# Determining the correct "target" value when retrieving notecard firmware

**URL:** <https://discuss.blues.com/t/determining-the-correct-target-value-when-retrieving-notecard-firmware/3391>\
**Category:** Firmware\
**Created:** [November 24, 2025, 2:48am UTC](https://discuss.blues.com/t/determining-the-correct-target-value-when-retrieving-notecard-firmware/3391 "2025-11-24T02:48:05Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![devElert](https://avatars.discourse-cdn.com/v4/letter/d/8e8cbc/32.png) [@devElert](https://discuss.blues.com/u/devElert)\
**Post date:** [November 24, 2025, 2:48am UTC](https://discuss.blues.com/t/determining-the-correct-target-value-when-retrieving-notecard-firmware/3391/1 "2025-11-24T02:48:05Z")

</div>

The `/firmware` API supports a `target` parameter, which filters the returned Notecard firmwares by the target type. Is there a way host firmware can determine the correct value of target from the Notecard?

---

<div class="post-metadata">

**Author:** ![devElert](https://avatars.discourse-cdn.com/v4/letter/d/8e8cbc/32.png) [@devElert](https://discuss.blues.com/u/devElert)\
**Post date:** [November 24, 2025, 3:04am UTC](https://discuss.blues.com/t/determining-the-correct-target-value-when-retrieving-notecard-firmware/3391/3 "2025-11-24T03:04:38Z")

</div>

Opps, relying too muchpon agentic coding. Sometimes you just gotta look yourself:

`card.version` returns `body.target` as the correct target needed.

---

<div class="post-metadata">

**Author:** ![abucknall](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.blues.com/abucknall/32/1335_2.png) [@abucknall](https://discuss.blues.com/u/abucknall)\
**Post date:** [November 24, 2025, 2:25pm UTC](https://discuss.blues.com/t/determining-the-correct-target-value-when-retrieving-notecard-firmware/3391/4 "2025-11-24T14:25:33Z")

</div>

Hi @devElert ,

We actually have a tool to help you with this! If you’re seeing that your agentic development is causing lots of hallucinations about Notecard API usage, then we recommend using the [Blues Expert MCP server](https://dev.blues.io/tools-and-sdks/generative-ai-tools/mcp-servers/#blues-expert). This plugs into your LLM and gives it access to up-to-date documentation as well as the ability to validate the API usage against our API schema. Check out the link to see how to connect it to your workflow.

Thanks,

Alex