Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add NVIDIA-TAO/tao-skill-bank --skill tao-generate-referring-expressionsgit clone --depth 1 https://github.com/NVIDIA-TAO/tao-skill-bankWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-generate-referring-expressions)<a href="https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-generate-referring-expressions"><img src="https://agentmods.dev/badge/skills/nvidia-tao/tao-skill-bank/tao-generate-referring-expressions/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-generate-referring-expressions"><img src="https://agentmods.dev/badge/skills/nvidia-tao/tao-skill-bank/tao-generate-referring-expressions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 74 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00143 | $0.02656 |
| Opus 5 | $0.00072 | $0.01328 |
| Sonnet 5 | $0.00029 | $0.00531 |
| Haiku 4.5 | $0.00014 | $0.00266 |
Grade A, and why
tao-generate-referring-expressions scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Image Referring Expression Pipeline
Standalone install? If this session was not initialized by the TAO skill bank plugin, run the
tao-setupskill 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)
- Step 0 (region_expr) — VLM emits one short discriminative phrase per KITTI bbox (
bbox_2d,type,color,description). - Step 1 (image_caption) — VLM emits a holistic, location-agnostic scene caption.
- Step 2 (grounding_expr) — VLM groups Step 0 objects into grouping phrases and returns one bbox list per group, optionally using Step 1's caption as extra context.
- Step 3 (double_check) — VLM re-checks each Step 2 bbox against the image; bad matches are removed, slightly-off boxes get tightened.
Steps 0 and 1 run in parallel within a single thread pool (they only depend on the seed records). Each step writes its own step_<N>_*/annotations.jsonl and skips already-processed images on re-run unless workflow.force_reprocess: true.
Instructions
Initial setup
When a user wants to run this pipeline, walk through these steps:
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 145 lines · 143 tokens per session scan A 8db9b45d1c2e
tao-generate-referring-expressions is a skill published in the GitHub repository NVIDIA-TAO/tao-skill-bank (88 stars, last pushed today), licensed Apache-2.0. It adds 143 tokens to every session and 2,656 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
vision
Nested swiss-knife reference for image understanding — a decision tree that routes between three paths depending on what the agent has access to: (1) the built-in vision tool if the LLM provider supports image input, (2) the sibling minimax-cli reference if a usable MiniMax preset/key slot is available, or (3) a local…
evalscope
LLM evaluation & inference performance testing via the evalscope CLI. Translates natural language requests into evalscope commands for: (1) Model accuracy evaluation — runs registered benchmarks against local checkpoints or API endpoints (OpenAI-compatible, Anthropic, LiteLLM); (2) Performance stress testing — TTFT…
clawhub
Dingo: A Comprehensive AI Data, Model and Application Quality Evaluation Tool.
multimodal-llm
Vision, audio, video generation, and multimodal LLM integration patterns. Use when processing images, transcribing audio, generating speech, generating AI video (Kling v3, Sora 2, Veo 3.1 std/lite/fast, Runway Gen-4.5 via gen4turbo), or building multimodal AI pipelines.
vss-build-vision-ai
Add agent-ready vision capabilities — dense captioning, detection, search, alerting, summarization — to an agent or application through a customizable, self-contained vision stack built on the NVIDIA VSS Blueprint. Use this skill when a developer or agent wants to give their app vision: pick capabilities via guided…
benchmark-vlm-qa
Benchmark video Q&A accuracy and latency of a deployed RT-VLM (Cosmos Reason 3) via vss vlm run, using questions and videos from the DSS vss-devx-base dataset. Replaces the deprecated nat eval / vss-agent QA path. Not for tool-calling or trajectory evaluation, and not for LVS summarization throughput.