skills AGENTS.md

skills AGENTS.md is an instructions file for Codex, OpenCode from iannuttall/skills. It costs 1,019 tokens per session, scanned A, original, MIT.

Repository instructions for maintaining Ian Nuttall’s public skills, including a skill that reconstructs PNG images as SVG files. SVG is a vector image format that stays sharp when resized.

In plain words
What is it for?
Use them when developing, testing, documenting, or packaging the skills repository, especially the png2svg skill and its Python package.
Why use it?
They explain the project’s design rule: image decomposition needs visual judgment, while measurements should remain exact and repeatable. They also prevent duplicate copies of the core package.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/iannuttall/skills/agents-md
Clone the repo
git clone --depth 1 https://github.com/iannuttall/skills

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for skills AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/iannuttall/skills/agents-md.svg)](https://agentmods.dev/instructions/iannuttall/skills/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/iannuttall/skills/agents-md"><img src="https://agentmods.dev/badge/instructions/iannuttall/skills/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,019 This file is loaded in full into every session.
When invoked 1,019 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.01019 $0.01019
Opus 5 $0.00509 $0.00509
Sonnet 5 $0.00204 $0.00204
Haiku 4.5 $0.00102 $0.00102

Measured 5d ago against content hash a8a9206a1399, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skills AGENTS.md 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 5d 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

AGENTS.md · 112 lines

How it starts

The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Working on Ian's agent skills

Every public skill lives at skills/<name>/. Keep each installed folder self-contained. Supporting tests, examples and private work belong outside the skill folder.

png2svg

This repo's product is the skill in skills/png2svg/. Everything else exists to build, test and document it.

The design principle

There is no single algorithm that reconstructs every logo. Decomposition — how many shapes, what covers what, which coincidences are real design constraints — needs eyes on the image. So:

  • the library makes every measurement deterministic and exact
  • the agent decides what to measure, what overlaps what, and when to stop

Do not try to collapse that into a one-shot reconstruct command. Effort is better spent making each primitive sharper, or moving a decision the agent keeps re-deriving into a call it can make once.

Layout gotcha

The Python package lives at skills/png2svg/scripts/png2svg/ — inside the skill, so the skill is self-contained. pyproject.toml points the build backend there via tool.uv.build-backend.module-root. There is exactly one copy. Never vendor a second one into src/; a drifting duplicate is worse than either copy alone.

skills/png2svg/          the deliverable
  SKILL.md               workflow; keep under 500 lines / 5000 tokens
  references/            loaded on demand — conventions, model, examples
  scripts/png2svg/       the engine
  scripts/*_template.py  what an agent copies and edits
examples/                real per-image reconstruction scripts
tests/                   ground-truth tests

Commands

uv sync
uv run pytest                    # must stay green
uv run png2svg --help
uvx --from skills-ref agentskills validate ./skills/png2svg

The bundled entry point must also work with nothing installed, since that is how the skill runs on someone else's machine:

uv run --no-project skills/png2svg/scripts/png2svg_cli.py --help

Testing: validate against ground truth

Read the full file on GitHub · 112 lines

Changes

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.

  1. 5d ago First seen · 112 lines · 1,019 tokens per session scan A a8a9206a1399

Subscribe to this mod's changes

skills AGENTS.md is an instructions file published in the GitHub repository iannuttall/skills (6 stars, last pushed 1mo ago), licensed MIT. It adds 1,019 tokens to every session, about $0.0051 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 2026-08-31.