magpie-write-skill

magpie-write-skill is a skill for Claude Code from apache/magpie. It costs 95 tokens per session (4,944 once invoked), scanned A, original, Apache-2.0.

A guide for writing or updating skills for the Apache Magpie framework, which is a system for extending coding agents with reusable instructions. It covers the framework's file format, conventions, safety checks, and validation tools.

In plain words
What is it for?
It helps scaffold new skills, refactor existing ones, add required metadata and resources, and validate the result with Magpie's tools.
Why use it?
It reduces guesswork when creating a Magpie skill and helps catch structural, security, or privacy problems before use.

Skill for Claude Code ✓ vendor

Written for Claude Code: when-to-use in frontmatter. Also seen: reads .claude/ paths; mentions Claude Code; mentions AGENTS.md.

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is │ │ │ [`docs/labels-and-capabilities.md`](../../docs/labels-and-capabilities.md)).

Part of the magpie plugin — 72 skills shipped together

Good fit It helps scaffold new skills, refactor existing ones, add required metadata and resources, and validate the result with Magpie's tools.

Compare 6 skills from other repositories ↓
About the project

Apache Magpie is a framework of agent-assisted workflows for maintaining Apache software projects, including issue triage, pull-request review, contributor mentoring, and security-report handling. Apache maintainers and developers use it to delegate repetitive project work to coding agents while retaining human review. Its catalogue entries are the skills, instructions, plugin, and settings that implement these workflows.

apache/magpie · 87 stars · on GitHub · magpie.apache.org

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/apache/magpie
agentmods
npx agentmods add skills/apache/magpie/write-skill

Made for: Claude Code.

Or install magpie, the plugin that ships this one along with the rest of its 72 skills.

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 magpie-write-skill

README.md
[![agentmods](https://agentmods.dev/badge/skills/apache/magpie/write-skill.svg)](https://agentmods.dev/skills/apache/magpie/write-skill)
Your own site
<a href="https://agentmods.dev/skills/apache/magpie/write-skill"><img src="https://agentmods.dev/badge/skills/apache/magpie/write-skill.svg" alt="Measured on agentmods" height="20"></a>
Per session 95 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,944 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00095 $0.04944
Opus 5 $0.00048 $0.02472
Sonnet 5 $0.00019 $0.00989
Haiku 4.5 $0.00010 $0.00494

Measured 3d ago against content hash 26b6cf579f7b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

magpie-write-skill scanned grade A with 1 finding 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 3d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/init_skill.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Reads agent configuration directorieslowAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

[framework `.claude/settings.json`](../../.claude/settings.json)

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

skills/write-skill/SKILL.md · 426 lines

How it starts

The opening of the file, as written. The whole thing — 426 lines — stays where its author put it; the contents beside it link to each section on GitHub.

write-skill

This skill walks the user through authoring a new skill for the Apache Magpie framework, or refactoring an existing one to pick up the framework's current conventions.

Provenance

This skill is adapted from the skill-creator skill in the JuliusBrussee/awesome-claude-skills repository, distributed under the Apache License 2.0. The upstream commit at the time of adoption is 5380239.

The framework's adaptations of the upstream content are substantial. They are summarised in the bullets below, in roughly the order they appear in this file. None of them are breaking-versus-upstream — anyone familiar with skill-creator will recognise the workflow shape:

  • Renamed from skill-creator to write-skill to match the framework's verb-prefixed naming convention. The trigger vocabulary in the when_to_use field includes both forms.
  • Frontmatter shape updated to the framework's schema: license: Apache-2.0 (not free-form licence text), when_to_use (the framework's convention) alongside description, SPDX comment + placeholder-convention comment after the frontmatter.
  • Step 3 (initialisation) uses the adapted scripts/init_skill.py that scaffolds the framework's expected structure (Adopter-overrides preamble, Snapshot-drift preamble, placeholder convention, SPDX header).
  • Step 5 (packaging) is dropped entirely — the framework distributes skills via the snapshot model documented in docs/setup/install-recipes.md, not as zip artefacts. The upstream's package_skill.py is not included; validation is performed by the existing tools/skill-and-tool-validator, which is the framework's superset of the upstream's quick_validate.py.
  • New Step 5 (security checklist) added — a hard walk-through of the prompt-injection-defence patterns that every framework skill ingesting external content must adopt. Sourced from the 2026-05 audit recorded at the gist. See the sibling security-checklist.md for the full pattern catalogue. This is the load-bearing adaptation: it ensures any new skill written through this flow inherits the lessons rather than rediscovering them in a future audit.

Read the full file on GitHub · 426 lines

Files

What ships with it

2 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.

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. 3d ago First seen · 426 lines · 95 tokens per session scan A 26b6cf579f7b

Subscribe to this mod's changes

magpie-write-skill is a skill published in the GitHub repository apache/magpie (87 stars, last pushed 5d ago), licensed Apache-2.0. It adds 95 tokens to every session and 4,944 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.