Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Yardstick ATS MCP Server move Sarah Chen to the offer stage for the backend engineer job"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Yardstick is an applicant tracking system. Its hosted MCP server
(https://api.yardstick.team/mcp) gives an AI assistant 176 tools for jobs, candidates,
applications, interviews, and scorecards — so you can ask an assistant to move a
candidate forward, draft outreach, or pull a hiring report, in whatever app you
already work in.
The server is hosted: nothing to install, nothing to run, and no API key to paste into a configuration file. Point your client at the URL and sign in through your browser. It works in Claude, ChatGPT, and Claude Code; Cursor connects through a local server instead, described at https://portal.yardstick.team/docs/mcp/developer-tools along with the Yardstick CLI.
Server URL: https://api.yardstick.team/mcp
Documentation: https://portal.yardstick.team/docs/mcp
Tool reference: https://portal.yardstick.team/docs/mcp/tools
Registry name: team.yardstick/yardstick
Claude Code
claude mcp add --transport http yardstick https://api.yardstick.team/mcpThen run /mcp and finish the browser sign-in.
Claude, ChatGPT, Perplexity, and other chat apps
Add a custom connector in the app's settings and paste in the URL above. The app reads the server and lists the tools it found.
Editors that use an mcp.json file
{
"mcpServers": {
"yardstick": {
"url": "https://api.yardstick.team/mcp"
}
}
}Step-by-step instructions for each app, including the ones that need a different shape, are at https://portal.yardstick.team/docs/mcp/chat-apps and https://portal.yardstick.team/docs/mcp/developer-tools.
Related MCP server: external-ai-ecosystem-gateway
Browsing the tool list needs no account: a client can connect and see what is available before anyone signs in. Everything that touches hiring data requires a signed-in Yardstick user, and the assistant can only reach what that person is allowed to reach.
Sign-in is OAuth 2.1 with PKCE, in your browser — there is no credential to copy anywhere.
Yardstick identifies a connecting app through a Client ID Metadata Document rather than OAuth Dynamic Client Registration. Apps that support the metadata-document flow find this on their own, and sign-in needs nothing from you. An app that can only register dynamically stops with an error naming dynamic client registration; that is not a misconfiguration on your side — use the Yardstick CLI or a local MCP server with that app instead.
The server reads and writes your organization's own Yardstick data, acting for whoever is signed in and limited to what that person can already reach in the app. Connecting an assistant does not widen anyone's access.
What leaves Yardstick is whatever the assistant asks for, and it goes to the assistant you connected — so that app's own terms govern what happens to it afterward. A Yardstick administrator can revoke a connection at any time from your account settings, which cuts off its access immediately.
Yardstick's own handling of your data is unchanged by this connection and is covered by the documents below.
Privacy Policy — https://www.yardstick.team/privacy-policy
Data Processing Agreement — https://www.yardstick.team/dpa
Terms of Service — https://www.yardstick.team/terms
A tool marked read-only will not change anything. The rest are not marked that way: most of them write, and a few only report status but carry no read-only guarantee. Expect your assistant to say what it is about to do before it calls one.
Tool | What it does | Read-only |
| Return the current Yardstick API principal context. | Yes |
| List Yardstick tasks visible to the current principal. | Yes |
| Create a public task deep link for a caller-supplied task id. | No |
| List people visible to the current principal. | Yes |
| Read one person by public id. | Yes |
| Create a person through the public Yardstick API. | No |
| Update one person with schema-bound public API fields. | No |
| List active Talent CRM prospects visible to the current principal. | Yes |
| Read one Talent CRM prospect by public id. | Yes |
| Create Talent CRM prospect relationship state for an existing person. | No |
| Preview conversion of one Prospect into one canonical Application; bind fields are rejected and no outreach/import/merge occurs. | No |
| Bind an approved prospect conversion preview with explicit approval metadata and matching idempotency key. | No |
| Preview a one-off prospect outreach draft without creating a draft, sending email, enrolling a sequence, or creating an application. | Yes |
| Create a pending review prospect outreach email draft. Sending stays on email draft send and requires communications:send. | No |
| Preview a structured-row Prospect CRM import. JSON rows only; no raw CSV/resume file reads, outreach, application conversion, source ranking, or silent person merge. | No |
| Read one Prospect CRM import dry-run/apply batch by public id. | Yes |
| Apply an approved Prospect CRM import preview with preview_hash and explicit row decisions for advisory matches. | No |
| Preview a structured-row ATS import of candidates, jobs, and applications (source_kind: csv). JSON rows only; no raw CSV/archive file reads, no application conversion side effects, and low-confidence stage/fuzzy matches require explicit apply decisions. | No |
| Read one ATS import dry-run/apply batch and its rows by public id. | Yes |
| Apply an approved ATS import preview with a preview_hash matching the stored dry-run. Creates people/jobs/interview_plans(stub)/job_postings/applications with idempotent replay; never sends outreach, merges people, or deletes records. | No |
| List source/referral attribution links without raw token material or source quality ranking. | Yes |
| Read one source/referral attribution link by public id without raw token material. | Yes |
| Create an opaque source/referral link for a published job posting. Raw source_token is returned only in the create response; reads return token_hint, and active token-bearing links also return public_url. | No |
| Disable a source/referral link so its public URL no longer stamps attribution. | No |
| Update bounded Talent CRM prospect relationship fields. | No |
| List bounded activity records for one Talent CRM prospect. | Yes |
| Create a bounded internal prospect activity note or touchpoint. | No |
| List Talent CRM pools visible to the current principal. Archived pools are excluded unless include_archived is true. | Yes |
| Read one Talent CRM pool by public id. | Yes |
| Create a Talent CRM pool. | No |
| Update bounded Talent CRM pool fields. | No |
| List active prospect memberships for one Talent CRM pool. | Yes |
| Add an existing prospect to a Talent CRM pool. | No |
| Remove an active prospect membership from a Talent CRM pool. | No |
| List application person + Job attachments visible to the current principal. | Yes |
| Read one application person + Job attachment by public id. | Yes |
| Create a recruiter or agent-sourced application for an existing person and job posting. | No |
| Update one application with schema-bound public API fields. | No |
| Directly write a non-terminal application transition; approval-only statuses are rejected. Optionally target a configured stage by id with pipeline_stage_id (the resolved stage is reported back). | No |
| Preview approval-required application transitions only; bind fields and direct-write targets are rejected. Optionally target a configured stage by id with pipeline_stage_id (status must match the stage's mapped status). | No |
| Bind an already-previewed application transition with explicit approval metadata and matching idempotency key. | No |
| Preview application archive approval requirements only; bind fields are rejected. | No |
| Bind an already-previewed application archive with explicit approval metadata and matching idempotency key. | No |
| Preview the fixed applications-owned Interviewing to Offer status/email approval POC without mutating status or email state. | No |
| Bind the fixed applications-owned status/email approval POC with explicit approval metadata and a matching idempotency key. | No |
| Read the decision brief for an application. | Yes |
| Regenerate a decision brief for an application with idempotency metadata. | No |
| Preview the candidate email decision for an application status change without mutating the application or drafts. | No |
| Apply an application status-change email decision with idempotency metadata. | No |
| List candidate email drafts with optional status and email-type filters. | Yes |
| Read a redacted candidate email draft. | Yes |
| Read rendered preview content for a candidate email draft. | Yes |
| Read candidate email draft history and provenance. | Yes |
| Save mutable candidate email draft fields or set/clear a validated sender override. | No |
| Send a pending candidate email draft with replay protection, or reclaim a never-delivered draft and send it. | No |
| Discard a pending candidate email draft without provider mutation. | No |
| Read candidate email automation settings. | Yes |
| Update candidate email automation settings. | No |
| Read redacted email sender settings with sanitized queue diagnostics. | Yes |
| List redacted sender provider connection state. | Yes |
| List candidate email sender defaults. | Yes |
| Read sanitized Workspace integration health diagnostics. | Yes |
| Set the default sender policy for a candidate email type. | No |
| List candidate email template policies without raw prompts. | Yes |
| Read candidate email template detail without raw prompts. | Yes |
| Preview a candidate email template for sample variables. | No |
| Validate candidate email template subject, HTML, and text bodies. | No |
| Publish a validated candidate email template version. | No |
| List immutable candidate email template versions. | Yes |
| Rollback a candidate email template to an immutable version. | No |
| Preview AI sample customization for a candidate email template without publishing. | No |
| Record direct final-recommendation evidence; this does not mutate application status. | No |
| List interview sessions for an application. | Yes |
| Ensure an application interview session for an explicit round id. | No |
| List read-only opportunity workflow views visible to the current principal. | Yes |
| Read one read-only opportunity workflow view by public id. | Yes |
| Read the next action state for an opportunity without mutating it. | No |
| Create a public deep link for an opportunity workflow view. | No |
| Search redacted recruiting reporting summaries through the public reporting API. | Yes |
| List Departments visible to the current principal. | Yes |
| Read one Department by public id. | Yes |
| Create a Department through the public Yardstick API. | No |
| Update one Department with schema-bound public API fields. | No |
| List Department members for an authorized Department. | Yes |
| Add an existing person to a Department. | No |
| Set one Department member as lead. | No |
| Remove a Department member, optionally with a replacement lead. | No |
| List redacted Department invitation views. | Yes |
| Create a Department-scoped invitation without returning token material. | No |
| List explicit access grants visible to the current principal. | Yes |
| Read one explicit access grant by public id. | Yes |
| Create a reasoned, revocable, audited access grant. | No |
| Revoke an explicit access grant with optional reason text. | No |
| List internal Job containers visible to the current principal. | Yes |
| Create an internal Job container without mutating job postings, applications, or interview plans. | No |
| Read one internal Job container by public id. | Yes |
| List deterministic Talent CRM prospect matches for one Job. Read-only; includes evidence, caveats, and blocked lifecycle states without outreach, conversion, import, or AI ranking. | Yes |
| Update one internal Job container with schema-bound public API fields. | No |
| Close one Job row only; linked job postings, applications, and interview plans are not mutated. | No |
| Archive one Job row only; linked job postings, applications, and interview plans are not mutated. | No |
| Link an existing interview plan, job posting, or application to a Job. Idempotent when already linked to the same Job; conflicts when linked elsewhere. Requires jobs:write and the artifact's own write scope. | No |
| Clear the link between an artifact and a Job. Idempotent when already unlinked. Requires jobs:write and the artifact's own write scope. | No |
| List the organization's pipeline templates with their ordered stages. Canonical default stages are flagged canonical_locked. | Yes |
| Read one pipeline template and its ordered stages by public id. | Yes |
| Create a custom pipeline template with optional inline stages. Does not affect the org default pipeline or candidate tracking. | No |
| Rename, re-describe, or toggle active for one pipeline template. | No |
| Append (or insert at a position) a stage to a pipeline with a kind and optional approval scope. | No |
| Edit a pipeline stage. Canonical default stages allow display-name change only; slug/stage_kind/approval_scope/is_terminal are locked. | No |
| Reorder all stages of a pipeline by supplying the complete stage id order. Constraint-safe; allowed for canonical stages. | No |
| Preview deleting a pipeline stage; returns the candidate-count impact and binding fields. Canonical default stages are rejected. | No |
| Bind an approved pipeline-stage delete with explicit approval metadata and a matching idempotency key. Refuses while candidates remain. | No |
| Assign a pipeline to a job (jobs.pipeline_template_id) or clear it with pipeline_id: null. Pipeline validated by org; job resolved by org+mode. | No |
| List the organization's stage automation rules. Rules are inert config — authoring fires no side effect; the executor child runs rules later. | Yes |
| List the organization's stage automation runs (what fired for which candidate, success/failure). Filterable by application-id and rule-id. Read-only; runs are system-written by the executor and include skipped (deferred) send_email rows so run state is fully observable. | Yes |
| Read one stage automation rule by public id (its trigger, typed action, target stage, and is_active state). | Yes |
| Create a stage automation rule (trigger + typed action from the 3-type palette: send_scheduling_link / create_task / send_email). Inert — fires no side effect. | No |
| Edit a stage automation rule's trigger, typed action, condition, position, or is_active state. | No |
| Delete a stage automation rule. Plain delete (no approval preview→bind): rules are inert config, so deleting unfired config moves no candidates. | No |
| Dry-run preview: resolve the active rules + their resolved action params for a sample application entering/leaving a stage. INERT — mints no scheduling link, creates no task, sends no email; executes nothing. | Yes |
| List the built-in-agent catalog with each built-in agent's current status, explicit defaults or disabled state, and latest work summary. | Yes |
| Read one built-in agent by catalog id (for example scorecard-chaser, nightly-screening, or interview-kit-prep): what it does, what it never does without you, current status, and latest work summary. | Yes |
| List the organization's built-in-agent run history as a flat work report. Nightly screening reports scheduled-work counts and held, released, or invalidated review outputs; interview-kit preparation reports the scheduled week and count-only results. It never includes candidate, draft, recipient, kit, or provider details. delivery_unknown is not delivered and is not retried. Filterable by built-in-agent id. Read-only and mode-scoped. | Yes |
| Preview enabling a built-in agent; returns the approval-required envelope stating exactly what starts happening (and what is never sent without you) plus binding fields. Side-effect free. | No |
| Bind an approved built-in-agent enable preview with explicit approval metadata and a matching idempotency key. Writes the built-in agent's declared settings home and records one audit row. | No |
| Preview configuring a built-in agent (config is validated against the built-in agent's declared schema; enabled is a first-class key, so { enabled: false } is the off switch). Side-effect free; pass config via --body-json. | No |
| Bind an approved built-in-agent configure preview (same config plus approval metadata and a matching idempotency key). Writes only the built-in agent's declared settings keys and records one audit row. | No |
| List job postings visible to the current principal. | Yes |
| Read one job posting by public id. | Yes |
| Validate a draft job posting without mutating it. | No |
| Create a draft job posting through the public Yardstick API. | No |
| Update a job posting with schema-bound public API fields. | No |
| Preview job-posting publish approval requirements only; bind fields are rejected. | No |
| Bind an already-previewed job-posting publish with explicit approval metadata and matching idempotency key. | No |
| Preview job-posting archive approval requirements only; bind fields are rejected. | No |
| Bind an already-previewed job-posting archive with explicit approval metadata and matching idempotency key. | No |
| Read one interview plan by public id. | Yes |
| Generate draft interview-plan outcomes from explicit role context, or pass accepted_outcomes to store caller-written outcomes verbatim with no AI generation (the deterministic draft bootstrap). | No |
| Confirm draft interview-plan outcomes before generating later stages. | No |
| Generate draft interview-plan competencies from confirmed outcomes. | No |
| Generate draft interview-plan skills from confirmed competencies. | No |
| Generate draft interview-plan rounds from confirmed skills. | No |
| Generate scorecard behavioral anchors (levels 1-4) for each draft competency, or one competency when competency_id is supplied. | No |
| Generate interview questions and interviewer/candidate directions for each draft round, or one round when round_id is supplied. | No |
| Read one interview-plan draft in full: every stage checkpoint (outcomes, competencies, job skills, interview sequence, questions and directions), stage_versions for source_stage preconditions, status, and source metadata. Question ids are always present (question_N fallback), so edits can target them directly. | Yes |
| Add competencies to a draft exactly as written — deterministic, no AI generation. Requires confirmed outcomes; the resulting stage must stay within 25 competencies. Optional source_stage {stage: competencies, version} returns 409 when stale. | No |
| Replace draft competencies by id and/or reorder them with order (a complete permutation of resulting ids) — deterministic, no AI generation. Updating in place keeps the item id. Optional source_stage {stage: competencies, version} returns 409 when stale. | No |
| Delete draft competencies by id. Prunes each deleted competency's behavioral anchors in the same write and reports the pruned ids; the stage must keep at least one competency. Optional source_stage {stage: competencies, version} returns 409 when stale. | No |
| Add job skills to a draft exactly as written — deterministic, no AI generation. Requires competencies; the resulting stage must stay within 50 skills. Optional source_stage {stage: skills, version} returns 409 when stale. | No |
| Replace draft job skills by id and/or reorder them with order (a complete permutation of resulting ids) — deterministic, no AI generation. Optional source_stage {stage: skills, version} returns 409 when stale. | No |
| Delete draft job skills by id; the stage must keep at least one skill. Optional source_stage {stage: skills, version} returns 409 when stale. | No |
| Add interview rounds to a draft exactly as written — deterministic, no AI generation. Requires skills; the resulting sequence must stay within 20 rounds. Optional source_stage {stage: interview_sequence, version} returns 409 when stale. | No |
| Replace draft interview rounds by id and/or reorder them with order (a complete permutation of resulting ids) — deterministic, no AI generation. A round updated in place keeps its id, so its questions and directions are preserved (not pruned). Optional source_stage {stage: interview_sequence, version} returns 409 when stale. | No |
| Delete draft interview rounds by id. Prunes each deleted round's questions and directions in the same write and reports the pruned ids; the sequence must keep at least one round. Optional source_stage {stage: interview_sequence, version} returns 409 when stale. | No |
| Add questions to one draft round exactly as written — deterministic, no AI generation. Requires rounds; each round holds at most 50 questions. The first question edit on a round persists question_N ids for every question in that round. Optional source_stage {stage: questions_and_directions, version} returns 409 when stale. | No |
| Replace one draft round's questions by id and/or reorder them with order (a complete permutation of that round's resulting question ids) — deterministic, no AI generation. Read the draft first: stored questions expose stable question_N ids. Optional source_stage {stage: questions_and_directions, version} returns 409 when stale. | No |
| Delete questions from one draft round by question id. Optional source_stage {stage: questions_and_directions, version} returns 409 when stale. | No |
| Set one draft round's interviewer and/or candidate directions exactly as written — deterministic, no AI generation. Provide at least one of the two direction fields. Optional source_stage {stage: questions_and_directions, version} returns 409 when stale. | No |
| Upsert behavioral anchors for draft items, keyed by item id with levels 1-4 — deterministic, no AI generation. Requires competencies; anchors keyed to unknown item ids are rejected by id. Optional source_stage {stage: questions_and_directions, version} returns 409 when stale. | No |
| Delete behavioral-anchor entries for draft items by item id. Optional source_stage {stage: questions_and_directions, version} returns 409 when stale. | No |
| Validate an interview-plan draft without publishing it. | No |
| Preview interview-plan publish approval requirements only; bind fields are rejected. | No |
| Bind an already-previewed interview-plan publish with explicit approval metadata and matching idempotency key. | No |
| Preview interview-plan active-modification approval requirements only; bind fields are rejected. | No |
| Bind an already-previewed interview-plan active modification with explicit approval metadata and matching idempotency key. | No |
| Read one interview session by public id. | Yes |
| List interviewers assigned to an interview session. | Yes |
| Assign an interviewer to an interview session. | No |
| Unassign an interviewer from an interview session. | No |
| Start an interview session for an explicit interviewer. | No |
| Complete an interview session for an explicit interviewer. | No |
| Submit an interview-session scorecard with explicit ratings. | No |
| Read interview-session scheduling readiness. | No |
| Generate a scheduling link for an interview session. | No |
| Cancel a scheduled interview session, releasing the provider booking and invalidating its scheduling links. | No |
| List public Yardstick API events visible to the current principal. | Yes |
| List public Yardstick audit log entries visible to the current principal. | Yes |
The 176 tools above are everything the hosted server offers. Another 6 exist — 182 in total — and cannot be reached over this connection. Some of them upload a file from your own machine, which the hosted server has no way to read. The rest connect Yardstick to an outside email provider, which needs that provider's own browser sign-in. Both groups work normally in the Yardstick CLI and in a local MCP server, documented at https://portal.yardstick.team/docs/cli and https://portal.yardstick.team/docs/mcp/developer-tools.
The tool reference gives each of them the same one-line explanation:
Not listed on hosted remote MCP; calling it there returns an explanation instead. Use the CLI, a local MCP client, or the dashboard.
Tool | What it does |
| Upload a Greenhouse export archive (.zip) for import (create session -> PUT bytes to a signed URL). Reads a single schema-bound local file path; not available over hosted MCP. Then run |
| Upload a PDF or DOCX resume for a person (create session -> PUT bytes -> finalize). Reads a single schema-bound local file path; not available over hosted MCP. |
| Finalize a previously created resume upload session: re-validate the uploaded object, attach it, and parse linked applications. Not available over hosted MCP. |
| Prepare a browser OAuth handoff for an email sender connection; hosted MCP refuses execution. |
| Refresh sender identity discovery for a provider connection; hosted MCP refuses execution. |
| Disconnect an email sender provider connection; hosted MCP refuses execution. |
This is the public home of a hosted service: documentation, plus the manifest files that directory listings read. There is no product source here and nothing to build.
server.json — the Model Context Protocol registry entry
glama.json and lhm.plugin.json — directory manifests
LICENSE — MIT, covering the contents of this repository
Issues and questions about connecting to the server are welcome here. The server itself is operated by Yardstick; for anything account-specific, contact support@yardstick.team.
This server cannot be deployed
The OpenRouter for tools. One MCP connection gives any AI agent 254 hosted tools, pay per call.
Zero-setup MCP gateway securely connecting AI to your tools with authentication and workflows
Unified MCP Server is a remote MCP connector for AI agents and vertical AI products that provides access to 22,000+ authorized SaaS tools across 400+ integrations and 24 categories directly inside LLMs (Claude, GPT, Gemini, Cohere). Tools operate only on explicitly authorized customer connections, enabling agents to safely read and write against live third-party systems.
7 recruiting tools over one MCP endpoint: ATS boards, LinkedIn jobs, profiles, companies, Naukri.
