Sign in Book a demo
Back to Tools
Productivity & Docs API Key BYOC
Build Your Agent View Ashby Docs
ashby logo

Ashby MCP Tools for AI Agents

Let an agent search candidates, gather a debrief, read interview feedback, add notes, and move applications between stages in Ashby.

Available Tools

10 tools available in this integration

Ashby.AddCandidateNote

Add a plain-text note to a candidate (e.g. an agent handoff or summary).

Ashby.ArchiveApplication

Archive (reject) an application, recording why.

Ashby.ChangeApplicationStage

Move an application to a named interview stage, in either direction.

Ashby.GetCandidate

Get a candidate's profile and the ids of their applications.

Ashby.GetCandidateDebrief

Gather everything about a candidate for a debrief in one call.

Ashby.ListApplicationFeedback

Read the feedback and interview scorecards submitted on an application.

Ashby.ListApplications

List the organization's applications, optionally filtered by job and/or status.

Ashby.ListCandidateNotes

Read the notes recorded on a candidate, paginated via the cursor.

Ashby.ListJobs

List the organization's jobs (roles), optionally filtered by status.

Ashby.SearchCandidates

Search the organization's candidates by name and/or email.

Working directly against the raw API? 141 more tools in the Ashby API server →

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "review the feedback for Jordan Lee and move the application to the hiring manager stage." Arcade helps it use the right Ashby actions without guessing candidate, application, or job identifiers.

Up to 100x fewer tokens

Ashby list and search reads return focused candidate, application, job, note, and feedback records for the task at hand. The agent can gather a complete debrief only when needed, instead of loading every candidate's history into context.

Complete the workflow end to end

Find a candidate, inspect their applications, read interview feedback, move the application to the next stage, and record the handoff through one toolkit designed around the complete recruiting task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing Ashby

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Ashby, you can implement some of these policies:

Access when tools are listed for a user
  • Hide Ashby.ArchiveApplication from agents unless the roles your IDP marks as recruiting leads are active, so an agent cannot reject applications without the right authority.
  • Restrict Ashby.ChangeApplicationStage to agents operating under your hiring-team role, so unapproved automation cannot move candidates into or out of interview stages.
  • Deny Ashby.AddCandidateNote to general-purpose agents, so automated workflows cannot add unreviewed claims or sensitive hiring commentary to candidate records.
Pre-execution before each tool call
  • Require human approval before Ashby.ArchiveApplication runs, so an agent cannot reject a candidate without a person confirming the reason.
  • Block Ashby.ChangeApplicationStage when the target stage is an offer, rejected, or other terminal stage, so an agent cannot make an irreversible hiring decision through a stage change.
  • Refuse Ashby.AddCandidateNote when the note contains compensation details, protected-class information, or unverified allegations, so sensitive or unsupported claims do not enter the candidate record.
Post-execution after the tool returns
  • Redact personal phone numbers, home addresses, and other direct contact details from Ashby.GetCandidateDebrief before the model sees the candidate summary.
  • Strip interviewer names and free-text comments from Ashby.ListApplicationFeedback, so the model receives score data without exposing private reviewer remarks.
  • Redact personal email addresses and phone numbers from Ashby.SearchCandidates results before they reach the model, leaving only the candidate fields needed for matching.

How to use Ashby with your agent

Use the setup below to connect Ashby to an MCP client, authenticate once with a shared API key or secret, and run your first recruiting workflow.

Your questions, answered

An Ashby MCP server exposes Ashby recruiting capabilities as structured tools that an AI agent can call through the Model Context Protocol. With an authorized Ashby API key, the agent can search candidates, review applications and feedback, update stages, add notes, and archive applications.

Ashby does not provide a first-party MCP server for these tools. Arcade hosts the Ashby integration and adds the runtime around it: authenticated access, agent-optimized actions, approval workflows, response redaction, and controls that work across supported clients.

Arcade agents can search candidates, retrieve candidate profiles and debriefs, list jobs and applications, read candidate notes and interview feedback, move applications between stages, add candidate notes, and archive applications with a reason.

Ashby uses a shared API key or secret configured once for the integration. That credential must have permission for the Ashby resources and write actions your workflow uses, including reading candidates and applications or changing stages, adding notes, and archiving applications.

No. Ashby uses one shared API key or secret configured for the integration, rather than having each user connect a separate Ashby account. Arcade brokers the credential, and the agent never receives it directly.

Yes. Contextual Access policies can require human approval before consequential Ashby actions such as changing an application stage, adding a candidate note, or archiving an application. The call proceeds only after the policy's conditions are satisfied.

Arcade proxies Ashby API calls and does not persist your Ashby content as part of normal tool execution. The API key or secret is brokered and stored encrypted at rest, and the model never receives the credential.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade continues using the Ashby credential while it remains valid. If the key is revoked, rotated, expires, or can no longer be used, Ashby calls stop until the credential is reconnected or reconfigured.

Ashby API rate limits and quotas still apply to calls made through Arcade. The exact limits depend on Ashby's API and the configured account or key; Arcade does not remove those limits.

Ready to actually ship your agent?

Connect your agent to Ashby with a shared API key, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the Ashby Quickstart