Home / softtor / nestjs-hexagonal · GitHub

jev-eval SKILL.md skillA

jev-eval from softtor/nestjs-hexagonal is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned

mdr add softtor/nestjs-hexagonal/jev-eval@git:20260921.98d05f3

Pins git:20260921.98d05f3 in mdr.lock. Pin the fingerprint instead.

Audit grade: A, 17 of 17 checks passed
Kind: SKILL.md skill
Repository: softtor/nestjs-hexagonal
Path: skills/jev-eval/SKILL.md
Versions recorded: 1
Latest version: git:20260921.98d05f3, committed Sep 21, 2026
Size: 6,785 bytes
Repository stars: 5
License: MIT

Calibrates or evaluates a semantic (Jev) rule of the nestjs-hexagonal rulebook — adds labelled golden cases, runs the calibration harness against the real model, fits thresholds, reads the report and decides whether the rule stays advisory or moves to ask; also produces the weekly pilot report from the hook logs. Use when a semantic rule misfires, when a new semantic rule is added, when asked to "calibrate a rule", "add golden cases", "raise a rule to ask" or "weekly Jev report".

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

# jev-eval

A semantic rule is a question the rulebook asks Jev (`jev-1.13.0`) about a code slice. Its decision thresholds are not written by hand: they are fitted from labelled golden cases by `calibration/fit.ts`, and the rule only gains a threshold it has earned. This skill is the procedure for adding cases, running the calibration and reading the result. The detailed workflow, the cost model and the report format are in `references/calibration-workflow.md`.

**Cost and key.** `calibration/run.ts` needs `TYPESAFE_API_KEY` and sends one request per golden case (about 300 ms and a fraction of a cent each; the E1 experiment cost about US$ 0.003 for 50 requests). Without the key the run refuses to start; never commit the key, never paste it in a report. Everything else (`fit.ts`, the tests, the report) is offline.

---

## 1. Decide what the rule must catch

Read the rule in `rulebooks/hexagonal.rulebook.yaml` (`question.instructions`, `question.options` or `levels`, `state.slice`, `thresholds.uncertain`). Write down, in one sentence each, the violation the rule exists for and the legitimate code it must leave alone. Every golden case is one of those sentences made concrete.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint

mdr add softtor/nestjs-hexagonal/jev-eval@sha256:25c1dff7517ddf4f

The line at the top of this page pins the label git:20260921.98d05f3, 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](https://markdownregistry.com/badge/art_zbnbquky5rmhu5pz.svg)](https://markdownregistry.com/a/art_zbnbquky5rmhu5pz)

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

Versions

The registry has recorded 1 version of this file, committed on Sep 21, 2026.

version	committed	commit	size	audit
git:20260921.98d05f3 latest	2026-09-21	98d05f3	6,785 B	A	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 (6785 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

softtor/nestjs-hexagonal · 5 stars · license MIT · pushed 2026-10-08 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_zbnbquky5rmhu5pz
GET https://markdownregistry.com/api/v1/resolve?ref=softtor/nestjs-hexagonal/jev-eval
GET https://markdownregistry.com/api/v1/blob/25c1dff7517ddf4f025a4038520292ffbf206e22582d07d77e220c24a0ba025d

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.