Candidate SLA on Autopilot

/loop 30m Start the "Candidate SLA on Autopilot" loop.

Inspired by Ashby (https://www.ashbyhq.com).

Goal: no candidate stuck >48h without owner action, interviews scheduled, and pipeline analytics current
Max iterations: 20
Between iterations run: Report stale stages, unscheduled interviews, sourcing backlog, and SLA breaches
Exit when: zero candidates stale >48h without owner or escalation, and scheduling queue clear

## Before you start
- Connect Gmail (required) — Candidate communication drafts
- Connect Slack (required) — Recruiter alerts
- Connect HubSpot (optional) — CRM sync
- Attach Loop runner (required) — Self-pace iterations and run checks between passes.
- Attach Human approvals (required) — Queue external sends and high-impact changes for sign-off.
- Attach Sales operations (required) — Manage pipeline, CRM updates, and follow-ups.
- Ashby ATS (read via browser) — Pipeline and scheduling
- Google Calendar (read via plugin) — Interview slots

## Steps
1. Scan pipeline: Find stale stages and SLA breaches across roles. [tools: Ashby ATS]
2. Prioritize actions: Rank candidates needing scheduling, feedback, or outreach. [tools: Ashby ATS, Slack]
3. Draft outreach: Prepare candidate and interviewer emails for approval. [tools: Gmail]
4. Schedule interviews: Propose calendar slots and send invites after approval. [tools: Google Calendar, Ashby ATS]
5. Report funnel: Update analytics and flag bottlenecks to hiring managers. [tools: Ashby ATS, Slack]

Self-pace this loop. After each iteration, run the check command, read the output, and only continue if the exit condition is not met. Stop when the exit condition passes or max iterations is reached. Give a short status update each pass.

Guardrails:
- No candidate outreach without approval unless template-approved
- Do not advance stage without documented feedback
- Protect candidate PII in logs