AGENTS.md@integrations/nemoclawA
AGENTS.md@integrations/nemoclaw from nvidia-tao/tao-skill-bank is an AGENTS.md file: the standing instructions coding agents read in that repo.
mdr add nvidia-tao/tao-skill-bank/integrations/nemoclaw/AGENTS.md@git:20260828.614fe51Pins git:20260828.614fe51 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 14 of 14 checks passed
- Kind
- AGENTS.md file
- Repository
- nvidia-tao/tao-skill-bank
- Path
integrations/nemoclaw/AGENTS.md- Versions recorded
- 4
- Latest version
- git:20260828.614fe51, committed Aug 28, 2026
- First version recorded
- Aug 3, 2026
- Size
- 7,991 bytes
- Repository stars
- 95
- 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
# TAO on NemoClaw — Agent Operating Guide You run NVIDIA TAO workflows on the host GPU from inside a NemoClaw (OpenShell) sandbox. TAO executes on the **host** through the `tao` MCP tools; you never run Docker or hold credentials yourself. **Do not ask which platform to use, and do not use the Brev, SLURM, Kubernetes or local-docker dispatch skills.** Under NemoClaw the execution platform *is* the host `tao` MCP server. Pick between the two execution tools by what the work needs — the tool descriptions cover the rest: - `tao_exec` — CPU shell over the whole workspace at `/workspace`. Everything that is not GPU compute: inspect data, move and unpack files, author specs, run a skill's helper scripts, and stage models or datasets (host network, `HF_TOKEN`/`NGC_KEY` forwarded). Never burn a GPU job on a file copy, a download, or a helper script. - `tao_run` — the GPU job. It never pulls, so `tao_pull` first or it fails fast. Never move a file through your own context (no base64, no chunking) — read and write it in the shell. ## Orchestration: run to completion A workflow runs to completion in this session. Do not finish a stage and wait to …
Read the whole file at its exact version.
How to install
mdr add nvidia-tao/tao-skill-bank/integrations/nemoclaw/AGENTS.md@sha256:d329174bf98e9d9fThe line at the top of this page pins the label git:20260828.614fe51, 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_jdh3icgoqd255sqw)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, between Aug 3, 2026 and Aug 28, 2026, about one every 8 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260828.614fe51 latest | 2026-08-28 | 614fe51 | 7,991 B | A | view · diff |
| git:20260811.7fd814a | 2026-08-11 | 7fd814a | 7,996 B | A | view · diff |
| git:20260811.e508439 | 2026-08-11 | e508439 | 7,124 B | A | view · diff |
| git:20260803.03f2daa | 2026-08-03 | 03f2daa | 9,358 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (7991 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
nvidia-tao/tao-skill-bank · 95 stars · license Apache-2.0 · pushed 2026-10-21 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_jdh3icgoqd255sqw GET https://markdownregistry.com/api/v1/resolve?ref=nvidia-tao/tao-skill-bank/integrations/nemoclaw/AGENTS.md GET https://markdownregistry.com/api/v1/blob/d329174bf98e9d9f232744e910c22d210de2524dc6708340deeb304b8fe79306
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: how to write a good AGENTS.md · AGENTS.md vs CLAUDE.md · which agent file does what.