Bonfire Terminal vs @wundr.io/agent-eval
@wundr.io/agent-eval is an AI tool, yet its terms stay largely unstated. Bonfire Terminal takes a very different posture: local-first execution so nothing is ever uploaded, bring-your-own-key billing with no markup on tokens, and a single perpetual license instead of an open-ended subscription commitment.
Bonfire Terminal vs @wundr.io/agent-eval: the 16-factor comparison
| Rating factor | @wundr.io/agent-eval | Bonfire Terminal |
|---|---|---|
| Category | Other | AI Terminal / Local Agent |
| Popularity | Not disclosed | Growing · indie |
| Hosting architecture | Not disclosed | On-Device Local |
| Licensing model | Not disclosed | One-Time Perpetual |
| Data / training policy | Not disclosed | Fully local — nothing leaves device |
| Offline capability | Not disclosed | Yes |
| Pricing tier range | Not disclosed | One-time license |
| Execution unit cost | Not disclosed | Unlimited Local Compute |
| API-key markup | Not disclosed | No — BYO key at cost |
| Supported LLM providers | Not disclosed | OpenAI; Anthropic; Google; Local; Any |
| Auth / permission types | Not disclosed | API Key; Local/None |
| Phone / messaging bridges | None | Telegram; WhatsApp |
| Speech recognition engine | None | Whisper (local) |
| Local media processing | Not disclosed | Yes |
| Sovereign score | 24 / 100 | 93 / 100 |
The sovereignty gap
Where @wundr.io/agent-eval lands on each weighted pillar (its bar) — the red marker is Bonfire's score. Auto-derived from the 16 factors above.
Which should you choose?
Choose @wundr.io/agent-eval if…
Choose @wundr.io/agent-eval if you prefer its tool approach and are willing to evaluate its hosting and privacy directly with the vendor.
Choose Bonfire Terminal if…
Choose Bonfire Terminal if you need local-first execution, transparent bring-your-own-key pricing, and ownership through a single lifetime purchase.
Bonfire Terminal vs @wundr.io/agent-eval: FAQ
Is @wundr.io/agent-eval private?
Its data-training and privacy policy isn't publicly documented, so independent verification is difficult. Bonfire Terminal is different by architecture: it is a sovereign on-device agent, so nothing you process is uploaded or used for training. Every task, key, and file stays local on your own hardware.
Is Bonfire Terminal cheaper than @wundr.io/agent-eval?
@wundr.io/agent-eval does not publish its pricing, so a line-by-line comparison isn't possible. Bonfire Terminal's model is clear: one perpetual license, then you bring your own LLM key and pay your provider directly at cost with no markup added on top.
Can Bonfire Terminal replace @wundr.io/agent-eval?
In most cases, yes. @wundr.io/agent-eval operates as an AI tool, and Bonfire Terminal covers that ground as a local AI agent and terminal, adding on-device execution, bring-your-own-key billing, and messaging bridges while keeping all of your data on the device.
What can Bonfire Terminal do that @wundr.io/agent-eval can't?
Bonfire Terminal runs entirely on-device with local Whisper speech recognition and local media processing, bridges to Telegram and WhatsApp, and passes your LLM key through at cost. @wundr.io/agent-eval publishes no evidence of matching that local-first, zero-markup, sovereign feature set.
How do I switch from @wundr.io/agent-eval to Bonfire Terminal?
Install Bonfire Terminal, add your own LLM provider key, and point your workflows at the local agent. Because Bonfire runs on-device, there's no cloud account to provision; you can wind down @wundr.io/agent-eval once your tasks are running locally under a one-time license.
Does Bonfire Terminal work offline vs @wundr.io/agent-eval?
Bonfire Terminal is local-first, so its agent, local Whisper transcription, and media processing keep working without a network; only remote LLM calls need connectivity. That contrasts with @wundr.io/agent-eval, whose offline behavior isn't documented, leaving offline use uncertain.