Agent skill

Onnxtxt

Read or write ONNX text format ("onnxtxt"). An agent skill from onnx/onnx.

Apache-2.0Auto-check passedAI & LLM Engineering

Install Onnxtxt

skills CLI

$ npx skills add onnx/onnx --skill onnxtxt -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install onnx/onnx onnxtxt --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/onnx/onnx.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/onnxtxt .claude/skills/onnxtxt && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "onnxtxt" agent skill from https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt into .claude/skills/onnxtxt/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "onnxtxt", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add onnx/onnx --skill onnxtxt -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install onnx/onnx onnxtxt --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/onnx/onnx.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/onnxtxt .agents/skills/onnxtxt && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "onnxtxt" agent skill from https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt into .agents/skills/onnxtxt/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "onnxtxt", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add onnx/onnx --skill onnxtxt -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install onnx/onnx onnxtxt --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/onnx/onnx.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/onnxtxt .cursor/skills/onnxtxt && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "onnxtxt" agent skill from https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt into .cursor/skills/onnxtxt/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "onnxtxt", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/onnx/onnx.git --path .agents/skills/onnxtxt

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add onnx/onnx --skill onnxtxt -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install onnx/onnx onnxtxt --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/onnx/onnx.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/onnxtxt .gemini/skills/onnxtxt && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "onnxtxt" agent skill from https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt into .gemini/skills/onnxtxt/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "onnxtxt", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install onnx/onnx onnxtxt

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add onnx/onnx --skill onnxtxt -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/onnx/onnx.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/onnxtxt .github/skills/onnxtxt && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "onnxtxt" agent skill from https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt into .github/skills/onnxtxt/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "onnxtxt", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add onnx/onnx --skill onnxtxt -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install onnx/onnx onnxtxt --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/onnx/onnx.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/onnxtxt .opencode/skills/onnxtxt && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "onnxtxt" agent skill from https://github.com/onnx/onnx/tree/main/.agents/skills/onnxtxt into .opencode/skills/onnxtxt/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "onnxtxt", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: onnxtxt
GitHub stars: 22k
Token cost: ~1.3k tokens
SKILL.md length: 395 words
Files: 1
Skills in repo: 4
Repo updated: Oct 7, 2026
First seen: Oct 7, 2026
Licence: Apache-2.0

At a glance

Read or write ONNX text format ("onnxtxt"). An agent skill from onnx/onnx.

Authoring .FunctionBody(RONNX(...)) blocks
SKILL.md covers Where the format appears, Core syntax, Argument-order convention and Testing idioms, plus 2 more sections
Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
Writing tests with onnx.parser.parsemodel / parsegraph

What it does

Onnxtxt is an agent skill from onnx/onnx. Read or write ONNX text format ("onnxtxt"). Use when authoring .FunctionBody(R"ONNX(...)") blocks, writing tests with onnx.parser.parsemodel / parsegraph, using the C++ OnnxParser, debugging parser errors, or interpreting Constant <value = ... and body-subgraph syntax.

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering, covering Deep learning. It works with ONNX, C++ and Python. The repository describes itself as: Open standard for machine learning interoperability. The licence is Apache-2.0.

When your agent uses it

Authoring .FunctionBody(RONNX(...)) blocks
Writing tests with onnx.parser.parsemodel / parsegraph
Using the C++ OnnxParser
Debugging parser errors

Example prompts

“onnxtxt”
“ONNX(...)”
“/onnxtxt”

Requirements

Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit ab429ea. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are python and cpp).
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Onnxtxt loads about 1.3k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 395 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~73
When it runs · the whole SKILL.md, loaded when a task matches: ~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from onnx/onnx at commit ab429ea, republished under its Apache-2.0 licence (© onnx). 395 words, ~1,319 tokens.

Download SKILL.mdSave it as .claude/skills/onnxtxt/SKILL.md (or your agent's skills folder).

name: onnxtxt
description: Read or write ONNX text format ("onnxtxt"). Use when authoring `.FunctionBody(R"ONNX(...)")` blocks, writing tests with `onnx.parser.parse_model` / `parse_graph`, using the C++ `OnnxParser`, debugging parser errors, or interpreting `Constant <value = ...>` and body-subgraph syntax.

ONNX has a compact text format implemented by onnx/parser.py (Python) and onnx/defs/parser.{h,cc} (C++). The formal grammar lives in docs/Syntax.md; this skill captures the practical conventions, idioms, and gotchas that matter when authoring or reviewing code that uses it.

Where the format appears

Surface	API
C++ function bodies	`.FunctionBody(R"ONNX( ... )ONNX")` and `FunctionBuilder::Add(...)`
Python test fixtures	`onnx.parser.parse_model("...")`, `onnx.parser.parse_graph("...")`
C++ tests	`OnnxParser` in `onnx/defs/parser.h` — parse a model, then call `shape_inference::InferShapes`

Core syntax

<var> = <OpName> <attr1 = value, attr2 = value> (<input1>, <input2>)

Variables on the LHS are local to the surrounding graph (function body, subgraph, or model main graph).
Inputs/outputs of a function body must match the names declared in the schema's .Input(...) / .Output(...).
Constants: Const = Constant <value = float {0.0}>() or Alpha = Constant <value_float: float = @alpha>().
Use CastLike (not Cast) when the target dtype depends on another input.
Reference enclosing-op attributes with @attr_name (only inside function bodies, and only for attributes declared on the schema).

Argument-order convention

Simple scalar/tensor attributes — keep the conventional Op<attrs>(inputs) form; reads well on one line:

Y = Transpose<perm = [2, 0, 1]>(X)

Subgraph attributes (Scan, Loop, If, ScanVarLen, …) — prefer Op(inputs)<body = ...>. The body spans multiple lines, so putting inputs first keeps the call site readable:

so, xo = Scan (s, x) <
    num_scan_inputs = 1,
    body = scan_body (float[1] s_in, float[1] x_in) => (float[1] s_out, float[1] x_out) {
        s_out = Add(s_in, x_in)
        x_out = Identity(x_in)
    }
>

Testing idioms

Test file	Recommendation
`tests/python/shape_inference_test.py`, `tests/python/reference_evaluator_test.py`	Use `onnx.parser.parse_model(...)` for one-off fixtures.
`onnx/backend/test/case/node/<op>.py`	Keep the outer `helper.make_node` + `expect(...)` (it drives data generation). For body-subgraph ops, build the body with `onnx.parser.parse_graph`.
`tests/cpp/shape_inference_test.cc`	Use `OnnxParser` (`onnx/defs/parser.h`); pair with `shape_inference::InferShapes`.
`tests/python/version_converter/automatic_upgrade_test.py` and similar harnesses	Keep the established `_test_op_upgrade` / `_test_op_downgrade` style — do not rewrite.

Empirical: PR #7962 (ScanVarLen) cut ~58–70% of test LOC by switching to parser-based fixtures.

Show full SKILL.md (152 more words)Show less

Python example — shape inference fixture

python

import onnx
import onnx.parser
import onnx.shape_inference

model = onnx.parser.parse_model("""
    <ir_version: 8, opset_import: ["" : 18]>
    g (float[2, 3, 4] X) => (float[4, 2, 3] Y) {
        Y = Transpose<perm = [2, 0, 1]>(X)
    }
""")
inferred = onnx.shape_inference.infer_shapes(model, strict_mode=True)

Python example — body subgraph for a node test

python

body = onnx.parser.parse_graph("""
    b (float[1] s, float[1] xi) => (float[1] so, float[1] xo) {
        so = Identity(s)
        xo = Identity(xi)
    }
""")
node = onnx.helper.make_node("Scan", ["s", "x"], ["so", "xo"], body=body, num_scan_inputs=1)

C++ example — shape inference test

cpp

#include "onnx/defs/parser.h"
#include "onnx/shape_inference/implementation.h"

ModelProto model;
OnnxParser parser(R"ONNX(
    <ir_version: 8, opset_import: ["" : 18]>
    g (float[2, 3, 4] X) => (Y) {
        Y = Transpose<perm = [2, 0, 1]>(X)
    }
)ONNX");
auto status = parser.Parse(model);
ASSERT_TRUE(status.IsOK()) << status.ErrorMessage();
shape_inference::InferShapes(model);

Gotchas

unk__* materialization in C++ shape-inference tests. Under InferShapes, unset output dims are materialized by MaterializeSymbolicShape into dim_param names like unk__0, unk__1, … Assertions on free dims must accept either an unset dim or an unk__* placeholder — write (or use) a helper that treats both forms as equivalent.
Variable-name collisions. Local variables in a function body must not reuse declared input/output names of the enclosing op.
@attr_name scope. Only valid inside a function body, and only for attributes declared on the enclosing schema's .Attr(...) calls.
CastLike vs Cast. Use CastLike when the desired target dtype is determined by another input; Cast requires a static to attribute.

References

Resource	Path
Formal grammar	docs/Syntax.md
C++ parser	`onnx/defs/parser.h`, `onnx/defs/parser.cc`
Python parser	`onnx/parser.py`
C++ parser tests	`tests/cpp/parser_test.cc`
Python parser tests	`tests/python/parser_test.py`
Empirical LOC win	PR #7962 (ScanVarLen test rewrite)

Files

Just SKILL.md in .agents/skills/onnxtxt of onnx/onnx.

Open the folder on GitHubat commit ab429ea

Compare with similar skills

Onnxtxt next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Onnxtxt compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Onnxtxt this skillonnx/onnx	22k	—	~1.3k	Automated safety check: Pass	Apache-2.0	yesterday
Quark Installamd/Quark	181	—	~1.8k	Automated safety check: Notes	MIT	10 days ago
Technology Selectiondotnet/skills	5.6k	2 repos	~2.1k	Automated safety check: Pass	MIT	today
Segment Anything Model GuideOrchestra-Research/AI-Research-SKILLs	13k	9 repos	~3.3k	Automated safety check: Pass	MIT	3 mo ago
Paddle BuildPaddlePaddle/Paddle	24k	—	~1k	Automated safety check: Pass	Apache-2.0	8 days ago
Paddle Eager GraphPaddlePaddle/Paddle	24k	—	~562	Automated safety check: Pass	Apache-2.0	8 days ago

Similar skills

Quark Install
amd/Quark
Install or verify the AMD Quark package and its dependencies.
181 GitHub stars~1.8k tokensUpdated 10 days ago
AI & LLM EngineeringAuto-check: notes
Technology Selection
dotnet/skills
Official
Guides technology selection and implementation of AI and ML features in .NET 8+ applications using ML.NET, Microsoft.Extensions.AI (MEAI), Microsoft Agent Framework (MAF), GitHub Copilot SDK, ONNX…
5.6k GitHub starsUsed in 2 repos~2.1k tokens
AI & LLM EngineeringAuto-check passed
Segment Anything Model Guide
Orchestra-Research/AI-Research-SKILLs
Guide to using Meta's Segment Anything Model for zero-shot image segmentation with point, box or mask prompts, or automatic mask generation.
13k GitHub starsUsed in 9 repos~3.3k tokens
AI & LLM EngineeringAuto-check passed
Paddle Build
PaddlePaddle/Paddle
A skill your agent uses when needing to compile, rebuild, or install Paddle from source after code changes.
24k GitHub stars~1k tokensUpdated 8 days ago
AI & LLM EngineeringAuto-check passed
Paddle Eager Graph
PaddlePaddle/Paddle
A skill your agent uses when navigating Paddle eager-mode (dynamic graph) source code, tracing forward/backward execution, debugging autograd issues, understanding PyLayer, or investigating…
24k GitHub stars~562 tokensUpdated 8 days ago
DevelopmentAuto-check passed
ExecuTorch Build Guide
pytorch/executorch
Builds ExecuTorch from source: the Python package, C++ runtime, model runners, Android and iOS cross-compilation and backend-specific builds, with environment checks.
5.1k GitHub stars~2.3k tokensUpdated today
DevelopmentAuto-check: notes

Questions about Onnxtxt

What does Onnxtxt do?

Read or write ONNX text format ("onnxtxt"). An agent skill from onnx/onnx. Onnxtxt is an agent skill from onnx/onnx. Read or write ONNX text format ("onnxtxt").

When should I use Onnxtxt?

Onnxtxt fits situations like: authoring .FunctionBody(RONNX(...)) blocks; writing tests with onnx.parser.parsemodel / parsegraph; using the C++ OnnxParser; debugging parser errors.

How do I install Onnxtxt in Claude Code?

Run `npx skills add onnx/onnx --skill onnxtxt -a claude-code`. Or copy the skill folder (.agents/skills/onnxtxt in onnx/onnx) into .claude/skills/onnxtxt in your project. Claude Code loads it when a task matches its description.

How do I install Onnxtxt in Codex?

Run `npx skills add onnx/onnx --skill onnxtxt -a codex`. Or copy the skill folder (.agents/skills/onnxtxt in onnx/onnx) into .agents/skills/onnxtxt in your project. Codex loads it when a task matches its description.

Can I use Onnxtxt in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add onnx/onnx --skill onnxtxt -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/onnxtxt, .gemini/skills/onnxtxt, .github/skills/onnxtxt and .opencode/skills/onnxtxt in your project.

What does Onnxtxt need to run?

SKILL.md names no scripts, command-line tools or credentials: Onnxtxt is instructions for the agent only. Our summary lists: Python 3.

Does Onnxtxt access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Onnxtxt safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Onnxtxt use?

Onnxtxt is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Onnxtxt use?

About 1.3k tokens (SKILL.md is roughly 5.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Onnxtxt?

Skills that share tags, products or a category with Onnxtxt: Quark Install (amd/Quark, 181 stars), Technology Selection (dotnet/skills, 5.6k stars), Segment Anything Model Guide (Orchestra-Research/AI-Research-SKILLs, 13k stars) and Paddle Build (PaddlePaddle/Paddle, 24k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Onnxtxt?

onnx (a GitHub organization) maintains it in onnx/onnx, which has 21,564 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on October 7, 2026.

Source: onnx/onnx on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Onnxtxt

Install Onnxtxt

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Where the format appears

Core syntax

Argument-order convention

Testing idioms

Python example — shape inference fixture

Python example — body subgraph for a node test

C++ example — shape inference test

Gotchas

References

Files

Compare with similar skills

Similar skills

Quark Install

Technology Selection

Segment Anything Model Guide

Paddle Build

Paddle Eager Graph

ExecuTorch Build Guide

More from onnx/onnx

Add Op

Add Function Body

Add Shape Inference

Questions about Onnxtxt

Install Onnxtxt

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Where the format appears

Core syntax

Argument-order convention

Testing idioms

Python example — shape inference fixture

Python example — body subgraph for a node test

C++ example — shape inference test

Gotchas

References

Files

Compare with similar skills

Similar skills

More from onnx/onnx

Related

Questions about Onnxtxt