Verify by Tailr Made
HRConnect AI agents to Verify by Tailr Made, an AI resume verification and fraud-detection tool for recruiters. Agents submit resume text for analysis and receive a risk level, summary, and structured red-flag report so screening happens before an interview is ever scheduled.
What This Integration Enables
Verify by Tailr Made does one thing. Send it the text of a resume and it returns a risk level, a plain text summary, a formatted HTML report, and structured analysis containing the companies, titles, employment dates, technologies, and contact details it detected, along with timeline, technology, and education concerns. One action, one decision it informs. A single action connector is not a thin connector when the action is this consequential.
Being precise about what it is helps more than a feature list. This is document analysis, not background screening. The service says so itself: it does not perform identity verification and it does not perform biometric checks. It has not called a previous employer, checked a registry, or confirmed a degree. It has read text and reasoned about whether that text is internally consistent and plausible. That is genuinely useful, because most resume problems are visible in the document, and it is genuinely limited, because a well written fabrication is internally consistent too.
For a recruiting operations team the value is consistency of attention, not judgment. Every application in scope gets the same structured pass, at the same depth, on the busiest week of the quarter as on the quietest. What comes back is a set of specific things to ask about, attached to the passages that raised them. Each analysis is billed per resume by Tailr Made, so a flow that runs this over a backlog is a spend decision as well as a screening decision, and it is worth putting a scope and a ceiling on the flow rather than on the invoice. Authentication is an API key sent in the X-API-Key header, created in the Tailr Made developer dashboard. FlowRunner's connectors are built and verified against each vendor's official API.
Without FlowRunner
With FlowRunner
Use Case Scenarios
Screening That Produces Questions, Not Verdicts
An application lands in the applicant tracking system with a CV attached. The agent extracts the text, confirms the extraction is clean enough to reason over, and runs Verify Resume. Most come back with a low risk level and nothing to discuss, and those move to the recruiter's normal review untouched. Where concerns are returned, the agent builds an interview preparation note rather than a decision: each flagged item, the resume passage it came from, and a suggested question a person could ask about it. That note is posted to the recruiter in Slack and attached to the candidate record in JOIN. The recruiter still decides who to interview. They just walk in knowing what to ask.
The Extraction Check That Comes First
Resume text arrives from parsed PDFs, and PDF parsing is uneven. A two column layout can interleave two jobs into one line, and a scanned document can lose date ranges entirely. Since Verify Resume reasons over whatever text it is given, a bad extraction produces confident concerns about a timeline that was never in the original document. So the agent scores the extraction first: does the text contain plausible date ranges, do company names appear as contiguous strings, is the character count consistent with the page count. Resumes that fail that check are routed to a person for manual extraction and never scored automatically. It is cheaper to spend a minute on the parse than to ask a candidate to explain a gap that a PDF library invented.
A Consistent Standard Across a Hiring Wave
A team opens twelve roles at once and application volume triples. The first fifty resumes get read carefully and the next four hundred get skimmed, which is a volume problem rather than a diligence problem. Instead, the agent applies the same pass to every application within the defined scope and records the risk level and concern categories against each candidate. A weekly view then shows where the concerns cluster. If one sourcing channel produces a much higher rate of flagged resumes than the others, that is a fact about the channel worth investigating, and the agent surfaces it to recruiting operations. It does not act on it. Suppressing a channel and rejecting the people who came through it are very different decisions.
Human-in-Loop Highlight
This is the sharpest gate in this category, and it is not a close call: a risk level from this connector must never reject an application by itself. The vendor gives the levels as examples rather than as a documented threshold, does not perform identity verification, and derives every concern from text. Meanwhile a career break for caregiving, a contractor holding two overlapping engagements, military service, a company that was acquired and renamed, and a CV translated from another language all produce exactly the pattern a timeline concern describes. The connector cannot tell those apart from a fabrication, because from the text they are the same shape.
So the agent's job ends at the question. When a medium or high risk level comes back, it assembles the flagged items with the resume passage that produced each one and hands the recruiter a decision, not a verdict: "Candidate resume flagged medium. Two concerns. Employment dates for Company A and Company B overlap by seven months. A stated certification is dated before the technology it covers was released. Extraction quality was clean. Ask the candidate about both, route to the hiring manager, or dismiss the flags?" The recruiter chooses. If the answer is to ask, the candidate gets asked about something specific and gets to answer, which is the outcome a human reviewer would have produced on their most attentive day.
That is human-in-the-loop doing the work it exists for. The agent handles the part that does not need judgment, which is reading every resume with the same care. The part that decides whether a named person gets a chance at a job stays with a person who can be asked why.
Agent Capabilities
1 actionsResume Analysis
1- Verify Resume Analyze resume text for fraud signals and red flags before an interview. Accepts the raw text of a resume, or text already extracted from a PDF, and returns a risk level, a plain text summary, a formatted HTML report, and structured analysis including detected companies, titles, employment dates, technologies, and contact details, plus timeline, technology, and education concerns. This does not perform identity verification or biometric checks, and each analysis is billed per resume by Tailr Made.
Frequently Asked Questions
What can FlowRunner do with Verify by Tailr Made?
FlowRunner agents can run Verify Resume in Verify by Tailr Made.
Does connecting Verify by Tailr Made to FlowRunner require OAuth?
No. Verify by Tailr Made connects to FlowRunner with an API key, no OAuth flow required.
Can Verify by Tailr Made trigger a FlowRunner workflow automatically?
Verify by Tailr Made doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Start building with Verify by Tailr Made
$100 in credits. No card required. Connect in minutes.