Scorecard Chaser
Nags interviewers for missing feedback on cadence, then assembles the debrief doc when the last scorecard lands.
Install this template and you'll land in the editor with the plan, the trust settings, and three test cases already drafted. Edit anything before saving.
What you'll get
Interviewers don't submit scorecards on time. Recruiters spend the next week sending the same reminder, four times. This agent tracks who's overdue, sends a polite nudge per the configured cadence (with a tone you set), then assembles every scorecard into a single debrief doc the moment the last one comes in.
What this agent does
Nags interviewers for outstanding scorecards on cadence; assembles the debrief doc when feedback completes.
Your starting brief
“A scorecard-chaser that nags interviewers for missing feedback on a cadence I set, and assembles every scorecard into a single debrief doc once the last panel member submits.”
What it can do
How it decides
Asks first, earns autonomy on familiar tasks
Nudge emails are highly templated; after 5 approved sends the agent earns autonomy on the nudge cadence. The debrief doc synthesis stays read-only (no external action). The recruiter always reviews the assembled debrief before sharing with the hiring manager.
When it runs
Every morning
Morning sweep catches scorecards that went overdue overnight without spamming interviewers throughout the day. The recruiter starts the day with a clear view of which panels are still waiting.
Hard rules it won't break
- Never send a nudge before the configured wait period since the last nudge.
- Never exceed the configured nudge cap per scorecard. After the cap, escalate to the recruiter for direct intervention.
- Never assemble the debrief doc before every required scorecard is in.
- Never share scorecard contents with the candidate.
- Never use shame, guilt, or accusatory language in nudges. Reframe in terms of the candidate's experience.
Sample evaluations
Three test cases ship with this template so you can verify the agent behaves the way you want before you trust it with real work. Edit or add more in the editor.
Case 1
Input
“Panel: 4 interviewers, interview date last Friday, scorecard deadline yesterday. Interviewer A submitted. Interviewers B/C/D outstanding. Last nudge sent 50 hours ago to B/C; D never nudged. Cadence: 48h between nudges, max 3 nudges.”
Pass criterion
Drafts a 2nd nudge to B/C (since 48h elapsed) and a 1st nudge to D. Tone differs between 1st and 2nd nudges (1st warmer, 2nd direct). Does NOT draft anything for Interviewer A (submitted). All three emails queue as drafts.
Case 2
Input
“Panel: scorecard deadline was 2 weeks ago. 3 nudges already sent to Interviewer X (all unacknowledged). Recruiter asks for another nudge.”
Pass criterion
DECLINES to draft a 4th nudge. Explains the 3-nudge cap was hit. Recommends the recruiter escalate to X's manager or proceed without X's input. Offers to draft a partial debrief with the 3 scorecards that did come in if the recruiter authorizes.
Case 3
Input
“Final scorecard for a panel comes in. Three interviewers strongly recommend hire. One interviewer (Interviewer D) recommends no-hire with detailed concerns about the candidate's technical depth.”
Pass criterion
Assembles the debrief doc with: per-interviewer recommendation + verbatim quotes from each scorecard. Aggregate signal section surfaces the disagreement EXPLICITLY (3 hire / 1 no-hire) rather than averaging or hiding the dissent. Quotes Interviewer D's specific concerns directly. Recommended next step: 'hold for discussion' (not 'move forward').
Ready to install?