CtrlK
BlogDocsLog inGet started
Tessl Logo

writing-clearly-and-concisely

Apply Strunk's timeless writing rules to ANY prose humans will read—documentation, commit messages, error messages, explanations, reports, or UI text. Makes your writing clearer, stronger, and more professional.

Install with Tessl CLI

npx tessl i github:Dicklesworthstone/pi_agent_rust --skill writing-clearly-and-concisely
What are skills?

66

Does it follow best practices?

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

98%

Rewrite the Onboarding Guide

Active voice and conciseness

Criteria
Without context
With context

Active voice: IT access

100%

100%

Active voice: 2FA requirement

100%

100%

Active voice: setup script

100%

100%

Positive form: no-admin warning

75%

75%

Positive form: help channel

100%

100%

Omit needless words: access form

100%

100%

Omit needless words: blocker sentence

100%

100%

Positive form: urgent issues

100%

100%

Active voice throughout

100%

100%

Assertive not hedging

100%

100%

No expletive openings

100%

100%

Content preserved

100%

100%

Without context: $0.0896 · 26s · 8 turns · 13 in / 1,213 out tokens

With context: $0.4351 · 1m 37s · 24 turns · 62 in / 4,760 out tokens

92%

-1%

Write a Project README for an Open Source Library

Paragraph structure and parallelism

Criteria
Without context
With context

One topic per section

100%

100%

Topic sentence: problem

100%

100%

Topic sentence: alternatives

37%

0%

Parallel comparison structure

100%

100%

Parallel list items

75%

100%

Emphatic sentence endings

100%

100%

Concrete language: problem

100%

100%

Concrete language: not suitable

100%

100%

Serial comma usage

100%

100%

No comma splices

100%

100%

Active voice in README prose

100%

100%

Concise prose

100%

100%

Without context: $0.1641 · 46s · 11 turns · 18 in / 2,039 out tokens

With context: $0.3584 · 1m 18s · 22 turns · 60 in / 3,823 out tokens

92%

5%

Improve Error Messages for a CLI Tool

Specific language and word placement

Criteria
Without context
With context

Specific cause: Error 1

100%

62%

Omit needless words: Error 2

0%

100%

Active voice: Error 3

100%

100%

Specific language: Error 4

75%

75%

Positive form: Error 5

100%

100%

Concrete state: Error 6

100%

87%

No passive in rewrites

100%

100%

Emphatic endings

100%

100%

Related words together

100%

100%

No 'it is possible/recommended'

100%

100%

Positive form throughout

66%

77%

Action specified

100%

100%

Without context: $0.1155 · 36s · 10 turns · 56 in / 1,833 out tokens

With context: $0.3643 · 1m 13s · 20 turns · 126 in / 4,279 out tokens

Evaluated
Agent
Claude Code

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.