calibrate SKILL.md skillA
calibrate from skymanbp/sci-paper is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add skymanbp/sci-paper/calibrate@git:20261002.7e38af6Pins git:20261002.7e38af6 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- skymanbp/sci-paper
- Path
skills/calibrate/SKILL.md- Versions recorded
- 6
- Latest version
- git:20261002.7e38af6, committed Oct 2, 2026
- First version recorded
- Aug 26, 2026
- Size
- 15,590 bytes
- Repository stars
- 6
- License
- a license file GitHub does not identify
Make the plugin yours. Walks the whole calibration chain for one field — corpus intake, profile extraction, per-axis calibration, optional model training, a held-out provenance measurement, and finding-level labelling against your own standard, style and target journal — then verifies the result. Every axis reports `measured`, `degraded` or `unmeasured` from its own floor; a stratum that cannot support a rate stays `unmeasured` rather than being given one. Produces evidence only: no threshold…
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
> **Not a policy source.** `docs/SCIPAPER_STANDARD.md` decides what good > scientific prose is; nothing produced here can change that. Calibration > supplies the *reference distributions* the advisory axes compare against, and > `docs/architecture/EVALUATION.md` records what each one is worth. A corpus > statistic cannot redefine a consequence class, create an authorship verdict, > or turn an advisory into a blocker. # calibrate — 校准：make the axes speak your field ## 0. What this skill does, and what it refuses A fresh clone ships **no** corpus and **no** profile. Every `measured` axis in this plugin is measured *against a reference distribution built from papers you supply*, so out of the box the field-referenced axes are honestly `unmeasured`. This skill is the path from that state to a calibrated one, for one field. It refuses three things, and will say so rather than do them: 1. **It will not lower a floor to make a number appear.** If a section bucket holds fewer than 30 reference passages, or a labelling cell fewer than 20 labels, the result is `unmeasured`. A rate computed from four observations is worse than no rate, because it reads like the others. …
Read the whole file at its exact version.
How to install
mdr add skymanbp/sci-paper/calibrate@sha256:57988415d9ba9271The line at the top of this page pins the label git:20261002.7e38af6, 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_xn7seavs6bukbw67)
5 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 6 versions of this file, between Aug 26, 2026 and Oct 2, 2026, about one every 7 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261002.7e38af6 latest | 2026-10-02 | 7e38af6 | 15,590 B | A | view · diff |
| git:20260927.2c70054 | 2026-09-27 | 2c70054 | 15,080 B | A | view · diff |
| git:20260927.b25f4cd | 2026-09-27 | b25f4cd | 15,021 B | A | view · diff |
| git:20260927.de925a9 | 2026-09-27 | de925a9 | 14,593 B | A | view · diff |
| git:20260827.b2a4a9b | 2026-08-27 | b2a4a9b | 14,656 B | A | view · diff |
| git:20260826.23e80b0 | 2026-08-26 | 23e80b0 | 13,645 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 (15590 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
skymanbp/sci-paper · 6 stars · license a license file GitHub does not identify · pushed 2026-10-17 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_xn7seavs6bukbw67 GET https://markdownregistry.com/api/v1/resolve?ref=skymanbp/sci-paper/calibrate GET https://markdownregistry.com/api/v1/blob/57988415d9ba927106de6b32b6dc41642398746c86ba8da85fbce07c8e2eaf31
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 skymanbp/sci-paper
Every file in skymanbp/sci-paper