> ## Documentation Index
> Fetch the complete documentation index at: https://docs.exterview.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Exterview Agents
> The 26 specialized agents that make up the talent workforce inside Exterview, grouped by hiring stage, with what each one is accountable for and its current status.
## Purpose
Exterview is the first application on [Smaya](/smaya/overview), and it runs a workforce of **26 specialized agents**. Each is accountable for one job. [AI Employees](/smaya/ai-employees) put them to work today. Upcoming: in a role run they work together as an [agent swarm](/smaya/agent-swarm). You see the reports, not the agents.
**Availability.** **Available** means you can use it in Exterview today.
**Upcoming** means it is on the way. The workforce is 26 roles either way — a
buyer asking "what does your AI do?" deserves the whole map, not only the
finished half.
## Design — 2 agents
Turning a hiring need into something that can be evaluated consistently.
| Agent | Accountable for | Status |
| -------------------------- | ----------------------------------------------------------------------------- | --------- |
| **Role Design Agent** | Turning a brief or job description into a structured role with clear criteria | Available |
| **Interview Design Agent** | Building the interview structure and questions that test those criteria | Available |
## Sourcing & Intake — 3 agents
Getting the right people into the process, and their information into a usable shape.
| Agent | Accountable for | Status |
| ---------------------------- | --------------------------------------------------------------------- | --------- |
| **Sourcing Agent** | Finding and surfacing candidates against an open role | Upcoming |
| **Candidate Outreach Agent** | First contact and follow-up with sourced candidates | Upcoming |
| **Application Intake Agent** | Reading resumes and applications into structured, comparable profiles | Available |
## Screening & Interviewing — 6 agents
The core evaluation work, across every channel a process uses.
| Agent | Accountable for | Status |
| ------------------------------ | ------------------------------------------------------------------------------------- | --------- |
| **Resume Screening Agent** | Matching a profile against a role and explaining the gaps | Available |
| **Phone Screening Agent** | Running a short structured screening call and reporting on it | Available |
| **Voice Interview Agent** | Running a full structured voice interview and producing the report | Available |
| **Avatar Interview Agent** | Running a face-to-face avatar interview and producing the report | Available |
| **Skills Assessment Agent** | Setting and evaluating practical skills and coding work | Upcoming |
| **Panel & HR Interview Agent** | Capturing panel and HR rounds in Microsoft Teams, and producing the competency report | Available |
## Trust & Integrity — 2 agents
The agents that check the other agents, and check the process itself.
| Agent | Accountable for | Status |
| ------------------------------------------------- | ------------------------------------------------------------------------------ | -------- |
| **Interview Integrity Agent** | Detecting impersonation, assistance and other integrity risks during a session | Upcoming |
| **Quality & Fairness Agent** | Independently reviewing every other agent's output before it enters the record | Upcoming |
**Not customer-configurable, by design.** It grades every
other agent — an organization able to soften it could pass a flawed outcome
through a clean scorecard. It stays fully visible and readable to you; only its
settings are fixed. Today, automated checks run on every scored output and show
in Scoring Logs. See [Governance & Control](/smaya/oversight).
## Operations — 7 agents
The coordination work that consumes most of a recruiting team's day.
| Agent | Accountable for | Status |
| --------------------------------- | --------------------------------------------------------------------------------------------------------- | --------- |
| **Scheduling Agent** | Finding times, booking rounds and handling reschedules | Available |
| **Task Reminder Agent** | Chasing outstanding actions with the people who owe them | Upcoming |
| **Candidate Communication Agent** | Keeping candidates informed at every stage, on email and WhatsApp | Available |
| **Journey Nudge Agent** | Bringing back candidates who drop out mid-interview. Upcoming: recovery across voice and avatar channels | Available |
| **Shortlisting Agent** | Ranking candidates from completed evaluations on the Leaderboard | Available |
| **Escalation Agent** | Routing anything outside a role's mandate to its named human owner | Upcoming |
| **HR System Sync Agent** | Sending reports and status to your applicant system. Upcoming: two-way sync with applicant and HR systems | Available |
## Knowledge & Insight — 4 agents
Answering questions, over your own data.
| Agent | Accountable for | Status |
| ----------------------- | ------------------------------------------------------------------- | --------- |
| **Talent Search Agent** | Finding people across your talent pool by capability, not keyword | Upcoming |
| **Knowledge Agent** | Bringing your company knowledge into agents' questions and answers | Available |
| **Report Q\&A Agent** | Answering questions about a specific evaluation, with citations | Available |
| **Analytics Agent** | Turning a question about your hiring data into an answer and a view | Upcoming |
## Candidate-facing — 1 agent
| Agent | Accountable for | Status |
| -------------------------------------------------- | -------------------------------------------------------------------------- | -------- |
| **Candidate Agent** | Helping candidates through their own process and answering their questions | Upcoming |
**A deliberately different permission model.** It
serves people who are not signed in to your organization, so it carries a fixed
minimal scope, no connections to your systems, and no access to your candidate
pool. That restriction is structural, not a setting.
## Post-decision — 1 agent
| Agent | Accountable for | Status |
| --------------------------- | ------------------------------------------------------------------ | -------- |
| **Offer Preparation Agent** | Preparing offer terms and routing them through your approval chain | Upcoming |
**Extending an offer is always a human act.** No agent dispatches an offer, and no agent records a final hiring outcome.
## What no agent may ever do
Hired, rejected and withdrawn are set by a named person. An agent prepares the evidence and the recommendation.
An agent may never suppress or de-prioritize a candidate who scored below the
bar. The record is complete or it is worthless.
Effective authority is the intersection of the agent's scope and the user's
permissions.
Every action, proposal and approval is captured and attributed.
## Related
*
[Gradientflo Agents](/agents/gradientflo). The ten HR agents, all Upcoming.
*
[Roadmap to 100](/agents/roadmap). How the library grows.
*
[AI Employees](/smaya/ai-employees). How these agents become a deployed role.