Early feedback: did Context Prism surface useful notes?

[stefanimp]

Thanks for trying Context Prism. I am looking for small, practical retrieval-quality feedback.

You do not need to share private notes. Synthetic examples, redacted excerpts, or a short description of the workflow are enough.

3-minute feedback

1. What kind of note did you test it on?

   • Example: project note, class note, language note, meeting note, writing draft, research note.

2. Did Context Prism surface notes you would have copied manually into an AI assistant?

   • Yes / partly / no.
   • If possible, mention what kind of notes appeared.

3. What was the biggest issue?

   • Missing important notes.
   • Irrelevant notes ranked too high.
   • Snippets were not useful.
   • Token estimates were unclear.
   • The workflow was hard to understand.
   • Something else.

Optional context

• Approximate vault size: small / medium / large.
• Main language or languages in the vault.
• Whether you use metadata, tags, aliases, or mostly plain Markdown notes.
• Whether you tested the context pack with ChatGPT, Claude, Codex, Cursor, or another assistant.

What helps most

The most useful feedback is not "nice plugin". It is a concrete answer to this question:

Did Context Prism find the notes you would actually want an AI assistant to read first?

If it failed, that is useful too. Please describe the failure with redacted or synthetic examples when possible.