v0.1.0 — early access

AI-moderated user studies.
Native to your agent workflow.

One command. Real users. Prioritized findings.
No forms. No dashboards. Just your terminal.

$curl -fsSL https://candor.sh | bash
view source on github →
terminal
// how it works
1
install
One curl command. Registers itself into Claude Code automatically.
2
create a study
Tell your agent what you want to learn. It does the rest.
3
participants are interviewed
Real users. AI voice moderator. Live on your product.
4
get results
Prioritized findings. Key quotes. Ready to act on.