aide-training-phase-b2-qlora-runner SKILL.md skillC
aide-training-phase-b2-qlora-runner from anonymousnomad/aide-sovereign-workbench is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add anonymousnomad/aide-sovereign-workbench/aide-training-phase-b2-qlora-runner@git:20260823.df7cbdePins git:20260823.df7cbde in mdr.lock. Pin the fingerprint instead.
- Audit grade
- C, 14 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- anonymousnomad/aide-sovereign-workbench
- Path
skills/packs/aide-training-phase-b2-qlora-runner/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260823.df7cbde, committed Aug 23, 2026
- Size
- 3,026 bytes
- Repository stars
- 53
- License
- Apache-2.0
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
# Training Phase B2 — Hardware-Aware QLoRA Job Runner
## What
TrainingManager v2: replace the allowlisted-command stub with a real fine-tuning runner — hardware-aware preset selection, a generated pinned-dependency Unsloth script, live loss streaming to the UI, checkpoint policy (keep-3, best-by-eval-loss), OOM-ladder advisor, approval-gated start/stop.
## Why
2026 consensus recipe: QLoRA+Unsloth, r=16, alpha≈r, all-linear targets, lr 2e-4 cosine, batch-1 + gradient accumulation, 1–3 epochs. Verified VRAM tables put this machine's bundled 0.5B (~3GB) and 1.5B (~4GB) within the GTX 1060 6GB budget. The current 45-line stub can run commands but knows nothing about models, progress, checkpoints, or failure modes.
## Code Plan
- `node/src/services/training.ts` (TS port, replaces daemon stub; keep old manifest jobs importable): presets computed from `hardware.ts` probe (VRAM total/free) × candidate base models (manifest GGUFs with param counts) → {model, max_seq_length, r, alpha, lr, epochs, batch, accum} per verified table; refuse combos over budget.
…Read the whole file at its exact version.
How to install
mdr add anonymousnomad/aide-sovereign-workbench/aide-training-phase-b2-qlora-runner@sha256:31be1078ac21433cThe line at the top of this page pins the label git:20260823.df7cbde, 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_pwibvpxremeyn5k7)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Aug 23, 2026.
Audit of the latest version
- fail: Frontmatter block present
- fail: Frontmatter declares a name
- fail: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (3026 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
Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.
Source
anonymousnomad/aide-sovereign-workbench · 53 stars · license Apache-2.0 · pushed 2026-10-26 · branch covert-production
API
GET https://markdownregistry.com/api/v1/artifacts/art_pwibvpxremeyn5k7 GET https://markdownregistry.com/api/v1/resolve?ref=anonymousnomad/aide-sovereign-workbench/aide-training-phase-b2-qlora-runner GET https://markdownregistry.com/api/v1/blob/31be1078ac21433ca415993d7deaa496be62263b67942260f7fe5a3f8d7ad851
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.