Menu
SkillsMP has collected 6 skills from agentculture/unsloth-cli. Open a skill to review its source and details.
Showing 6 of 6 collected skills.
Explain how to use unsloth-cli — the agent-first CLI and its fine-tuning verbs. Covers the two faces of the tool (GPU-free introspection verbs: whoami / learn / explain / overview / doctor / cli; and the LoRA/QLoRA fine-tuning verbs: train / eval / export…
Drive a LoRA/QLoRA adapter training run end-to-end for this repo's fine-tuning verbs: validates the dataset and scope guard (GPU-free), resolves the plan via `sloth train --dry-run`, then (for a real run) runs the full adapter job (`sloth train`), evaluates…
Search the shared eidetic memory store and get back ranked, provenanced records. Drives `eidetic recall` with four search modes — exact (verbatim substring), approximate (vector/semantic), keyword (BM25 lexical), and hybrid (a weighted blend of…
Ingest records into the shared eidetic memory store so they can be recalled later. Drives `eidetic remember`: accepts one record as a JSON object, or a batch as NDJSON on stdin for bulk ingest. Upsert is idempotent by id (and dedups by content hash) —…
unsloth-cli's CI/CD lane, layered on `devex pr`. Delegates lint / open / read / reply / delta to devex; adds two extensions — `status` (SonarCloud quality gate + hotspots + unresolved-thread tally) and `await` (read --wait + status with non-zero exit on Sonar…
Cross-repo + mesh communication from unsloth-cli: file tracked GitHub issues on sibling repos, fetch issues from sibling repos to inline current state into briefs, and send live messages to Culture mesh channels. Use when the next step lives outside…