How is passive voice detected?

The detector uses a heuristic — a form of "to be" (is, are, was, were, been, being) followed by a past participle, optionally with a "by" agent. It is fast and local but not perfect; it can miss or over-flag, so treat results as suggestions.

Do I need an API key just to detect passive voice?

No. Detection and highlighting run entirely in your browser with no key. The key is only required for the optional rewrite step that converts flagged sentences to active voice.

Where does my API key go?

It stays in your browser tab and is sent directly to OpenAI or Anthropic with the rewrite request you trigger. It is never stored, logged, or routed through any Gera server, and refreshing the tab clears it.

Is passive voice always bad?

No. Passive voice is appropriate when the actor is unknown, irrelevant, or deliberately de-emphasized, and in scientific or formal writing. Use the detector to find candidates and apply judgement rather than rewriting everything.

Who pays for the rewrite calls?

You do, on your own provider account. Each rewrite is one real API call billed at your usage rate. Detection is free and unlimited; only the rewrite consumes tokens on your key.

What is the Passive Voice Detector & Rewriter (BYO-key)?

Highlights likely passive-voice constructions in LLM output using a local heuristic, with a per-sentence count. With your own OpenAI or Anthropic API key, it rewrites the flagged sentences in active voice while preserving meaning. Your key stays in your browser. It runs free in your browser on Gera Tools, with nothing uploaded.

Passive Voice Detector & Rewriter (BYO-key)

Name: Passive Voice Detector & Rewriter (BYO-key)
Creator: Gera Tools
License: https://creativecommons.org/licenses/by/4.0/

Get one useful tool a week

Like this tool? Enter your email and we'll send you one genuinely useful Gera tool a week — plus a link to come back to this one. No spam, one-click unsubscribe any time.

Passive voice creeps into LLM output constantly — “the report was generated by the system” instead of “the system generated the report.” It is wordier and vaguer about who did what. This tool flags likely passive-voice sentences with a fast local heuristic, and — with your own OpenAI or Anthropic key — rewrites the flagged ones in active voice while keeping the meaning intact.

Why passive voice is common in LLM output

Language models are trained on large bodies of text that include formal writing, academic papers, technical documentation, and corporate communications — all genres that favour the passive voice. As a result, LLMs default toward passive constructions even for tasks where direct, active prose would serve better: marketing copy, product descriptions, customer emails, instructional guides. The passive also lets a model avoid naming a concrete subject when it is uncertain who did what, which is a subtle form of hedging.

How it works

Detection is local and free: the tool scans each sentence for a form of to be (is, are, was, were, been, being) followed by a past participle, optionally with a by agent, and highlights matches with a count. No key is needed for this step. If you want rewrites, add your API key and run the rewrite — the tool sends one direct request from your browser asking the model to convert only the flagged sentences to active voice and return the full revised text.

Your key never reaches a Gera server — it is held only in the tab and sent straight to the provider (with the official direct-browser-access header for Anthropic). Refreshing clears it.

Examples of the heuristic in action

The detector catches these patterns:

“The document was reviewed by the team” — auxiliary was + past participle reviewed + by agent.
“Errors are logged automatically” — auxiliary are + past participle logged.
“The process had been completed” — auxiliary had been + past participle completed.

It is less reliable on:

Progressive passives: “The system is being updated” is caught; “The system keeps being updated” may not be.
Stative passives that read as adjectives: “The door was open” scores as passive but open is an adjective, not a past participle.

These edge cases are why the output is a list of candidates to review, not a mandate to rewrite every flagged sentence.

Passive voice isn’t always wrong

The detector finds candidates, not crimes. Passive voice is the right choice when the actor is unknown, unimportant, or intentionally downplayed, and it is standard in much scientific and formal writing. Use the flags as prompts for judgement, not a mandate to rewrite everything.

Tips

Detect first, read the highlights, and only rewrite when the active version is genuinely clearer.
Cheaper models (gpt-4o-mini, claude-3-5-haiku) handle straightforward rewrites well and minimise cost.
After rewriting, re-run detection to confirm the passive count actually dropped.
If you are editing marketing or product copy, a passive count above roughly 20% of sentences is a signal the draft reads too formally for a consumer audience.