AI with Michal

OAuth and API security in recruiting stacks

The set of authentication protocols, scoped token permissions, and integration security practices that govern how recruiting tools connect to each other and to candidate data - with OAuth being the dominant standard for granting limited access without sharing passwords.

Michal Juhas · Last reviewed May 4, 2026

What is OAuth and API security in recruiting stacks?

OAuth and API security in recruiting stacks refers to the authentication standards and integration hygiene practices that govern how your recruiting tools connect, share candidate data, and execute automated actions. OAuth is the most common authorization protocol used when one tool requests access to another: it issues a scoped token rather than sharing a password, making it possible to revoke access without disrupting the connected account.

API security extends beyond the handshake to cover how tokens are stored, what permissions they carry, how long they stay active, and whether anyone audits what they accessed. In recruiting stacks that connect an ATS, sourcing tools, enrichment vendors, and communication platforms, the API layer is where most data leaks and unauthorized-access incidents begin.

Illustration: OAuth token flow connecting recruiting tools through scoped permissions with a revocation control and an audit log node

In practice

  • When a TA ops manager says "we had a token leak," they mean an API key or OAuth token was exposed - often through a shared Postman collection or a Slack message - allowing someone outside the team to query or write to the ATS using the tool's permissions.
  • A recruiter clicking "Connect to ATS" on a sourcing platform is authorizing an OAuth flow. The scopes they approve determine what that tool can do to candidate records, and most users accept all scopes without reviewing the list.
  • A quarterly revocation audit, removing OAuth connections for vendors the team no longer uses, is the single most high-leverage security maintenance task for a small TA function without dedicated IT support.

Quick read, then how hiring teams use it

This is for recruiters, TA ops, and HR leaders who need to understand integration security well enough to make smart vendor decisions, run basic audits, and answer compliance questions. Skim the first section for the vocabulary. Use the second when you are configuring live integrations.

Plain-language summary

  • What it means for you: Every time you click "Authorize" to connect a recruiting tool to your ATS or email, you are issuing a credential that carries real permissions. Treat it like a key to the office.
  • How you would use it: Review the permission scopes before authorizing any new integration. Document active tokens somewhere the team can see and update them. Set a reminder to revoke unused connections every quarter.
  • How to get started: Open your ATS admin panel and list every active OAuth connection. For each one, confirm the vendor is still in use, the scopes are still appropriate, and the DPA is signed.
  • When it is a good time: Before any new vendor goes live and during any vendor offboarding.

When you are running live reqs and tools

  • What it means for you: Over-permissioned tokens and unrevoked connections from past vendors are the leading causes of unauthorized data access in recruiting stacks. They are also the hardest to detect because no alert fires when a stale token is used.
  • When it is a good time: Audit tokens whenever a vendor relationship ends, when a team member with admin access leaves, and quarterly as a standard hygiene check.
  • How to use it: Scope every token to minimum necessary permissions. Store credentials in a team vault, not in shared documents or chat. Log API calls at the ATS level and review anomalies monthly.
  • How to get started: Map each integration in your stack against three questions: what data can it read, what can it write, and when was the token last reviewed? Fix any that answer "full access" or "never."
  • What to watch for: Integrations that bundle too many scopes into a single authorization request, vendors who cannot provide a clear DPA, and webhooks configured to retry on failure without a dead-letter inbox.

Where we talk about this

On AI with Michal live sessions integration security comes up in the sourcing automation track every time we wire a new tool to an ATS: we stop at each token request and walk through scope review, credential storage, and what to log. The AI in recruiting track covers the policy side: how to explain API security to a GDPR officer, what to include in a vendor data flow map, and how to write a one-page integration security standard a small TA team can actually follow. Bring your current tool list to Workshops.

Around the web (opinions and rabbit holes)

Third-party content on API security ranges from deeply technical to practitioner-friendly. These links are starting points; verify any advice against your ATS vendor documentation and your legal team.

YouTube

Reddit

Quora

OAuth versus API key: when to use which

Credential typeBest forKey risk
OAuth token (scoped)User-authorized third-party toolsOver-permissioned scopes at authorization
API key (service account)Server-to-server automationLeaked key has full scope until rotated
OAuth with refresh tokensLong-running integrationsRefresh token stored insecurely

Related on this site

Frequently asked questions

What is OAuth in the context of recruiting tools?
OAuth is an authorization protocol that lets one app request limited access to data in another app without the user sharing their password. In recruiting, it appears every time you click "Connect with LinkedIn" or "Authorize this ATS integration": a token is issued with a defined set of scopes, specifying what data the integration can read or write and for how long. The recruiter grants access without exposing credentials, and the token can be revoked without changing the underlying account password. Most modern ATS API integrations and sourcing tool connections use OAuth 2.0 as the handshake layer between systems.
Why should recruiting teams care about API security?
Recruiting stacks process candidate PII across multiple connected tools: ATS, sourcing platforms, background check vendors, HRIS, email, and calendar. Each integration is a potential data path. A misconfigured webhook, an over-permissioned API key, or a token left active after a vendor offboarding can expose candidate records, leak compensation data, or create unauthorized writes to the ATS. In workflow automation setups, a single leaked key can trigger bulk actions affecting hundreds of candidates before anyone notices. GDPR and CCPA both treat unauthorized access to candidate data as a reportable incident, so security posture here is also a compliance posture.
What are the most common API security mistakes in recruiting stacks?
Five patterns appear repeatedly in team audits. Over-permissioned tokens: granting full read-write access when the integration only needs to read stage names. Shared credentials: a team Postman collection or Slack message containing a production API key. No rotation schedule: tokens issued during a pilot that were never revoked after the vendor evaluation ended. Missing dead-letter handling: recruiting webhooks that retry on failure without rate limiting, creating duplicate records. And no access log: nobody reviews which integrations called the ATS API last week, so a compromised token can operate quietly for months before someone audits the logs.
How should a small TA team set up API security without a dedicated IT function?
Start with three habits that do not require an IT team. First, scope tokens to minimum necessary permissions when authorizing any new integration: read-only unless the integration writes back. Second, document every active token in a shared credential store (1Password Teams or similar) with the issuing vendor, the scopes granted, and an expiry reminder set 30 days before renewal. Third, do a quarterly revocation audit: list every OAuth connection in your ATS admin panel and remove any that connect to vendors you no longer use. These three steps close the most common exposure paths before a dedicated security review is possible.
What does OAuth scoping mean in practice for an ATS integration?
When you authorize a sourcing tool to connect to your ATS, the OAuth flow asks you to approve a list of scopes: read candidates, write notes, read job postings, post offers, and so on. Accepting all scopes is the default path of least resistance, but it means the sourcing tool can write to your ATS even if you only needed it to read. Best practice is to review the scope request before authorizing. If a tool cannot explain why it needs a particular scope, decline that scope and ask the vendor. Most enterprise ATS platforms expose granular scopes; consumer-grade integrations often bundle too many together and require you to push back during procurement.
How does API security connect to GDPR compliance in recruiting?
GDPR requires a documented lawful basis for every data transfer and a record of where candidate data flows. Every API integration is a data transfer. If a sourcing tool sends candidate PII to an enrichment provider, and that provider is outside the EU without a valid transfer mechanism, you have a potential Article 46 violation regardless of whether a breach occurred. Start by mapping the data flows for each integration in your stack: what data leaves your ATS, to which vendor, stored in which jurisdiction, under what DPA. The ATS API integration article covers the technical plumbing; GDPR compliance is the policy layer on top of it.
Where can I learn how to secure recruiting integrations in a real stack?
The sourcing automation and AI in recruiting tracks at Workshops cover integration security in the context of live stack builds, including credential management, webhook error handling, and what to ask vendors during a DPA negotiation. The Starting with AI: the foundations in recruiting course addresses API concepts at the level a recruiter needs before they start connecting tools. Membership office hours are useful for specific questions like "is this token scope reasonable for this vendor" from practitioners who have already audited similar stacks.

← Back to AI glossary in practice