Agent skill
eli5
Quellcode ansehen: mblode/agent-skills
Installation
npx skills add mblode/agent-skills --skill eli5252
Installationen
EU-hosted inference API
Power your AI agent skills with open-source models.
Drop-in OpenAI-compatible API. No data leaves Europe.
MiniMax
MiniMax M3
$0.50 / $1.90
per M tokens
Z.ai
GLM 5.3 Flash
$0.20 / $0.60
per M tokens
MoonshotAI
Kimi K3
$4.00 / $18.00
per M tokens
DeepSeek
DeepSeek V4.1 Flash
$0.40 / $1.40
per M tokens
Plain-language house style
- IS NOT: product copy or documentation (use
ghostwriter), or slide decks (usepresentation-creator).
Use this style for the explanation requested. Keep it for later replies only when the user asks for an ongoing mode; a single ELI5 request does not change the session permanently.
- Preserve exact identifiers, paths, commands, errors, numbers, and quoted source text. Explain around them.
- Avoid minimizers: simply, obviously, just, easy, of course, as you know.
- No em dashes in authored prose; do not substitute a spaced hyphen. For promotional vocabulary and other machine tells, run
ghostwriter/references/tells.md, the collection's one list; literal technical uses and quotations stay. - Use an analogy only when it clarifies the mechanism; identify its limit if that affects the answer. If an explanation did not land, change the framing instead of making the same analogy longer.
- Put the explanation or result first. Include a next action only when the reader needs to act. Do not assign the user work the agent is already authorized to complete.
Return the explanation itself. No activation announcement, fixed sentence count, compulsory recap, or pre-send checklist.
Pick the form
Prose is the default. Move up a rung when the user names the form, or when the content has a shape that prose hides. Each rung costs more to make and to check, so take the lowest one that carries the idea.
| Form | Use when | Cost |
|---|---|---|
| Plain prose | Most answers | None |
| Controlled English | User asks for STE100 or "Simplified Technical English", or the answer is a procedure or a dense mechanism a non-native reader must follow | None |
| Diagram | Parts and the links between them, a sequence over time, states, or a data path | Low |
| HTML page | The reader learns by changing an input and seeing the result, or the topic needs several linked views | Medium |
| Explainer video | User asks for a video, or the idea is motion (an algorithm stepping, a curve changing) | High: render, narration, review |
When the user names a form, use it; do not argue down to prose. When choosing on your own, never go past a diagram without asking, and offer the next rung in one line only when it would clearly help.
"80% STE100", "soft STE100", or similar means the relaxed profile in references/formats.md, not the full specification. Diagram, HTML, and video mechanics (tool choice, narration keys, review steps) are in the same file; load it before making any form other than plain prose.
Boundaries
This changes assistant prose and the explanations it makes. Product copy, technical docs, and README structure belong to ghostwriter, PR bodies to pr-creator, slide decks to presentation-creator, and data charts to an installed dataviz skill where one exists.
Maintenance
evals/evals.json contains behavior and routing scenarios for changes to this skill; do not load it during an explanation.
Verwandte Skills
Mehr aus dieser Quelle: mblode/agent-skills
So verwenden Sie diesen Skill
- 1
Install eli5 by running
npx skills add mblode/agent-skills --skill eli5in your project directory. Führen Sie den obigen Installationsbefehl in Ihrem Projektverzeichnis aus. Die Skill-Datei wird von GitHub heruntergeladen und in Ihrem Projekt platziert. - 2
Keine Konfiguration erforderlich. Ihr KI-Agent (Claude Code, Cursor, Windsurf usw.) erkennt installierte Skills automatisch und nutzt sie als Kontext bei der Code-Generierung.
- 3
Der Skill verbessert das Verständnis Ihres Agenten für eli5, und hilft ihm, etablierte Muster zu befolgen, häufige Fehler zu vermeiden und produktionsreifen Code zu erzeugen.
Was Sie erhalten
Skills sind Klartext-Anweisungsdateien — kein ausführbarer Code. Sie kodieren Expertenwissen über Frameworks, Sprachen oder Tools, das Ihr KI-Agent liest, um seine Ausgabe zu verbessern. Das bedeutet null Laufzeit-Overhead, keine Abhängigkeitskonflikte und volle Transparenz: Sie können jede Anweisung vor der Installation lesen und prüfen.
Kompatibilität
Dieser Skill funktioniert mit jedem KI-Coding-Agenten, der das skills.sh-Format unterstützt, einschließlich Claude Code (Anthropic), Cursor, Windsurf, Cline, Aider und anderen Tools, die projektbezogene Kontextdateien lesen. Skills sind auf Transportebene framework-agnostisch — der Inhalt bestimmt, für welche Sprache oder welches Framework er gilt.
Data sourced from the skills.sh registry and GitHub. Install counts and security audits are updated regularly.