Skip to content
Back to skills

Speq Writing Guardrails

ASecurity

Nine prose rules covering conclusion-first order, one idea per sentence, named actors, consistent names, plain explanations, no filler, neutral tone, current results only, and no em dashes. Triggered by /speq-mission, /speq-audit, /speq-implement, /speq-plan-pr, /speq-implement-pr, planner-agent, plan-reviewer, recorder-agent, architecture-agent, and code-reviewer.

  • 56 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 9, 2026
ai-agentsgoexpressrailscode-reviewgit

Security analysis

A100/100

Scanned October 9, 2026

npx -y skills add marconae/speq-skill --skill speq-writing-guardrails --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Speq Writing Guardrails?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Speq Writing Guardrails
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/marconae-speq-writing-guardrails/badge)](https://www.skillsdirectory.com/skills/marconae-speq-writing-guardrails)

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

Download with Pro
SKILL.md
---
name: speq-writing-guardrails
description: Nine prose rules covering conclusion-first order, one idea per sentence, named actors, consistent names, plain explanations, no filler, neutral tone, current results only, and no em dashes. Triggered by /speq-mission, /speq-audit, /speq-implement, /speq-plan-pr, /speq-implement-pr, planner-agent, plan-reviewer, recorder-agent, architecture-agent, and code-reviewer.
---

# Writing Guardrails

Follow these rules for governed prose.

## Scope

- Govern: prose in `plan.md`, `spec.md`, `mission.md`, `architecture.md` (project and plan level), decision logs, ADR fragments in `specs/_decision/`, `review-findings.md`, verification reports, implementation summaries, GitHub PRs, issues, and comments.
- Leave unchanged: Gherkin, Background bullets, tables, ASCII diagrams, delta markers, and validator-owned RFC keyword casing.

## Rules

1. Lead with the conclusion. State the decision, result, or finding first. Give reasons afterward.
2. Express one idea per sentence. Split sentences that make separate claims.
3. Name the actor. State who or what performs each action.
4. Name each thing once and reuse that name. Do not switch to a synonym for the same thing later in the text.
5. Assume the reader does not know the method, concept, technology, or system. Explain unfamiliar concepts in simple language before using specialized terms.
6. Remove filler, hedges, superlatives, intensifiers, hyperbole, and promotional language. Use measurable evidence when making claims about scale or quality.
7. Use neutral, matter-of-fact language. Avoid colloquial, metaphorical, dramatic, and rhetorical language.
8. State the current result. Do not narrate attempts, revisions, or reasoning history unless that history is relevant.
9. Do not use em dashes. Use a period, comma, or colon.

Attribution

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

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.