- Home
- Skills
- Technical Documentation
- active voice reviewer
Works with the AI Agents you already use
active voice reviewer
Eliminate passive voice and weak prose in technical documentation through a rigorous five-category editorial audit.
Free
active voice reviewer
Example session with this skill installed
Please review this documentation snippet for active voice: 'There are several steps that should be followed. The database is updated by the script after a connection is made by the user.'
- Read your context and instructions
- Compiled the active voice reviewer
I have completed the exhaustive active voice audit of your documentation snippet. I identified 4 issues across 4 categories.
⚡ EXHAUSTIVE ACTIVE VOICE AUDIT
Category 1: Formal Passive Voice
- Location: ## Database Updates
- Search String: "The database is updated by the script after a connection is made by the user."
- Fixed (Active): "The script updates the database after the user makes a connection."
Rationale: The actors (script and user) were at the end of their respective clauses. Moving them to the subject position makes the process flow logically.
Category 3: Weak "To-Be" Wind-ups (Deadwood)
- Location: ## Introduction
- Search String: "There are several steps that should be followed."
- Fixed (Active): "Follow several steps."
- Rationale: "There are" is filler text. Removing it creates a direct command.
Category 4: Nominalizations (Smothered Verbs)
- Location: ## Connection Logic
- Search String: "...after a connection is made by the user."
- Fixed (Active): "...after the user connects."
- Rationale: "Made a connection" uses a weak noun. "Connects" is a stronger, more concise verb.
Category 5: Modal Hesitation (Soft Commands)
- Location: ## Instructions
- Search String: "several steps that should be followed."
- Fixed (Active): "Follow these steps."
Rationale: "Should be followed" is a soft suggestion. Using the imperative "Follow" is standard for technical instructions.
📊 ACTIVE VOICE SUMMARY
| Category | Issues Found |
|---|---|
| 1. Formal Passive Voice | 1 |
| 2. Hidden Actors | 0 |
| 3. Weak "To-Be" Wind-ups | 1 |
| 4. Nominalizations | 1 |
| 5. Modal Hesitation | 1 |
| TOTAL IMPROVEMENTS | 4 |
Next steps
- Apply the imperative fixes to the "Instructions" section.
- Review the "Database Updates" paragraph for additional technical actors.
- Scan subsequent sections for "There are" filler phrases.
Connects securely to your tools. The creator never sees your data.
About this skill
Technical documentation often suffers from passive phrasing that obscures responsibility and slows down the reader. This skill audits your prose to eliminate weak "to-be" verbs, hidden actors, and wordy nominalizations. It converts hesitant advice into direct, imperative instructions that developers can follow without ambiguity.
What it does
- Identifies passive voice by locating "to-be" verbs paired with past participles to restore the subject to the start of the sentence.
- Uncovers hidden actors in agentless sentences, ensuring the reader knows exactly who or what performs an action.
- Removes deadwood wind-ups like "There is" or "It is" to reduce word count and get straight to the point.
- Converts nominalizations back into strong verbs, changing "conduct an analysis" into "analyze."
- Eliminates modal hesitation by replacing soft words like "should" or "might" with clear, imperative commands.
How it works
- Input text: Provide the technical prose, README, or documentation file you want to audit.
- Category scan: The agent performs five distinct passes over the text based on the defined error categories.
- Drafting fixes: The agent generates a side-by-side comparison of the original sentence and the active correction.
- Summary report: You receive an exhaustive list of every instance found, organized by document heading for easy navigation.
Frameworks & tools
This skill is framework-agnostic and works with any Markdown, plain text, or technical documentation format. It is optimized for instructional writing, API references, and developer guides.
Why this beats prompting it yourself
Generic AI editing often misses subtle nominalizations or keeps "soft" language to sound polite. This skill uses a rigid five-category protocol to ensure zero omissions and forces a transition to the imperative mood required for high-quality technical docs.
Use cases
- README audits to make setup instructions more authoritative and easier to scan.
- API documentation reviews to clarify which component performs specific background tasks.
- Technical blog editing to remove filler phrases and improve the reading experience.
- Onboarding guide cleanup to ensure new hires receive direct commands rather than suggestions.
Known limitations
This skill ignores code blocks inside triple backticks to prevent breaking syntax. It focuses strictly on prose clarity rather than technical accuracy or code logic.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.
Ignores all content within triple backticks. Focuses on linguistic clarity and active voice rather than verifying technical accuracy or correcting code logic.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 25 days ago
- Free with an Agensi account
Permissions
No special permissions declared or detected