PonoLens is for when that stops being enough
| Need | Free hook script | PonoLens |
|---|---|---|
| Tools & maintenance | You connect each tool and maintain its script. | Supported adapters maintained in one app. |
| Broader risks | Only the checks you write or add. | Configured checks for keys, contact, health, and payment formats. |
| What gets saved | You choose what to log and redact. | Sanitized audit records; saved prompts redacted by default. |
| A client asks what happened | You interpret the output and prepare an answer. | Redacted, plain-language records and exportable reports. |
A custom script can add these features too. The difference is who builds and maintains the workflow.
You use more than one tool.
Claude Code, Cursor, Codex, and others expose activity through different hooks, plugins, and session formats. PonoLens brings their reported activity into one place.
Your risk isn’t just keys.
Client names, emails, health and payment details slip into prompts too. A check for api_key alone won’t catch those. PonoLens checks configured sensitive-information formats beyond API keys. You can add your own custom flags for exact private values or patterns.
Your log shouldn’t become the leak.
A basic logger can save the private information it catches. PonoLens sanitizes commands, paths, and tool inputs before saving audit records, and redacts saved prompts by default.
A client needs an answer.
When a client asks whether their data went to an AI company, a grep result isn’t an answer. PonoLens gives you redacted, plain-language records and reports you can share with that client, based on the activity your integrations actually reported.
Someone has to maintain it.
Hook formats change with tool updates. We keep the PonoLens adapters current. With a DIY script, maintaining and retesting the connection is your job.
Local records, optional cloud archive
Detection runs locally. Saved prompts are redacted by default. Optional Solo cloud archiving uploads new redacted records for 365 days; encrypted originals stay on the Mac that created them. See what’s stored, where, and what’s uploaded →
Know what was actually stopped
PonoLens is honest about what it knows.
It recorded an action during or after it happened.
The integration indicates data left the device.
A supported adapter actually enforced a block.
You can tell what was seen, what was sent, and what was truly stopped.
Report Only is the stable default. Redact and Block are experimental and require supported before-action hooks. Codex side-chat is observation-only.
A custom script can grow into these features too. The choice is whether you want to build and maintain that workflow yourself. Neither approach can see activity a tool never exposes, and neither establishes regulatory compliance.
Explore the demo → · See current plans · Read the security model