosa, el copy de una UI, la legibilidad de un refactor, la salida de un prompt, la sensación de un diseño. Juzga el cambio por sus méritos con la autoría oculta, y luego CONSÉRVALO o REVIÉRTELO. Un empate revierte. Solo aterriza la mejora probada. ## Cuándo ejecutarla - Antes de aterrizar cualquier cambio donde "¿es mejor?" es una pregunta de gusto o calidad. - Como la puerta dentro de un loop de mejora: proponer → probar → medir → conservar o descartar. - Cada vez que el autor sienta la tentación de declarar su propio trabajo una mejora. ## El método 1. **Escribe qué significa "mejor" ANTES de mirar.** Una meta en lenguaje llano. Una medida primaria o un eje de rúbrica con una barra dura — un nivel que superar, no un número que empujar. Ejes secundarios en orden de prioridad …
Read the whole file at its exact version.
How to install
mdr add tcuzzo/backs-aios-skills/blind-eval@sha256:cd275e997deb3192The line at the top of this page pins the label git:20261031.160d2a2, 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_hgj5bhwmehw2ybn6)
0 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Aug 27, 2026 and Oct 31, 2026, about one every 16 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261031.160d2a2 latest | 2026-10-31 | 160d2a2 | 7,518 B | A | view · diff |
| git:20261030.9f0598a | 2026-10-30 | 9f0598a | 7,498 B | A | view · diff |
| git:20261015.d6c2fce | 2026-10-15 | d6c2fce | 7,658 B | A | view · diff |
| git:20260828.045e846 | 2026-08-28 | 045e846 | 7,201 B | A | view · diff |
| git:20260827.a68a1ea | 2026-08-27 | a68a1ea | 7,195 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 (7518 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
tcuzzo/backs-aios-skills · 2 stars · license a license file GitHub does not identify · pushed 2026-11-03 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_hgj5bhwmehw2ybn6 GET https://markdownregistry.com/api/v1/resolve?ref=tcuzzo/backs-aios-skills/blind-eval GET https://markdownregistry.com/api/v1/blob/cd275e997deb319263ef97e563d7d631e0a8f7af77323cb3784948d95352d13f
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 tcuzzo/backs-aios-skills
Every file in tcuzzo/backs-aios-skills