evidence-first-research SKILL.md skillA
This file is byte-identical to the first copy the registry indexed. Same content hash, so the same checks give both the same grade.
evidence-first-research from arasz/ai-badger is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add arasz/ai-badger/evidence-first-research@v1.0.0Pins v1.0.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- arasz/ai-badger
- Path
.ai-badger/skills/evidence-first-research/SKILL.md- Versions recorded
- 4
- Latest version
- v1.0.0, committed Sep 11, 2026
- First version recorded
- Aug 6, 2026
- Size
- 6,268 bytes
- Repository stars
- 3
- License
- MIT
Use when a question needs investigating and the answer will be acted on — "research X", "look into whether Y", "find out how Z works", "is this worth doing", "compare these options", a benchmark someone will quote, or a claim that has to survive being challenged. Produces a dated record where every finding carries how it is known — measured, read, inferred, or unverified — plus a self-contained HTML view. Not for locating code (that is explore-codebase), tracing one symptom (debug-issue), or…
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
# Evidence-first research Answer the question, and make how you know it inseparable from what you found. **Why this exists.** A report where *"the gate takes 75 seconds, measured"* and *"the gate takes about five minutes, roughly"* look identical is a report that will eventually be quoted wrong. That happened here: a cost claim of "3-5 minutes per push, about an hour across fifteen pushes" reached a changelog and a PR body. Measured, it was 75 seconds and nineteen minutes — off by four times, and caught only because someone re-derived it by hand. Nothing in the format made the unmeasured number look unmeasured. So every finding is graded, the grades are a closed set, and the renderer **refuses a record that grades badly** rather than producing a page that looks finished. ## The four grades | Grade | Means | Needs | |---|---|---| | `MEASURED` | You ran it and read the result | The command, the machine, the conditions | | `READ` | You read it in a source that is authoritative for this | `path:line`, a spec section, a doc URL | | `INFERRED` | You reasoned to it from things above | Say what from — reasoning is not a citation | …
Read the whole file at its exact version.
How to install
mdr add arasz/ai-badger/evidence-first-research@sha256:1f89e49e40652260The line at the top of this page pins the label v1.0.0, 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_jomrp6gzcuckmcv4)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, between Aug 6, 2026 and Sep 11, 2026, about one every 12 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.0.0 latest | 2026-09-11 | c8da0f0 | 6,268 B | A | view · diff |
| v1.0.0 | 2026-08-08 | 9006bf5 | 6,266 B | A | view · diff |
| v1.0.0 | 2026-08-07 | 013ee4d | 6,253 B | A | view · diff |
| v1.0.0 | 2026-08-06 | 48efdac | 6,035 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 (6268 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
arasz/ai-badger · 3 stars · license MIT · pushed 2026-10-20 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_jomrp6gzcuckmcv4 GET https://markdownregistry.com/api/v1/resolve?ref=arasz/ai-badger/evidence-first-research GET https://markdownregistry.com/api/v1/blob/1f89e49e4065226091a2a1ce1ee4ca0d5009442fe6fbf4d2284c654c461c3c9e
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.