tao-generate-referring-expressions SKILL.md skillA

tao-generate-referring-expressions from nvidia-tao/tao-skill-bank is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add nvidia-tao/tao-skill-bank/tao-generate-referring-expressions@v0.1.0

Pins v0.1.0 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
nvidia-tao/tao-skill-bank
Path
skills/data/tao-generate-referring-expressions/SKILL.md
Versions recorded
6
Latest version
v0.1.0, committed Oct 15, 2026
First version recorded
Jun 3, 2026
Size
10,415 bytes
Repository stars
95
License
Apache-2.0

"Four-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region.

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

# Image Referring Expression Pipeline

> **Standalone install?** If this session was not initialized by the TAO skill bank plugin, run the `tao-setup` skill first (host preflight, credentials, cross-skill discovery).

Generate referring-expression and grounding annotations from images with KITTI-format bounding box labels. A single VLM (Gemini or any OpenAI-compatible endpoint) runs four steps: per-object region descriptions, holistic image captions, grouped grounding expressions tied to bboxes, and an optional double-check verification pass.

## Purpose

Transform `(image, KITTI labels)` pairs into a unified `annotations.jsonl` containing rich, grounded referring expressions. The VLM acts as a "teacher" annotator: Steps 0-1 see the image; Step 2 groups Step 0 outputs into grouping phrases with bbox lists; Step 3 (optional) re-examines those bboxes against the image and corrects mismatches.

## Pipeline Architecture

```
Step 0: Region expression  ──┐
                              ├──▶  Step 2: Grounding expression  ──▶  [Step 3: Double check]
Step 1: Image caption  ──────┘                                                   (optional)
```
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add nvidia-tao/tao-skill-bank/tao-generate-referring-expressions@sha256:c166a294b1e30d37

The line at the top of this page pins the label v0.1.0, 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_whqhgjtf4lp646ep.svg)](https://markdownregistry.com/a/art_whqhgjtf4lp646ep)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 6 versions of this file, between Jun 3, 2026 and Oct 15, 2026, about one every 27 days.

versioncommittedcommitsizeaudit
v0.1.0 latest2026-10-15 f1e6f8d 10,415 BA view · diff
v0.1.02026-08-28 ae022a9 10,434 BA view · diff
v0.1.02026-08-03 03f2daa 10,415 BA view · diff
v0.1.02026-06-08 0ccd383 10,191 BA view · diff
v0.12026-06-03 94ef697 10,189 BA view · diff
v0.12026-06-03 2fd0a48 10,175 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (10415 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

GitHub

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_whqhgjtf4lp646ep
GET https://markdownregistry.com/api/v1/resolve?ref=nvidia-tao/tao-skill-bank/tao-generate-referring-expressions
GET https://markdownregistry.com/api/v1/blob/c166a294b1e30d372f0ca1a86b591dc0563a58a74959155ccf92afe1122e0774

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.

More from nvidia-tao/tao-skill-bank

AGENTS.md AGENTS.md file
nvidia-tao/tao-skill-bank · AGENTS.md
git:20260922.75b4f9f · audit A · 95 stars
AGENTS.md@integrations/nemoclaw AGENTS.md file
nvidia-tao/tao-skill-bank · integrations/nemoclaw/AGENTS.md
git:20260828.614fe51 · audit A · 95 stars
tao-analyze-changenet-rca SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-analyze-changenet-rca/SKILL.md · Performs deep Root Cause Analysis (RCA) on NVIDIA TAO Visual ChangeNet classification experiments with
v0.1.0 · audit A · 95 stars
tao-finetune-huggingface-model SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-finetune-huggingface-model/SKILL.md · Fine-tune any HuggingFace CV / VLM / LLM model on local NVIDIA GPUs inside an NGC PyTorch container when no dedicated…
v0.1.0 · audit A · 95 stars
tao-port-huggingface-model SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-port-huggingface-model/SKILL.md · Integrate a HuggingFace Computer Vision model into the NVIDIA TAO Toolkit ecosystem (tao-core config, tao-pytorch…
v0.1.0 · audit A · 95 stars
tao-run-automl-deft-pipeline SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-automl-deft-pipeline/SKILL.md · Run the canonical NVIDIA AOI three-phase training pipeline — Phase 1 AutoML baseline (HPO), Phase 2 DEFT loop (RCA →…
v0.1.0 · audit A · 95 stars
tao-run-automl SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-automl/SKILL.md · Run container-backed AutoML / hyperparameter optimization (HPO) for NVIDIA TAO networks using AutoMLRunner. Handles…
v0.1.1 · audit A · 95 stars
tao-run-deft-aoi-cosmos3 SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-deft-aoi-cosmos3/SKILL.md · Run the disk-backed DEFT AOI improvement loop for NVIDIA Cosmos Reason 3 / Cosmos3 models, using Nano by default and…
v0.2.0 · audit A · 95 stars
tao-run-deft-aoi SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-deft-aoi/SKILL.md · Run the full DEFT AOI improvement loop for NVIDIA TAO VisualChangeNet / ChangeNet PCB inspection models: baseline…
v0.1.0 · audit A · 95 stars
tao-run-deft-cr-its-mining SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-deft-cr-its-mining/SKILL.md · Run the mining-based DEFT improvement workflow for ITS Cosmos-Reason binary video questions, focused on the…
v0.1.0 · audit A · 95 stars
tao-run-deft-object-detection SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-deft-object-detection/SKILL.md · Run the full DEFT smart-data-augmentation loop for NVIDIA TAO Grounding DINO object detection: zero-shot baseline…
v0.1.0 · audit A · 95 stars
tao-run-deft-od-aoi SKILL.md skill
nvidia-tao/tao-skill-bank · skills/applications/tao-run-deft-od-aoi/SKILL.md · Run a binary industrial-inspection DEFT loop with TAO RT-DETR: measure on frozen KPI/test roles, mine gap-similar real…
v0.1.0 · audit A · 95 stars

Every file in nvidia-tao/tao-skill-bank

Browse by kind, by grade A, or by owner.