Candidate Availability Call

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Candidate Availability Call

Inspiration Recruiting teams lose time in back-and-forth scheduling, especially when a candidate is easier to reach by phone than by email. Candidate Availability Call focuses on one narrow workflow: safely turning a disclosed phone call into reviewable interview availability windows. ## What it does Candidate Availability Call lets a recruiter or hiring coordinator authorize one disclosed CALL-E phone call to a candidate for interview availability. The workflow validates the request, previews the call goal in dry-run mode, masks the destination phone number, asks only scheduling questions, and returns a structured result with availability windows, timezone, constraints, preferred follow-up channel, consent, and transcript evidence. It intentionally does not screen candidates, discuss compensation, confirm interviews, update calendars, or write to an ATS. A human coordinator reviews the result before any scheduling commitment. ## How we built it The project was built as a CALL-E Agent Skill contribution with a complete skill package, safety references, safe and unsafe examples, a fictional request fixture, no-network validation, and a no-network dry-run preview generator. The public pull request is https://github.com/CALLE-AI/awesome-phone-call-agents/pull/167 ## Safety and responsibility The skill discloses that the caller is AI, limits the purpose to scheduling, validates E.164 phone format and IANA timezone names, uses a fictional reserved sample number, redacts the destination in dry-run output, and refuses screening, compensation, protected-topic, immigration, and background-check questions. It also keeps consent explicit and requires human review before calendar action. ## Challenges The main challenge was drawing a bright line between helpful scheduling automation and hiring decisions that should not be delegated to a phone agent. The implementation keeps the agent narrow, evidence-backed, and dry-run first so reviewers can inspect the intended call before any live execution. ## Accomplishments The project includes a reusable CALL-E skill, sample input, validation script, preview script, generated example preview, repository validation, a public demo video, and an upstream pull request. ## What comes next Next steps would be connecting authenticated CALL-E execution after human authorization, adding coordinator review UI, and integrating the approved structured result into calendar or ATS workflows without automatic booking.

Built With

• agent-skill
• call-e
• github-actions
• javascript
• node.js
• recruiting
• scheduling
• voice-ai