Agent skill

Create Skill

by PINA-org in PINA-org/PINA

Create, modify, and improve PINA skills. An agent skill from PINA-org/PINA.

MITAuto-check passedAgent Workflows

Install Create Skill

skills CLI
$ npx skills add PINA-org/PINA --skill create-skill -a claude-code

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

GitHub CLI
$ gh skill install PINA-org/PINA create-skill --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/PINA-org/PINA.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.opencode/skills/create-skill .claude/skills/create-skill && 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/

Facts

Skill name
create-skill
GitHub stars
797
Token cost
~1.8k tokens
SKILL.md length
598 words
Files
1
Skills in repo
10
Repo updated
First seen
Licence
MIT

At a glance

Create, modify, and improve PINA skills. An agent skill from PINA-org/PINA.

  • Works in 5 steps: Capture intent → Study existing skills as templates → Write the SKILL.md → …
  • The user wants to create a new skill for PINA development guidance
  • SKILL.md covers Overview, Step 1 — Capture intent, Step 2 — Study existing skills… and Step 3 — Write the SKILL.md, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Create Skill is an agent skill from PINA-org/PINA. Create, modify, and improve PINA skills. Use when the user wants to create a new skill for PINA development guidance, edit an existing PINA skill, or add cross-tool compatibility (opencode, Claude Code, Codex CLI) to a skill. Covers skill directory structure, SKILL.md front matter conventions, symlinks for Claude Code compatibility, and updating AGENTS.md for Codex CLI.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts. Compatibility notes: opencode, codex, claude

It sits in Agent Workflows, covering Markdown, Deep learning and Agent instruction files. The repository describes itself as: Physics-Informed Neural networks for Advanced modeling. The licence is MIT.

When your agent uses it

  • The user wants to create a new skill for PINA development guidance
  • Edit an existing PINA skill
  • Add cross-tool compatibility (opencode
  • Codex CLI) to a skill

Example prompts

  • “/create-skill”

Requirements

  • Compatibility (from SKILL.md): opencode, codex, claude

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Capture intent
  2. Study existing skills as templates
  3. Write the SKILL.md
  4. Verify cross-tool setup
  5. Review and iterate

What it can do on your machine

Read from SKILL.md and the folder at commit 0cd8afb. 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 bash, yaml and markdown).

    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.

  • Compatibility

    opencode, codex, claude

    From compatibility in the SKILL.md frontmatter.

Context cost

Create Skill loads about 1.8k tokens when it runs. Until then it costs about 96 tokens; SKILL.md has 598 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~96
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k

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 PINA-org/PINA at commit 0cd8afb, republished under its MIT licence (© PINA-org). 598 words, ~1,809 tokens.

Download SKILL.mdSave it as .claude/skills/create-skill/SKILL.md (or your agent's skills folder).
name
create-skill
description
Create, modify, and improve PINA skills. Use when the user wants to create a new skill for PINA development guidance, edit an existing PINA skill, or add cross-tool compatibility (opencode, Claude Code, Codex CLI) to a skill. Covers skill directory structure, SKILL.md front matter conventions, symlinks for Claude Code compatibility, and updating AGENTS.md for Codex CLI.
compatibility
opencode, codex, claude
license
MIT
metadata.audience
users
metadata.workflow
skill-creation

Create a PINA Skill

[!IMPORTANT] Read RULES.md before using this skill — it applies to all skills.

Use this skill to create new PINA development skills or modify existing ones. Each PINA skill lives at .opencode/skills/<name>/SKILL.md and must be made available to all three tools (opencode, Claude Code, Codex CLI).

Overview

A PINA skill is a markdown file with YAML front matter that provides guided instructions for a specific PINA development task. Skills are auto-discovered by opencode, symlinked for Claude Code, and discoverable via the root AGENTS.md for Codex CLI.

What makes a good PINA skill
  • Narrow scope — One focused task (e.g. defining domains, not "all of PINA")
  • Concrete templates — Copy-paste-ready code snippets using PINA's API
  • Checklists — Verifiable steps so the user knows when they're done
  • Progressive disclosure — Start with the common case, then branch to edge cases
  • Why, not just what — Explain why PINA conventions exist (e.g. why operators trigger a backward pass per call)
Anatomy of a skill
.opencode/skills/<name>/
├── SKILL.md (required)
│   ├── YAML front matter (name, description, license, compatibility, metadata)
│   └── Markdown instructions
└── (optional resources)

Step 1 — Capture intent

Understand what the user wants the skill to do:

  1. What PINA task should this skill guide? (e.g. "setting up boundary conditions", "defining custom PDEs", "discretising domains")
  2. Who is the audience? (new PINA users, experienced users, or both)
  3. What existing skill is closest? — Check .opencode/skills/ for related skills. If this skill extends an existing workflow, reference it.
  4. What PINA APIs does it need to cover? — List the classes, functions, and import paths.

Step 2 — Study existing skills as templates

Read the existing PINA skills for style and structure. Key patterns:

FilePurposeReference pattern
create-problem/SKILL.mdEntry point, decision tree, templatesMultiple templates per audience, step-by-step flow, checklist
define-equations/SKILL.mdEquation definitionZoo reference table, custom signatures, edge cases
define-domains/SKILL.mdDomain creationDomain types table, common patterns, discretisation
condition-setup/SKILL.mdCondition bindingCondition types table, data type options
Front matter template

Use the same front matter format as existing skills:

yaml
---
name: <kebab-case-name>
description: >-
  One or two sentences describing when this skill triggers. Include specific
  user phrases and contexts that should activate it. Be slightly "pushy" —
  err on the side of including near-miss use cases so the skill triggers when
  it should.
license: MIT
compatibility: opencode, codex, claude
metadata:
  audience: users
  workflow: <workflow-name>
---
Writing guidelines
  • Imperative voice — "Define the domains", "Create the condition dict"
  • Explain the why — LLMs work better when they understand the reasoning behind conventions
  • Use tables for API references and option lists (consistent with existing skills)
  • Provide templates — At least one complete code template the user can adapt
  • Checklists at the end — Verifiable completion criteria
  • Front matter description is the primary trigger — include when-to-use context that matches real user phrasing
Show full SKILL.md (231 more words)Show less

Step 3 — Write the SKILL.md

Organise content with progressive disclosure:

markdown
# [Title]

> [!IMPORTANT]
> Read [RULES.md](../RULES.md) before using this skill — it applies to all skills.
> (If sub-skill: This is a sub-skill of **parent-skill**. Load the entry-point skill first.)

Use this skill to [concise purpose statement].

## Step 1 — [First step]
...
## Step N — [Last step]

## Templates
...
## Checklist
- [ ] ...
Cross-tool compatibility

Every new skill must be made available to all three tools. Include this checklist item and guide the user through the steps:

  1. opencode — The skill at .opencode/skills/<name>/SKILL.md is auto-discovered. No extra step needed.
  2. Claude Code — Create a symlink:
    bash
    mkdir -p .claude/skills/<name>
    ln -sf ../../../.opencode/skills/<name>/SKILL.md .claude/skills/<name>/SKILL.md
  3. Codex CLI — Create a symlink (same pattern):
    bash
    mkdir -p .agents/skills/<name>
    ln -sf ../../../.opencode/skills/<name>/SKILL.md .agents/skills/<name>/SKILL.md
    The relative path ../../../.opencode/skills/<name>/SKILL.md works from both symlink directories (3 .. to root, then into .opencode/).

Step 4 — Verify cross-tool setup

After writing the skill, verify both symlinks resolve:

bash
test -f .claude/skills/<name>/SKILL.md && echo "claude: OK"
test -f .agents/skills/<name>/SKILL.md && echo "codex: OK"

Step 5 — Review and iterate

Present the skill to the user for review. Check:

  • Does the front matter description clearly convey when to use this skill?
  • Are the templates syntactically correct PINA code?
  • Is the cross-tool compatibility checklist complete?
  • Does the skill reference RULES.md and link to related skills?

Templates

Template 1: New PINA skill (minimal)
yaml
---
name: my-new-skill
description: >-
  Guides users through [task]. Use when the user asks about [topic] or needs
  help with [specific problem].
license: MIT
compatibility: opencode, codex, claude
metadata:
  audience: users
  workflow: my-workflow
---

After creating .opencode/skills/my-new-skill/SKILL.md:

bash
mkdir -p .claude/skills/my-new-skill
ln -sf ../../../.opencode/skills/my-new-skill/SKILL.md .claude/skills/my-new-skill/SKILL.md
mkdir -p .agents/skills/my-new-skill
ln -sf ../../../.opencode/skills/my-new-skill/SKILL.md .agents/skills/my-new-skill/SKILL.md
test -f .claude/skills/my-new-skill/SKILL.md && echo "claude: OK"
test -f .agents/skills/my-new-skill/SKILL.md && echo "codex: OK"

Checklist

  • name is kebab-case, unique across .opencode/skills/
  • description front matter includes trigger context (when to use)
  • compatibility: opencode, codex, claude is set
  • metadata.audience and metadata.workflow are set
  • Skill references RULES.md at the top
  • Sub-skills reference their parent entry-point skill
  • At least one complete code template is provided
  • Cross-tool setup is complete:
    • .claude/skills/<name>/SKILL.md symlink exists and resolves
    • .agents/skills/<name>/SKILL.md symlink exists and resolves

© PINA-org, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .opencode/skills/create-skill of PINA-org/PINA.

Open the folder on GitHubat commit 0cd8afb

Compare with similar skills

Create Skill 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.

Create Skill compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Create Skill this skillPINA-org/PINA797—~1.8kAutomated safety check: PassMIT
Writing Docsc15t/c15t1.9k—~872Automated safety check: PassApache-2.0
Lint Repository Markdowncodsen/codsen214—~832Automated safety check: PassMIT
Creating agents.md Filespr-pm/prpm122—~2kAutomated safety check: PassMIT
EnforceQuentinCody/interlinked-cli178—~19kAutomated safety check: PassMIT
Render HTMLwanshuiyin/ARIS-in-AI-Offer580—~4.9kAutomated safety check: NotesMIT

Similar skills

  • Writing Docs

    c15t/c15t

    Author or edit c15t documentation in docs//.mdx — the source for both the c15t.com site and the docs bundled into published packages.

    1.9k GitHub stars~872 tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Keep the repository's linted Markdown passing npm run lint:markdown.

    214 GitHub stars~832 tokensUpdated 9 days ago
    DevelopmentAuto-check passed
  • Guides writing an agents.md project-context file: plain markdown with no frontmatter, focused on what an AI coding assistant cannot already know.

    122 GitHub stars~2k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Enforce

    QuentinCody/interlinked-cli

    Distill imperative markdown guidance (AGENTS.md, CLAUDE.md, .clinerules/, GEMINI.md, SKILL.md with hard imperatives) into deterministic Interlinked harness hook rules with verbatim source provenance.

    178 GitHub stars~19k tokensUpdated 6 days ago
    Agent WorkflowsAuto-check passed
  • Render HTML

    wanshuiyin/ARIS-in-AI-Offer

    Render an ARIS Markdown / JSON artifact (IDEAREPORT, AUTOREVIEW, KILLARGUMENT, PAPERPLAN, research-wiki state, etc.) into a single-file HTML view designed for human reading.

    580 GitHub stars~4.9k tokensUpdated 2 days ago
    AI & LLM EngineeringAuto-check: notes
  • Issue Triage

    intel/torch-xpu-ops

    Official

    Shallow, text-only triage of a GitHub issue on pytorch or torch-xpu-ops.

    115 GitHub stars~2.5k tokensUpdated today
    AI & LLM EngineeringAuto-check passed

More from PINA-org/PINA

All 10 skills in this repo
  • Pina Workflow

    PINA-org/PINA

    Orchestrates a complete session from problem definition through trained solver.

    797 GitHub stars~1.1k tokensUpdated 2 days ago
    Auto-check passed
  • Select Model

    PINA-org/PINA

    Guides users through selecting, configuring, or creating a PyTorch model for a PINA problem.

    797 GitHub stars~2.4k tokensUpdated 2 days ago
    Auto-check passed
  • Select Solver

    PINA-org/PINA

    Guides users through selecting the right PINA solver for their problem, or creating a custom solver when no built-in fits.

    797 GitHub stars~2.9k tokensUpdated 2 days ago
    Auto-check passed
  • Select Trainer

    PINA-org/PINA

    Guides users through configuring the PINA Trainer for solver training, including batching strategy, data splitting, and Lightning options.

    797 GitHub stars~1.3k tokensUpdated 2 days ago
    Auto-check passed
  • Skill Sync Checker

    PINA-org/PINA

    Audits SKILL.md files in this repo's skills directory for references to functions, classes, or modules (mentioned by name in prose, e.g.

    797 GitHub stars~1.5k tokensUpdated 2 days ago
    Auto-check passed
  • Condition Setup

    PINA-org/PINA

    Set up conditions for PINA problems. An agent skill from PINA-org/PINA.

    797 GitHub stars~1k tokensUpdated 2 days ago
    Auto-check passed

Questions about Create Skill

What does Create Skill do?

Create, modify, and improve PINA skills. An agent skill from PINA-org/PINA. Create Skill is an agent skill from PINA-org/PINA. Create, modify, and improve PINA skills.

When should I use Create Skill?

Create Skill fits situations like: the user wants to create a new skill for PINA development guidance; edit an existing PINA skill; add cross-tool compatibility (opencode; Codex CLI) to a skill.

How do I install Create Skill in Claude Code?

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

How do I install Create Skill in Codex?

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

Can I use Create Skill 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 PINA-org/PINA --skill create-skill -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/create-skill, .gemini/skills/create-skill, .github/skills/create-skill and .opencode/skills/create-skill in your project.

What does Create Skill need to run?

SKILL.md names no scripts, command-line tools or credentials: Create Skill is instructions for the agent only. Compatibility (from SKILL.md): opencode, codex, claude.

Does Create Skill 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 Create Skill 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 Create Skill use?

Create Skill is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Create Skill use?

About 1.8k tokens (SKILL.md is roughly 7.2k 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 Create Skill?

Skills that share tags, products or a category with Create Skill: Writing Docs (c15t/c15t, 1.9k stars), Lint Repository Markdown (codsen/codsen, 214 stars), Creating agents.md Files (pr-pm/prpm, 122 stars) and Enforce (QuentinCody/interlinked-cli, 178 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Create Skill?

PINA-org (a GitHub organization) maintains it in PINA-org/PINA, which has 797 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on October 6, 2026.

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