CtrlK
BlogDocsLog inGet started
Tessl Logo

docs-writer

Always use this skill when the task involves writing, reviewing, or editing files in the `/docs` directory or any `.md` files in the repository.

Install with Tessl CLI

npx tessl i github:google-gemini/gemini-cli --skill docs-writer
What are skills?

67

Does it follow best practices?

Agent success when using this skill

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

97%

36%

Document the Gemini CLI MCP Integration Feature

Voice, tone, and naming conventions

Criteria
Without context
With context

Active voice / present tense

80%

100%

Second-person address

87%

100%

Contractions used

0%

100%

No 'please'

100%

100%

No anthropomorphism

100%

80%

No Latin abbreviations

0%

100%

'Lets you' phrasing

100%

100%

Meaningful example names

100%

100%

Correct project name

50%

100%

BLUF introduction

100%

100%

Overview paragraph before lists

50%

80%

Sentence case headings

42%

100%

Next steps section

0%

100%

Without context: $0.1871 · 59s · 9 turns · 9 in / 2,988 out tokens

With context: $0.2900 · 1m 20s · 12 turns · 13 in / 4,021 out tokens

87%

10%

Document the Gemini CLI Code Assist Feature

Procedure formatting and experimental feature docs

Criteria
Without context
With context

Experimental feature note

0%

100%

No table of contents

100%

100%

Procedures introduced with complete sentence

100%

100%

Imperative verb starts each step

100%

70%

Numbered steps for sequences

62%

100%

Conditions before instructions

100%

100%

Optional steps labelled

100%

80%

'Must' vs 'we recommend' (no 'should')

100%

100%

Bold for UI / code font for commands

100%

100%

Serial comma

100%

100%

Next steps section

0%

0%

Without context: $0.1521 · 33s · 8 turns · 9 in / 1,921 out tokens

With context: $0.2819 · 1m 3s · 14 turns · 14 in / 2,793 out tokens

99%

29%

Document the Gemini CLI Rate Management and Usage Tracking

Quota and limit terminology

Criteria
Without context
With context

'Quota' for admin context

33%

91%

'Limit' for real-time ceiling

75%

100%

Combine limit and quota when blocked

46%

100%

'Usage' for consumption tracking

70%

100%

'Reset' for refresh

100%

100%

No 'quota' in error message examples

20%

100%

Meaningful example text

100%

100%

BLUF introduction

100%

100%

Correct project name

100%

100%

No 'e.g.' or 'i.e.'

100%

100%

Without context: $0.2227 · 1m 8s · 10 turns · 11 in / 3,382 out tokens

With context: $0.3815 · 1m 39s · 17 turns · 1,163 in / 4,871 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.