device-training-1060 SKILL.md skillA
device-training-1060 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/device-training-1060@git:20260823.df7cbdePins git:20260823.df7cbde in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- anonymousnomad/aide-sovereign-workbench
- Path
skills/packs/device-training-1060/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260823.df7cbde, committed Aug 23, 2026
- Size
- 8,415 bytes
- Repository stars
- 53
- License
- Apache-2.0
Device-specific training discipline for THIS machine (GTX 1060 6GB Pascal mobile + i7-8750H + 16GB RAM + Windows). The execution layer under training-from-scratch and model-scaling: what Big Tech recommends vs what this exact card can actually do. Use whenever setting up, launching, resuming, or optimizing ANY training run (pretrain/SFT/distill/preference) on this machine — precision choice, batch/data-pipeline construction, throughput budgets, Windows DataLoader pitfalls, and the 27%-GPU-util…
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
# Device Training: GTX 1060 6GB + i7-8750H + 16GB + Windows Grounded in research (sources below). This skill is the DEVICE layer; it does not replace training-from-scratch (big-tech pipeline) or model-scaling (size ceiling). All three must agree — if they ever conflict, this device skill wins on physics (what the card can do), training-from-scratch wins on recipe (how big-tech trains), model-scaling wins on size (never >100-150M). ## Device facts (measured + sourced) | Component | Spec | Consequence for training | |---|---|---| | GPU | GTX 1060 Mobile 6GB GDDR5, Pascal CC 6.1, 1404MHz | NO Tensor Cores | | FP16 rate | 1/64th of FP32 on CC 6.1 (NVIDIA CUDA guide; Anandtech 1080 = 1/64) | FP16 is SLOWER for math here | | Memory bandwidth | 192 GB/s GDDR5 | THE throughput limiter for decode & attention | | PCIe | 3.0 x16 (~16 GB/s effective) | H2D bulk transfers matter; keep data contiguous | | CPU | i7-8750H 6C/12T 2.2-4.1GHz | thermal throttles ~9-14% sustained (Notebookcheck) | | RAM | 16GB total | num_workers>0 duplicates data per worker (numpy) | | Power | laptop battery | GPU -53% on battery (Notebookcheck) | ## Hard rule #1 — FP32 ONLY. Never AMP/TF32/FP16 for speed. …
Read the whole file at its exact version.
How to install
mdr add anonymousnomad/aide-sovereign-workbench/device-training-1060@sha256:d71854f83b34f096The 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_iyczx3p7vduolkex)
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
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (8415 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
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_iyczx3p7vduolkex GET https://markdownregistry.com/api/v1/resolve?ref=anonymousnomad/aide-sovereign-workbench/device-training-1060 GET https://markdownregistry.com/api/v1/blob/d71854f83b34f09638b8076db48894e4092a2efd151bdf5e1a2dcf3a73f1acac
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.