AGENTS.md@learnings agentsA
AGENTS.md@learnings is agent-read markdown (agents) from robium-ai/robium.
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Learning-engine guidance This file applies under `learnings/` and to `scripts/engine/` work. Load the `learning-loop` skill for consolidation, absorption, refining, experiments, or retention maintenance. ## Capture without interrupting builds - Hooks silently queue corrections/errors and archive session transcripts. - Do not stop implementation to write a finding after every event. At a natural milestone, batch meaningful flags into `observations/<skill>.md` as `tentative` entries and discard context-free noise. A tentative entry needs only a status and a signal; consolidation completes it. - A retro is useful only when it carries a real signal; clean per-skill score lines are not mandatory. - Project-local facts belong in the owning app/site documentation, not in the cross-project learning corpus. ## Promotion and absorption - Queue flags are pointers, not knowledge. Promote only when the transcript establishes expected behavior, actual behavior, and a skill or `new-skills`. - Only `status: ready` observations can be absorbed. Preserve tentative, rejected, and absorbed entries as the dedup/audit trail. …
Read the whole file at its exact version.
How to install
mdr add robium-ai/robium/learnings/AGENTS.md@git:20260922.d17aa3emdr add robium-ai/robium/learnings/AGENTS.md@sha256:5fd9b0a27603ab5fPin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_val57m4d2zbcohwg)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260922.d17aa3e latest | 2026-09-22 | d17aa3e | 2,656 B | A | view · diff |
| git:20260922.a4fc9b1 | 2026-09-22 | a4fc9b1 | 2,526 B | A | view · diff |
| git:20260907.863840e | 2026-09-07 | 863840e | 1,845 B | A | view · diff |
| git:20260828.a7b345e | 2026-08-28 | a7b345e | 1,780 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (2656 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
robium-ai/robium · 14 stars · license MIT · pushed 2026-09-22 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_val57m4d2zbcohwg GET https://markdownregistry.com/api/v1/resolve?ref=robium-ai/robium/learnings/AGENTS.md GET https://markdownregistry.com/api/v1/blob/5fd9b0a27603ab5f7fc9a3e1531af9d923347572ec05f656da7f2173ea5456fc
Your agent 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.