Automated summary
Walks a workbench maintainer through creating a new skill: justification against existing skills, proposed name/description/triggers/evals, then scaffolding and script-based validation.
Strengths
- Enforces a reuse-before-creating analysis before any writing, requiring justification that a new skill beats extending the closest existing one
- Specifies the opening proposal precisely: jarvis-<workbench>-<verb-noun> naming, third-person description with a "Does not trigger" clause, collision-checked trigger phrases, and 3 eval scenarios
- Explicitly halts implementation until human approval, then requires ./lint-platform.sh to exit 0 before the work may be called done
- Ends with a structured after-report covering files touched, validation, versioning, documentation impact, and whether lessons.md needs an entry
Limitations
- Tightly coupled to the shipfoundry repo: docs/platform/* paths and the new-skill.sh, build-dist.sh, lint-platform.sh scripts must exist in that layout
- Assumes familiarity with the platform's class taxonomy (foundation / orchestrator / pattern / standalone) and its naming conventions
- Does not cover authoring a skill outside an existing workbench, nor publishing steps beyond lint
Use when adding a new skill to an existing shipfoundry workbench and you want to be guided from justification through validation.
Avoid it when editing an existing skill, bootstrapping a workbench from scratch, or working in a repo that doesn't follow the shipfoundry operating model and lint scripts.
Security analysis
CautionThe skill operates within a development workflow, using local scripts for scaffolding and validation. No destructive, exfiltrating, or obfuscated actions are present, but the script execution and file modifications warrant caution.
- •Instructs execution of local shell scripts (./new-skill.sh, ./build-dist.sh, ./lint-platform.sh) without declared allowed-tools; these scripts modify repository files and could have side effects if misused.
- •Requires updating CLAUDE.md and documentation, which writes to files.
Examples
/skill-new notre workbench a besoin d'une skill qui relit les migrations SQL avant merge, déclenchée quand on touche au dossier db/migrate — propose-la moi (justification, nom, description, triggers, evals) avant d'implémenter/skill-new je veux une skill qui génère des fixtures de test à partir d'un schéma Prisma — dis-moi d'abord si ça ne devrait pas être une extension d'une skill existante, puis propose le squelette/skill-new jarvis-api-audit-endpoints — la proposition est approuvée, passe au scaffolding, aux 3 evals, au build et au lint, puis donne-moi l'after-reportdescription: Start a new skill in an existing workbench — platform-owner mode, routed to the new-skill playbook + gates argument-hint: <what the skill does, who triggers it, which workbench>
You are the shipfoundry maintainer — operate per docs/platform/operating-model.md (Type A — new skill).
New skill requested: $ARGUMENTS
Before writing anything:
- Read docs/platform/operating-model.md, docs/platform/new-skill-playbook.md, and docs/platform/skill-taxonomy.md.
- Tell me why this should be a NEW skill and not an extension of the closest existing one (reuse before creating); its class (foundation / orchestrator / pattern / standalone); and where it sits in the platform.
- Propose: the name (jarvis-<workbench>-<verb-noun>), the description (third person, "Use when …" + a "Does not trigger" clause, ≤1024 chars, {braces} not angle brackets), the trigger phrases (verify they don't collide with existing skills), the reference files, and the 3 eval scenarios.
Stop there and get my approval before implementing.
After I approve:
4. Scaffold with ./new-skill.sh <name> <class>, then fill in SKILL.md + the 3 grounded evals.
5. ./build-dist.sh <name> then ./lint-platform.sh <name> — must exit 0.
6. Update CLAUDE.md skill counts + structure tree.
7. Give me the after-report: Summary · Files touched · Validation · Versioning · Documentation impact · whether lessons.md needs an entry. Do not claim done while ./lint-platform.sh is non-zero.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/create-new-skill-in-platform-ms8gxn9q/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/create-new-skill-in-platform-ms8gxn9q' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/create-new-skill-in-platform-ms8gxn9q/download' --output '.skill-downloads/create-new-skill-in-platform-ms8gxn9q/SKILL.md.part' && mv '.skill-downloads/create-new-skill-in-platform-ms8gxn9q/SKILL.md.part' '.skill-downloads/create-new-skill-in-platform-ms8gxn9q/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
This file does not provide a usable name/description header for this native procedure. It remains available to download and inspect; use the manual method below.
Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/create-new-skill-in-platform-ms8gxn9q/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.