Skip to content
cccskills
FreePublished on GitHub

wtf

Make AI coding-agent messages to humans clear and easy to follow. Use for every user-facing update, explanation, review, question, and handoff in this repository, or when the user asks to simplify or rewrite an agent message.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md2.5 KB
  • agents/openai.yaml215 B

SKILL.md (original)

Review the instructions the agent will receive before you install.

Human-facing writing

Apply this guide to all prose written for people. This includes chat messages, status updates, reviews, questions, handoffs, and pull-request summaries. Preserve exact commands, identifiers, schemas, and required templates.

Make the message easy to understand

  • Assume the reader is a mid-level or senior software engineer with working AWS knowledge. Explain project-specific context and unfamiliar concepts. Do not assume the reader remembers the full session.
  • Lead with the outcome or the information the reader needs most. Add background only when it helps the reader understand a decision, consequence, or next step.
  • Use plain, conventional sentence structure. Prefer an explicit subject, active verb, and clear object.
  • Split dense sentences. If a sentence requires backtracking, shorten it or divide it.
  • Avoid detached afterthoughts after commas, such as “..., extracted.” Write a complete clause or a separate sentence.

Keep the structure purposeful

  • Group related information. Use sections only when each section answers a distinct reader question.
  • Make relationships between sections and steps obvious. Avoid a wall of text and avoid a collection of unrelated headings.
  • When a message asks several questions, keep them together, preferably at the end, instead of scattering them through the text or across several messages. This does not apply to the harness's native question tool: use that tool whenever the user has to decide something, and let it carry the questions.
  • In a handoff, state what changed, which checks ran, and what remains uncertain. Mention implementation details only when they help the reader understand or maintain the result.

Remove distracting language

  • Remove filler, repetition, unnecessary jargon, and needless qualifiers.
  • Avoid sycophancy and stock chatbot phrases such as “Certainly” or “Found the smoking gun.”
  • Avoid rhetorical, literary, promotional, and marketing language. State the fact directly.

When invoked manually without a new task, rewrite the previous agent message using this guide. If there is no previous message to rewrite, confirm the change in one short sentence. Do not recite the guide unless the user asks for it.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Run, diagnose, or manually test the private Stacktape Console locally, especially when local API/UI behavior must use the shared dev RDS data, Cognito users, AWS services, OAuth callbacks, or provider webhooks. Use for Console dev-mode setup, deciding between UI-only and full mode, shared-dev migrations, and GitHub/GitLab/Bitbucket end-to-end testing.

The English summary is being prepared. Showing the original description.

stacktape/stacktape81Updated Oct 19, 2026

load

Free

Load context from an earlier Claude Code or Codex session, selected by conversation name or recency. Explicit invocation only: `/load` in Claude Code, `$load` in Codex. Never trigger this from ordinary wording such as "load" or "continue".

The English summary is being prepared. Showing the original description.

stacktape/stacktape81Updated Oct 19, 2026

Redesign Stacktape's marketing website or Console through visual exploration, independent design critique, and browser verification. Use for substantial page, flow, or visual-system design work, including agent-led redesigns.

The English summary is being prepared. Showing the original description.

stacktape/stacktape81Updated Oct 19, 2026

wtf

Free

Make AI coding-agent messages to humans clear and easy to follow. Use for every user-facing message in this repository, or when the user asks to simplify or rewrite an agent message.

The English summary is being prepared. Showing the original description.

stacktape/stacktape81Updated Oct 19, 2026

See all skills by stacktape

Report a problem with this skill