# Survey After

> Design participant-feedback collection for one edition of a technical event, then write the organizers' design memo
> or the survey participants answer. Enter the format and size, whether the event already happened, the roles you can
> reach, the decisions waiting on the answers, the draft questions, the anonymity promise and the send plan. Built on the
> SkillSafe skill @samber/event-feedback (samber/dev-event-organizer-skills, MIT).

https://survey-after.skillsafe.ai/

## What is free
- Every draft question tied to a named decision and owner, or flagged for cutting; hypothetical, leading and unbounded wording; the closed-to-open balance; talent-pipeline questions at a community event; the level question for a mixed-level meetup.
- The depth the draft makes and the page's pick (single-question pulse, short core, core plus one role module, full instrument with the hackathon layer), with modules for absent roles deleted and the length penalty.
- The anonymity the questions need (judging, mentors, inclusion, a named person) and the free-text rule.
- The channel menu against measured response rates - emailed 10-30%, in-room 40-60%, an incentive about half again - with in-room capture deleted once the event has happened and the prize draw deleted under anonymity.
- The 24-48 hour send, at most two reminders, the hackathon follow-up one to two weeks later, and decisions due before the answers can arrive.
- Response rate per role against the published floor (about one in ten emailed, four in ten in-room), small roles read by name.
- What goes back to respondents: a you-said-we-changed note by default, a public findings post only with consent and a planned recap.
- Questions CSV (spreadsheet-safe), the send plan as a calendar file (.ics: each send, reminder, follow-up and decision due date), Markdown and JSON exports.

## The metered lanes
Two tasks on gpt-terra, metered per run, routed by `task`. Input string fields: `task` (design or survey), `event`, `instrument` and `facts`.
- `design` replies with `summary`, `depth`, `channel`, `questions`, `anonymity`, `timing`, `loop`, `rates`, `signals` and `caveats`; the page holds every pick, keep-or-cut, promise, date and rate to its own.
- `survey` replies with `title`, `intro`, `questions`, `closing` and `note`; the page requires exactly the kept questions in order, no hypothetical or leading wording, the anonymity promise and question count in the intro, and no public post promised where the page deletes it.

## Links
- API tutorial: https://survey-after.skillsafe.ai/api.html
- Source skill: https://skillsafe.ai/skill/@samber/event-feedback