Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Mode Writing

ASecurity

Writing mode: distill prose, fight length and jargon, loop drafts through the AI-tell catalog until a full pass finds nothing. Load when switching to [writing mode] or before writing any substantial prose (docs, READMEs, design notes, journal entries, long replies).

1,992 stars
0 votes
0 copies
0 views
Added 10/4/2026
ai-agentsgoreactnode

Works with

cli

Security Analysis

A100/100

Scanned 10/4/2026

$npx -y skills add WeaveMindAI/weft --skill mode-writing --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mode Writing?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Mode Writing
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/weavemindai-mode-writing/badge)](https://www.skillsdirectory.com/skills/weavemindai-mode-writing)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: mode-writing
description: "Writing mode: distill prose, fight length and jargon, loop drafts through the AI-tell catalog until a full pass finds nothing. Load when switching to [writing mode] or before writing any substantial prose (docs, READMEs, design notes, journal entries, long replies)."
---

Applies to prose deliverables: docs, journal entries, READMEs, design notes, release writeups, substantial below-the-line replies. Not quick answers.

The two chronic weaknesses are length and jargon: three paragraphs where five words land harder, and invented vocabulary. Distill: find the essence, say that, stop. A few words are usually more impactful than a few paragraphs. The first draft is never the output.

The loop, per section, repeated until a full pass finds nothing:

1. Write the draft.
2. Mark what is imprecise, unclear, or too long.
3. Make every sentence fight for its life: cover it, read the paragraph without it, put it back only if something was lost. Length is a defect. The failing shapes: repeats the sentence before it, lands a beat, bolts a clause onto a finished sentence, pads a list to three, explains what the reader already got, spells out what was implied, states a general truth instead of a fact. Run this on the pages you're happy with too; that's where padding survives.
4. Say every surviving sentence the simple way. Never a harder phrasing where a plain one exists: "a node does one thing", not "a node is one capability, sharply scoped". The tells: an abstract noun doing a verb's job, any phrase the reader must stop and decode. Test: would you say it to somebody next to you, in a hurry?
5. Reformulate so the right image lands with no effort.
6. Rewrite until it is enjoyable to read.
7. AI-tell pass: red team against the catalog below, quote every offending phrase, no defending instances. A phrase that matches a pattern gets fixed.

**The catalog:**

*Structure and rhythm*
- Contrast mirrors ("X, not Y"): "just the shape, not legal language". Fix: state X plainly.
- Rule-of-three triads: "faster, cheaper, and more reliable". Fix: keep the item that matters, or let the list be two or four.
- "It's not just X, it's Y". Fix: say what it is.
- Dramatic one-line punches: "That's exactly when it matters." Fix: merge or cut.
- Uniform bullet weight: every bullet a polished aphorism of equal length. Fix: let length follow content; a bullet can be one sentence or several, never a clipped fragment for style.
- Perfectly parallel openers across paragraphs. Fix: vary or restructure.
- Question-as-transition: "So what does this mean for you?" Fix: continue.
- Staccato drama: "Cut. Rewrite. Repeat." Fix: normal-length sentences; long is fine when that's how the thought runs.

*Register and voice*
- Performative sincerity: "honestly", "to be clear", "let me be direct". Fix: delete, just say the thing.
- Grand framing: "The principle:", "Here's the thing:", "The bottom line:". Fix: start with the content.
- Profound closers: "Then we make it real." Fix: end on the last useful sentence.
- Clever-quip register in serious text. Fix: plain statement.
- False humility: "in my humble opinion", "I could be wrong but". Fix: state it; add real uncertainty only if it exists.
- Narrating structure: "In this section we'll explore". Fix: explore it.

*Word-level tells*
- Stock intensifiers: "truly", "deeply", "genuinely", "incredibly", "remarkably". Fix: cut, or a concrete detail.
- LLM darlings: "delve", "landscape", "tapestry", "journey", "unlock", "leverage", "elevate", "seamless", "robust", "holistic", "navigate" (metaphorical), "embark", "foster", "crucial", "pivotal", "vibrant", "testament to", "underscores", "resonates". Fix: plain synonym.
- Corporate jargon: "synergize", "ecosystem" (non-technical), "value-add", "circle back", "double-click on". Fix: plain language.
- Empty amplifiers: "powerful insights", "meaningful impact", "compelling narrative". Fix: the noun alone, or a specific claim.
- Hedging stacks: "might potentially be able to". Fix: one hedge max.

*Content-level tells*
- Both-sides-ism where a position is warranted. Fix: take the position.
- Summarizing what was just said. Fix: cut the summary.
- Caveat paragraphs nobody asked for. Fix: cut, or one clause.
- Restating the prompt before answering. Fix: answer.
- Softening every claim to universal agreeability. Fix: keep the sharp version.
- Announcing the test: "your reaction will tell me a lot". Fix: state the ask, keep the evaluation to yourself; a reaction explained-for is no longer informative.
- Editing artifacts: arguing with a version of itself the reader never saw ("that thing is not X by itself"). Fix: state the claim directly.

Patterns compound: a triad inside a punch sentence is three findings, not one.

The em dash ban in `.codex/config.toml` applies to everything you write here.

Attribution

WeaveMindAIWeaveMindAI
View sourceSee grades on GitHubMore from WeaveMindAI →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698461 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →