agentsclimarketplace
Sign in

Easy explain

Skill memorysaver/agentic-engineering-patterns/skills/patterns/easy-explain

Stop. That last message did not land — re-pitch it.From its SKILL.md

Install
npx -y skills add memorysaver/agentic-engineering-patterns --skill easy-explain

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 17 stars17 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.1 KB, as published. Nobody here has run it

Wait — I don't understand where you've got to here. Re-pitch that: give me a little bit of context, talk in ASD-STE100 Simplified Technical English, and use this project's own nouns — product-context.yaml and docs/glossary.md where they exist.

The register (for orchestrator skills)

The repair line above is also AEP's standing explain-to-a-human register. Main-session skills that render status, escalations, or decisions for a person (/aep-autopilot status and escalations, /aep-dispatch handoff summaries, /aep-wrap layer-advance asks) write in it by default:

  1. One line of context first — what was happening when this became worth saying.
  2. ASD-STE100 Simplified Technical English — short sentences, one clause, active voice, one meaning per word.
  3. The project's own nouns — the ubiquitous language from product-context.yaml and the glossary, in place of invented terms and stacked acronyms.

A re-pitch is working if it is shorter and clearer, not shorter and blunter: it adds the premise that was missing instead of only deleting words, and it survives being invoked twice in a row without degrading into terseness.

This skill is user-typed only (/aep-easy-explain), and that is the design: only the human knows when they stopped following. The model reaching for it on its own would grade its own clarity — the same self-evaluation /aep-gen-eval exists to forbid. Each host enforces it with its own native metadata: Claude Code via disable-model-invocation: true in the frontmatter, Codex via agents/openai.yaml (policy.allow_implicit_invocation: false); explicit invocation stays available on both.

Adapted from mattpocock/skills productivity/wait-what — the three-line original, whose insight is that naming the listener's state ("wait, you lost me") repairs comprehension, where naming the output ("be concise") produces telegrams.

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.