# ApplySent (remote · applysent.com)

Search jobs, preview each application and apply with your OK on Workday, Greenhouse, Ashby and more.

- Trust score: 83/100 (high trust)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `applysent.com`: 83/100 (this document), [markdown](https://verifymcp.io/servers/com-applysent-applysent/api-mcp.md), [page](https://verifymcp.io/servers/com-applysent-applysent/api-mcp)

## Channel facts

- Endpoint: `https://applysent.com/api/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `1.5.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 97/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
  - The authorisation server supports Client ID Metadata Documents, the current MCP client-registration mechanism.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 77/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 3552 tokens (~101/item across 35 items; 35 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 15/100
  - Stability check failed: schema churn in the 9 days we've observed: 4 tool removals, 0 breaking changes, 0 auth/transport breaks, 13 additions.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 36 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the ApplySent MCP server?

ApplySent is a hosted endpoint at https://applysent.com/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http com-applysent-applysent 'https://applysent.com/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-applysent-applysent": {
      "url": "https://applysent.com/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-applysent-applysent": {
      "type": "http",
      "url": "https://applysent.com/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-applysent-applysent]
url = "https://applysent.com/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-applysent-applysent": {
      "type": "remote",
      "url": "https://applysent.com/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-applysent-applysent --url 'https://applysent.com/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-applysent-applysent:
    url: "https://applysent.com/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-applysent-applysent": {
      "Transport": "http",
      "Url": "https://applysent.com/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-applysent-applysent -t streamable-http -u 'https://applysent.com/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-applysent-applysent": {
      "type": "http",
      "url": "https://applysent.com/api/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 83, +1)

No change was recorded against any check on this day. Stability & Change Management went from 11 to 15.

### 2026-11-19 (score 82, +1)

No change was recorded against any check on this day. Stability & Change Management went from 5 to 8.

### 2026-11-15 (score 81, +1)

- [security regression] Stability: 0.07 → fail
- [security regression] Tool “get_autopilot” was removed
- [security regression] Tool “preview_autopilot” was removed
- [security regression] Tool “set_autopilot” was removed
- [security regression] Tool “update_autopilot_filters” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] New tool “set_job_feedback”, which the server declares destructive
- [security] Tool “fetch” rewrote its description, which is the text the model reads
- [security] Tool “get_account” rewrote its description, which is the text the model reads
- [security] Tool “get_application” rewrote its description, which is the text the model reads
- [security] Tool “get_application_review” rewrote its description, which is the text the model reads
- [security] Tool “get_job_from_link” rewrote its description, which is the text the model reads
- [security] Tool “get_profile_status” rewrote its description, which is the text the model reads
- [security] Tool “get_thread” rewrote its description, which is the text the model reads
- [security] Tool “queue_application_for_review” rewrote its description, which is the text the model reads
- [security] Tool “answer_question” rewrote its description, which is the text the model reads
- [security] Tool “apply_to_job” rewrote its description, which is the text the model reads
- [security] Tool “list_pending_questions” rewrote its description, which is the text the model reads
- [security] Tool “preview_application” rewrote its description, which is the text the model reads
- [security] Tool “regenerate_tailored_resume” rewrote its description, which is the text the model reads
- [security] Tool “search” rewrote its description, which is the text the model reads
- [security] Tool “regenerate_tailored_resume” is now declared destructive
- [functional regression] Schema quality: 85 → 101
- [functional regression] Schema quality: 85 → 102
- [functional improvement] Tool coverage: 66% → 100%
- [functional] Schema quality: good → excellent
- [functional] Server version: 1.3.0 → 1.5.0
- [functional] Server version: 1.1.0 → 1.3.0
- [functional] New tool “get_application_board”
- [functional] New tool “get_application_review”
- [functional] New tool “get_metrics”
- [functional] New tool “get_notification_settings”
- [functional] New tool “list_blocked_companies”
- [functional] New tool “list_notifications”
- [functional] New tool “list_saved_jobs”
- [functional] New tool “mark_notification_opened”
- [functional] New tool “queue_application_for_review”
- [functional] New tool “set_company_blocked”
- [functional] New tool “set_job_saved”
- [functional] New tool “update_notification_settings”
- [cosmetic] “answer_question” reworded the description of “idempotency_key”
- [cosmetic] “apply_to_job” reworded the description of “idempotency_key”
- [cosmetic] “approve_tailored_resume” reworded the description of “idempotency_key”
- [cosmetic] “cancel_application” reworded the description of “idempotency_key”
- [cosmetic] “get_application_board” reworded the description of “limit”
- [cosmetic] “get_application_board” reworded the description of “offset”
- [cosmetic] “get_job_from_link” reworded the description of “idempotency_key”
- [cosmetic] “get_metrics” reworded the description of “days”
- [cosmetic] “list_blocked_companies” reworded the description of “limit”
- [cosmetic] “list_blocked_companies” reworded the description of “offset”
- [cosmetic] “list_notifications” reworded the description of “limit”
- [cosmetic] “list_notifications” reworded the description of “offset”
- [cosmetic] “list_saved_jobs” reworded the description of “limit”
- [cosmetic] “list_saved_jobs” reworded the description of “offset”
- [cosmetic] “list_saved_jobs” reworded the description of “sort”
- [cosmetic] “mark_notification_opened” reworded the description of “idempotency_key”
- [cosmetic] “preview_tailored_resume” reworded the description of “idempotency_key”
- [cosmetic] “queue_application_for_review” reworded the description of “idempotency_key”
- [cosmetic] “regenerate_tailored_resume” reworded the description of “idempotency_key”
- [cosmetic] “search_jobs” reworded the description of “salary_currency”
- [cosmetic] “set_company_blocked” reworded the description of “idempotency_key”
- [cosmetic] “set_job_feedback” reworded the description of “idempotency_key”
- [cosmetic] “set_job_feedback” reworded the description of “signal”
- [cosmetic] “set_job_saved” reworded the description of “idempotency_key”
- [cosmetic] “update_notification_settings” reworded the description of “idempotency_key”
- [cosmetic] “answer_question” reworded the description of “batch_id”
- [cosmetic] “answer_question” reworded the description of “question_index”
- [cosmetic] “fetch” reworded the description of “id”
- [cosmetic] “search_jobs” reworded the description of “nationals_only”
- [cosmetic] “get_jobs” reworded the description of “job_ids”
- [cosmetic] “get_message” reworded the description of “message_id”
- [cosmetic] “get_thread” reworded the description of “message_id”
- [cosmetic] “list_applications” reworded the description of “limit”
- [cosmetic] “list_applications” reworded the description of “offset”
- [cosmetic] “list_applications” reworded the description of “started_by_this_client”
- [cosmetic] “list_applications” reworded the description of “status”
- [cosmetic] “list_inbox” reworded the description of “category”
- [cosmetic] “list_inbox” reworded the description of “limit”
- [cosmetic] “list_pending_questions” reworded the description of “limit”
- [cosmetic] “list_pending_questions” reworded the description of “offset”
- [cosmetic] “search_jobs” reworded the description of “city”
- [cosmetic] “search_jobs” reworded the description of “include_applied”
- [cosmetic] “search_jobs” reworded the description of “limit”
- [cosmetic] “search_jobs” reworded the description of “min_salary”
- [cosmetic] “search_jobs” reworded the description of “min_score”
- [cosmetic] “search_jobs” reworded the description of “offset”
- [cosmetic] “search_jobs” reworded the description of “seniority”
- [cosmetic] “search_jobs” reworded the description of “sort”
- [cosmetic] “search_jobs” reworded the description of “workplace”
- [cosmetic] Tool “fetch” changed its title: Read a job (citations) → Fetch a job's text
- [cosmetic] Tool “get_account” changed its title: Read plan and usage → Read application usage
- [cosmetic] Tool “get_jobs” changed its title: Read several jobs → Read jobs
- [cosmetic] Tool “regenerate_tailored_resume” changed its title: Tailor a résumé again → Regenerate tailored résumé
- [cosmetic] Tool “search” changed its title: Search jobs (citations) → Search jobs

### 2026-11-14 (score 80, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-13 (score 79, 0)

- [functional improvement] Stability: unverified → 0.03

### 2026-11-12 (score 79, +40)

- [security improvement] Transport: unverified → pass
- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Authorization: Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Tool coverage: unverified → 100
- [functional] First check of Schema quality: fail
- [functional] First check of Destructive annotations: 100
- [functional] First check of Tool coverage: 66
- [functional] First check of Schema quality: pass
- [functional] First check of Schema quality: good
- [functional] First check of Tool coverage: 100

### 2026-11-11 (score 39)

First indexed and scored.

## MCP tools (35)

### `search_jobs` (~426 tokens)

Search matching jobs

Search the jobs that match the user's résumé and filters, best match first, a page at a time. Leaves out jobs already applied to unless include_applied is true, and postings open only to nationals unless the profile lists a nationality or nationals_only says otherwise. Reads only.

Input parameters:

- `city` (string): City the posting is in, e.g. "Dubai" or "London".
- `country` (string): Country name or ISO code, e.g. "US", "United Kingdom", "AE".
- `employment_type` (string): e.g. full_time, part_time, contract.
- `include_applied` (boolean): Include jobs already applied to; they are left out otherwise.
- `limit` (integer): How many jobs to return.
- `min_salary` (number): Only jobs stating pay of at least this; a job with no stated salary is left out.
- `min_score` (integer): Only jobs the résumé matches at least this well; leave it out for no floor.
- `nationals_only` (string): Whether postings open only to a country's nationals are left out, included, or the only ones kept; leaving it out excludes them unless the profile lists a nationality.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.
- `posted_since` (string): ISO date; only jobs posted on or after it. Jobs with no known posting date are left out.
- `query` (string): Words in the title, employer or description.
- `salary_currency` (string): ISO code of the currency the stated pay must be in: with min_salary the floor is read in that currency, on its own it keeps the jobs that state pay in it.
- `seniority` (string): Seniority as postings state it, e.g. "senior", "junior", "lead".
- `sort` (string): Order of the page: best match first, or newest posting first.
- `workplace` (string): Where the work is done; leave it out for any.

Output parameters:

- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `jobs` (array): The page of matching jobs, best match first or newest first as asked.
- `more` (boolean): True when later pages hold more jobs.
- `offset` (integer): The offset this page starts at.
- `total` (integer): How many jobs the filters match in all.

### `list_saved_jobs` (~94 tokens)

List saved jobs

The jobs the user saved, with each one's match assessment, best match first or newest first, a page at a time. Reads only.

Input parameters:

- `limit` (integer): How many saved jobs to return.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.
- `sort` (string): Order of the page: best match first, or newest first.

Output parameters:

- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `jobs` (array): The page of saved jobs, best match first or newest first as asked.
- `more` (boolean): True when later pages hold more saved jobs.
- `offset` (integer): The offset this page starts at.
- `total` (integer): How many jobs the user saved.

### `set_job_saved` (~97 tokens)

Save or unsave a job

Saves one job to the user's saved jobs, or removes it, and returns the job with its match assessment.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.
- `saved` (boolean, required): true saves the job, false removes it.

Output parameters:

- `apply_supported` (boolean): Whether ApplySent can fill and submit this posting's form by itself.
- `apply_url` (string): The employer's posting URL, where the application is filled.
- `city` (string|null): City the posting is in.
- `company` (string): The employer's name.
- `country` (string|null): Country the posting is in, as the catalogue recorded it.
- `excerpt` (string): The start of the posting's text, plain, for a list row.
- `id` (string): The job id the other job tools take.
- `match` (object): The assessment of this job against the user's résumé, targets and eligibility.
- `posted_at` (string|null): When the posting went up, ISO 8601 in UTC, when the catalogue knows it.
- `provider` (string): The employer's hiring system the posting came from, e.g. workday, greenhouse or ashby.
- `salary` (object|null): The pay the posting states, or null when it states none.
- `saved` (boolean): Whether the user saved this job.
- `title` (string): The posting's title as the employer states it.

### `set_job_feedback` (~118 tokens)

Skip or restore a job

Skips a job or restores a skipped job. Skipping cancels a swipe only if it has not yet been queued; it does not cancel an already queued application.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.
- `signal` (string, required): Whether the user is not interested in this job, or wants an earlier skip taken back.

Output parameters:

- `job_id` (string): The job the feedback was recorded for.
- `signal` (string): The signal stored: not_interested, or undo once the skip was taken back.
- `skipped` (boolean): True while the job is skipped.

### `list_blocked_companies` (~74 tokens)

List blocked companies

The employers the user excluded from matches, by name, with when each was excluded; a page at a time. Reads only.

Input parameters:

- `limit` (integer): How many companies to return.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.

Output parameters:

- `companies` (array): The page of excluded employers, by name.
- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `more` (boolean): True when later pages hold more employers.
- `offset` (integer): The offset this page starts at.
- `total` (integer): How many employers the user excluded.

### `set_company_blocked` (~92 tokens)

Hide or restore a company

Excludes one employer from the user's matches, or removes its exclusion, and returns the full list of excluded employers.

Input parameters:

- `blocked` (boolean, required): true excludes the company from matches, false restores it.
- `company` (string, required): The employer name as postings state it.
- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.

Output parameters:

- `blocked` (boolean): True when the employer is now excluded, false when it was restored.
- `changed` (boolean): True when this call changed the list, false when it already stood that way.
- `companies` (array): Every employer the user excludes now.
- `company` (string): The employer name the exclusion was recorded for.

### `get_job` (~50 tokens)

Read a job

The full posting, the match assessment against the user, and any application status. Reads only.

Input parameters:

- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `application_status` (string|null): The application's ledger status when the user applied to this job, else null.
- `apply_supported` (boolean): Whether ApplySent can fill and submit this posting's form by itself.
- `apply_url` (string): The employer's posting URL, where the application is filled.
- `city` (string|null): City the posting is in.
- `company` (string): The employer's name.
- `country` (string|null): Country the posting is in, as the catalogue recorded it.
- `description` (string): The posting's text with whitespace collapsed, cut to the tool's length.
- `description_truncated` (boolean): True when the posting's text was cut.
- `facts` (object): Facts the catalogue derived from the posting; a key is absent when the catalogue established nothing for it.
- `id` (string): The job id the other job tools take.
- `match` (object): The assessment of this job against the user's résumé, targets and eligibility.
- `posted_at` (string|null): When the posting went up, ISO 8601 in UTC, when the catalogue knows it.
- `provider` (string): The employer's hiring system the posting came from, e.g. workday, greenhouse or ashby.
- `salary` (object|null): The pay the posting states, or null when it states none.
- `saved` (boolean): Whether the user saved this job.
- `title` (string): The posting's title as the employer states it.

### `get_jobs` (~63 tokens)

Read jobs

Up to 20 postings by id with their match assessments, descriptions shortened; ids not found come back in missing. Reads only.

Input parameters:

- `job_ids` (array, required): Job ids from search_jobs, get_job_from_link or list_applications; a repeated id is read once.

Output parameters:

- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `jobs` (array): The requested postings that are in this account, in the order asked.
- `missing` (array): The requested ids this account does not hold.

### `get_job_from_link` (~98 tokens)

Read a job from its link

Reads the posting at an employer or job-board link and returns the full posting text, the match assessment against the user and the job id, keeping it in the user's saved jobs. Fetches the employer's page; applies to nothing.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `url` (string, required): The posting URL the user shared.

Output parameters:

- `application_status` (string|null): The application's ledger status when the user applied to this job, else null.
- `apply_supported` (boolean): Whether ApplySent can fill and submit this posting's form by itself.
- `apply_url` (string): The employer's posting URL, where the application is filled.
- `city` (string|null): City the posting is in.
- `company` (string): The employer's name.
- `country` (string|null): Country the posting is in, as the catalogue recorded it.
- `description` (string): The posting's text with whitespace collapsed, cut to the tool's length.
- `description_truncated` (boolean): True when the posting's text was cut.
- `facts` (object): Facts the catalogue derived from the posting; a key is absent when the catalogue established nothing for it.
- `id` (string): The job id the other job tools take.
- `match` (object): The assessment of this job against the user's résumé, targets and eligibility.
- `posted_at` (string|null): When the posting went up, ISO 8601 in UTC, when the catalogue knows it.
- `provider` (string): The employer's hiring system the posting came from, e.g. workday, greenhouse or ashby.
- `salary` (object|null): The pay the posting states, or null when it states none.
- `saved` (boolean): Whether the user saved this job.
- `title` (string): The posting's title as the employer states it.

### `search` (~65 tokens)

Search jobs

The same matching search as search_jobs, best match first, as citable results: each job's id, its title and employer, and the employer's posting URL; fetch reads one in full. Reads only.

Input parameters:

- `query` (string, required): Words in the title, employer or description.

Output parameters:

- `results` (array): The matching jobs as citable results, best match first.

### `fetch` (~45 tokens)

Fetch a job's text

The full text of one job from search, with its posting URL and match details, for citing it. Reads only.

Input parameters:

- `id` (string, required): A job id from a search result.

Output parameters:

- `id` (string): The job id.
- `metadata` (object): The posting's facts beside its text.
- `text` (string): The posting's full text.
- `title` (string): The posting's title and employer in one line.
- `url` (string): The employer's posting URL, for citing it.

### `list_applications` (~148 tokens)

List applications

The user's applications, newest first, with status and the employer's receipt when sent. status needs_you keeps those waiting on the user, uncertain those sent without a confirmed receipt (never sent again); started_by_this_client keeps those this assistant started. Reads only.

Input parameters:

- `limit` (integer): How many applications to return.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.
- `started_by_this_client` (boolean): Only applications this assistant started; leave it out for every application.
- `status` (string): Which applications to list: every one, those waiting on the user, or those sent without a confirmed receipt.

Output parameters:

- `applications` (array): The page of applications, newest first.
- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `more` (boolean): True when later pages hold more applications.
- `offset` (integer): The offset this page starts at.

### `get_application` (~73 tokens)

Read an application

One application: status and step (blocked with its reason and retry_at while ApplySent waits to retry), receipt, the answers sent, any open questions, suggested_next_tool and poll_after_seconds. Reads only.

Input parameters:

- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `answer_batches` (array): The open question batches this application waits on.
- `answers` (array): The answers this application sends.
- `apply_url` (string): The employer's posting URL.
- `company` (string|null): The employer's name, when the ledger has it.
- `hold` (object|null): What the employer waits on before this application can go on, when it waits on something ApplySent recognizes.
- `id` (string): The job id of this application.
- `missing_prerequisites` (array): What still stops this application from being sent, in words.
- `not_sent_reason` (string|null): Why a failed application was not sent, when ApplySent refused it.
- `poll_after_seconds` (integer|null): Seconds to wait before reading this application again; null when nothing will change by itself.
- `receipt` (object|null): The employer's confirmation of this application, or null when none was recorded.
- `recorded_at` (string|null): When the user started this application, ISO 8601.
- `retry_at` (string|null): When ApplySent retries a blocked application, ISO 8601.
- `review` (object): How far the review of this application has come.
- `review_reasons` (array): Why the application waits for the user's review.
- `stage` (string|null): The hiring stage the employer's messages established (applied, interview, offer, rejected), when known.
- `status` (string): The application's ledger status, e.g. queued, submitted, uncertain or failed.
- `step` (string): Where the application stands in ApplySent's pipeline, e.g. preparing_application, needs_answer or sent.
- `suggested_next_tool` (string): The tool that answers what this application needs next.
- `title` (string|null): The posting's title, when the ledger has it.
- `uncertain` (boolean): True when the send went out without a confirmed receipt; ApplySent never sends such a job again.
- `waiting` (string|null): Set while the background worker is on this application right now.

### `get_application_board` (~183 tokens)

Read the application board

The Applications board under its filters: how many applications stand in each bucket, how many are being worked on now, and light rows of one bucket's page or of the first rows of every bucket, work running now first. Paging with offset needs bucket. Reads only.

Input parameters:

- `before` (string): Only applications recorded before this ISO date.
- `bucket` (string): One bucket to page through; leave it out for the first rows of every bucket.
- `company` (string): Only applications at this employer, matched without case.
- `limit` (integer): Rows per bucket.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.
- `q` (string): Words to match in the job title or employer.
- `since` (string): Only applications recorded on or after this ISO date.

Output parameters:

- `bucket` (string|null): The column paged through, or null when every column's first rows came back.
- `counts` (object): How many applications stand in each board column, under the filters.
- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `filters` (object): The filters this page was read under.
- `offset` (integer): The offset this page starts at.
- `rows` (array): The light rows of one column's page, or the first rows of every column.
- `working` (integer): How many applications are being worked on right now.

### `get_metrics` (~64 tokens)

Read application metrics

Receipt-confirmed application metrics for a window of UTC days: confirmations, their daily and rolling 30-day counts, and the rolling interview rate; internal AI spend is left out. Reads only.

Input parameters:

- `days` (integer): How many UTC days back to count, ending today.

Output parameters:

- `confirmed_count` (integer): Applications confirmed by a receipt inside the window.
- `confirmed_per_day` (number): Confirmed applications per day across the window.
- `daily_confirmed` (object): Confirmed applications by UTC date inside the window.
- `days` (integer): How many UTC days back the window counts.
- `receipts_by_kind` (object): Confirmations by receipt kind.
- `receipts_by_provider` (object): Confirmations by hiring system.
- `rolling_30d_confirmed` (integer): Applications confirmed in the last 30 days.
- `rolling_30d_interview_rate` (number|null): Interviews per confirmed application over those 30 days, or null when nothing was confirmed.
- `rolling_30d_interviews` (integer): Applications that reached interview or offer in the last 30 days.
- `uncertain_count` (integer): Applications sent without a confirmed receipt, account-wide now.

### `get_application_review` (~142 tokens)

Read application review

Everything needed to review one application before it is sent: the packet revision and its approval, the documents (name, format, size, hash) and form answers it would send, open employer questions, every blocker in words, field errors and the public timeline. Use it when the user wants to check or fix an application that waits for review; use get_application for status alone and get_application_receipt for proof after sending. Pass job_id. Reads only: approving, answering and sending are done by answer_question, preview_application and apply_to_job.

Input parameters:

- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `answers` (array): The answers this application sends.
- `blockers` (array): Everything that stops this application from being sent, in words.
- `company` (string): The employer's name.
- `digest` (string|null): The current packet's hash, when one exists.
- `events` (array): The public timeline steps, oldest first.
- `field_errors` (object): Form fields whose answers the review rejected, by field name.
- `job_id` (string): The job id of this application.
- `packet` (object|null): The packet the application would send, or null when none was built.
- `questions` (array): The open question batches this application waits on.
- `review_available` (boolean): True when the application has a packet the user may review.
- `review_reasons` (array): Why the application waits for the user's review.
- `revision` (integer|null): The current packet revision, when one exists.
- `stage` (string|null): The hiring stage the employer's messages established, when known.
- `status` (string): The application's ledger status.
- `step` (string|null): Where the application stands in ApplySent's pipeline.
- `title` (string): The posting's title.
- `uncertain` (boolean): True when the send went out without a confirmed receipt.

### `get_application_receipt` (~67 tokens)

Show an application receipt

The employer's confirmation evidence for one application: the receipt, the answers sent, the proof files and the confirmation screenshot as an image when one was captured. Reads only.

Input parameters:

- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `answers_sent` (array): The answers this application sends.
- `company` (string): The employer's name.
- `job_id` (string): The job id of this application.
- `proof` (array): The proof files the send left behind.
- `receipt` (object|null): The employer's confirmation of this application, or null when none was recorded.
- `screenshot` (string): The name of the proof image also returned as an image part, when one was captured.
- `status` (string): The application's ledger status.
- `title` (string): The posting's title.
- `uncertain` (boolean): True when the send went out without a confirmed receipt.

### `list_pending_questions` (~100 tokens)

List open employer questions

Questions employers asked that only the user can answer; the applications asking them wait. Each is asked once and its answer is reused for every application asking it. Each group gives its batch_id, the application, its questions and how many applications ask each. Reads only.

Input parameters:

- `limit` (integer): How many question groups to return.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.

Output parameters:

- `applications_waiting` (integer): How many applications wait on those answers.
- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `groups` (array): The page of question groups, widest reach first.
- `total` (integer): How many question groups wait on the user.

### `get_account` (~34 tokens)

Read application usage

How many applications the account has left and has sent today (or unlimited), and anything stopping applications from being sent. Reads only.

Output parameters:

- `applications` (object): The allowance the sender counts against.
- `send_blockers` (array): Everything that stops applications from being sent now, in words.

### `get_profile_status` (~42 tokens)

Check profile readiness

What the user's ApplySent profile still needs before applications can go out, and how many employer questions wait. Reports what is missing, not the values. Reads only.

Output parameters:

- `missing` (array): What the profile still needs, in words.
- `pending_questions` (integer): How many employer questions wait on the user.
- `ready` (boolean): True when the profile has everything an application needs.
- `tailor_resume` (boolean): Whether the user asked for a résumé tailored to each job.
- `targets` (object): What the user targets, as the profile records it.

### `list_inbox` (~81 tokens)

List employer emails

Employer emails in the user's application inbox, newest first, classified (interview, rejection, verification, assessment...) and linked to the application they belong to. Reads only; nothing is sent.

Input parameters:

- `category` (string): Only emails the classifier put in this category; leave it out for every category.
- `limit` (integer): How many emails to return.

Output parameters:

- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `messages` (array): The page of employer emails, newest first.

### `get_message` (~40 tokens)

Read an employer email

One email from the application inbox with its text and the application it belongs to. Reads only.

Input parameters:

- `message_id` (required): A message id from list_inbox.

Output parameters:

- `application` (object|null): The application this email belongs to, or null when it belongs to none.
- `body` (string): The email's text, cut to the tool's length.
- `body_truncated` (boolean): True when the email's text was cut.
- `category` (string): What the email is, as classified: interview, rejection, assessment, offer, recruiter, verification, application_confirmation or unknown.
- `id` (integer): The message id get_message and get_thread take.
- `received_at` (string): When the email arrived, ISO 8601.
- `sender` (string): The sender's address.
- `subject` (string): The email's subject as received.

### `get_thread` (~77 tokens)

Read an email conversation

The conversation around one email: every email of the same application, or of the same sender and subject when it belongs to none, oldest first. Each message gives its id, sender, subject, received time, category and the application it belongs to, without bodies. Reads only.

Input parameters:

- `message_id` (required): A message id from list_inbox.

Output parameters:

- `messages` (array): Every email of the conversation, oldest first, without bodies.

### `get_notification_settings` (~31 tokens)

Read notification settings

The user's push categories and email and Telegram notification preferences, as the app and site show them. Reads only.

Output parameters:

- `notifications` (object): The email and Telegram notification preferences.
- `push` (object): The push categories, as the app and site show them.

### `update_notification_settings` (~93 tokens)

Change notification settings

Changes only the given notification preferences and returns the saved settings.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `notifications` (object): Email and Telegram preferences to change; leave any out to keep it as it is.
- `push` (object): Push categories to change; leave any out to keep it as it is.

Output parameters:

- `notifications` (object): The email and Telegram notification preferences.
- `push` (object): The push categories, as the app and site show them.

### `list_notifications` (~70 tokens)

List notifications

The user's notification feed, newest first, a page at a time, with coverage saying what was left out. Reads only.

Input parameters:

- `limit` (integer): How many notifications to return.
- `offset` (integer): Rows to skip for the next page: the previous page's coverage names the offset to pass.

Output parameters:

- `coverage` (object): What this page covers, so the assistant can say honestly what it read.
- `more` (boolean): True when later pages hold more notifications.
- `notifications` (array): The page of notifications, newest first.
- `offset` (integer): The offset this page starts at.

### `mark_notification_opened` (~66 tokens)

Mark a notification opened

Records the first opening of a notification. Repeating it changes nothing.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `notification_id` (integer, required): A notification id from list_notifications.

Output parameters:

- `id` (integer): The notification that was opened.
- `opened` (boolean): True once the opening is recorded; a repeat leaves it true.

### `answer_question` (~150 tokens)

Answer an employer question

Saves the given answer to one open question and applies it to every application asking it; those applications then continue. The answer is stored exactly as given, as the user's own words or the option they chose. Returns the applications answered and those now ready to continue. Sends nothing.

Input parameters:

- `answer` (string, required): The user's own answer, or the exact option they picked.
- `batch_id` (integer, required): The question group from list_pending_questions the question belongs to.
- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `question_index` (integer, required): Which question of that group to answer, as list_pending_questions numbers it.

Output parameters:

- `answered` (array): The applications the answer was applied to.
- `banked` (integer): How many reusable answers the answer was stored as.
- `profile_field` (string|null): The profile field the answer was saved to, when it is a fact the profile keeps.
- `ready` (array): The applications left with nothing to ask, now ready to go on.

### `preview_tailored_resume` (~104 tokens)

Preview a tailored résumé

The résumé tailored to one job, as the PDF an application would attach, with what changed against the saved résumé. Made once with an AI model from facts already in the résumé, then reused; nothing is sent.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `approved` (boolean): Whether this résumé is the one approved to send with the application.
- `changes` (object): What changed against the saved résumé.
- `company` (string): The employer's name.
- `job_id` (string): The job id the résumé is tailored to.
- `notes` (array): What the tailoring did and left out, as it reported it.
- `resume_sha256` (string): The tailored PDF's hash; approve_tailored_resume takes it.
- `title` (string): The posting's title.

### `approve_tailored_resume` (~105 tokens)

Approve a tailored résumé

Marks the previewed tailored résumé (by its resume_sha256) as the one to send with this job's application. Sends nothing.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.
- `resume_sha256` (string, required): resume_sha256 from preview_tailored_resume.

Output parameters:

- `approved` (boolean): True once the résumé is the one to send.
- `job_id` (string): The job the résumé is approved for.
- `resume_sha256` (string): The approved résumé's hash.

### `regenerate_tailored_resume` (~106 tokens)

Regenerate tailored résumé

Discards the job's tailored résumé and its approval, then tailors it again with an AI model from the saved résumé, and returns the new résumé with what changed. Refused once the application was sent or is sending.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `approved` (boolean): Whether this résumé is the one approved to send with the application.
- `changes` (object): What changed against the saved résumé.
- `company` (string): The employer's name.
- `job_id` (string): The job id the résumé is tailored to.
- `notes` (array): What the tailoring did and left out, as it reported it.
- `resume_sha256` (string): The tailored PDF's hash; approve_tailored_resume takes it.
- `title` (string): The posting's title.

### `preview_application` (~91 tokens)

Preview an application

Shows what applying to one job would send and use, and returns a single-use confirmation_token for apply_to_job. The user must review this preview and say yes in the conversation before apply_to_job; the token works for this job and this connection only and expires after 10 minutes. Sends nothing.

Input parameters:

- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `confirmation_token` (string): The single-use token apply_to_job takes for this job.
- `expires_in` (integer): How many seconds the token stays valid.
- `job` (object): The job the application would go to.
- `match` (object): The assessment of the job against the user.
- `resume` (string): Which résumé the application would attach, in words.
- `sends` (string): What ApplySent would fill and submit, in words.
- `uses_applications` (integer): How many applications this send draws from the allowance.

### `queue_application_for_review` (~107 tokens)

Queue an application for review

Prepares one application for the user to review and approve in ApplySent; it is sent only after the user approves it there. Sends nothing. Refuses if a send is already in progress or the posting has already been sent.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `answer_batches` (array): The open question batches this application waits on.
- `answers` (array): The answers this application sends.
- `apply_url` (string): The employer's posting URL.
- `company` (string|null): The employer's name, when the ledger has it.
- `hold` (object|null): What the employer waits on before this application can go on, when it waits on something ApplySent recognizes.
- `id` (string): The job id of this application.
- `missing_prerequisites` (array): What still stops this application from being sent, in words.
- `not_sent_reason` (string|null): Why a failed application was not sent, when ApplySent refused it.
- `poll_after_seconds` (integer|null): Seconds to wait before reading this application again; null when nothing will change by itself.
- `receipt` (object|null): The employer's confirmation of this application, or null when none was recorded.
- `recorded_at` (string|null): When the user started this application, ISO 8601.
- `retry_at` (string|null): When ApplySent retries a blocked application, ISO 8601.
- `review` (object): How far the review of this application has come.
- `review_reasons` (array): Why the application waits for the user's review.
- `stage` (string|null): The hiring stage the employer's messages established (applied, interview, offer, rejected), when known.
- `status` (string): The application's ledger status, e.g. queued, submitted, uncertain or failed.
- `step` (string): Where the application stands in ApplySent's pipeline, e.g. preparing_application, needs_answer or sent.
- `suggested_next_tool` (string): The tool that answers what this application needs next.
- `title` (string|null): The posting's title, when the ledger has it.
- `uncertain` (boolean): True when the send went out without a confirmed receipt; ApplySent never sends such a job again.
- `waiting` (string|null): Set while the background worker is on this application right now.

### `apply_to_job` (~174 tokens)

Apply to a job

Sends one real application to one job on the employer's site, only after the user said yes to this job in the conversation, and uses one application from the account's allowance. Requires the confirmation_token preview_application returned for this job. A job already applied to, or sent with an uncertain outcome, is never sent again. A posting held on an employer account that is connected now is woken and reported as applying (status queued, no hold, poll suggested) instead of returning the old hold.

Input parameters:

- `confirmation_token` (string, required): From preview_application for this job.
- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `answer_batches` (array): The open question batches this application waits on.
- `answers` (array): The answers this application sends.
- `apply_url` (string): The employer's posting URL.
- `company` (string|null): The employer's name, when the ledger has it.
- `hold` (object|null): What the employer waits on before this application can go on, when it waits on something ApplySent recognizes.
- `id` (string): The job id of this application.
- `missing_prerequisites` (array): What still stops this application from being sent, in words.
- `not_sent_reason` (string|null): Why a failed application was not sent, when ApplySent refused it.
- `poll_after_seconds` (integer|null): Seconds to wait before reading this application again; null when nothing will change by itself.
- `receipt` (object|null): The employer's confirmation of this application, or null when none was recorded.
- `recorded_at` (string|null): When the user started this application, ISO 8601.
- `retry_at` (string|null): When ApplySent retries a blocked application, ISO 8601.
- `review` (object): How far the review of this application has come.
- `review_reasons` (array): Why the application waits for the user's review.
- `stage` (string|null): The hiring stage the employer's messages established (applied, interview, offer, rejected), when known.
- `status` (string): The application's ledger status, e.g. queued, submitted, uncertain or failed.
- `step` (string): Where the application stands in ApplySent's pipeline, e.g. preparing_application, needs_answer or sent.
- `suggested_next_tool` (string): The tool that answers what this application needs next.
- `title` (string|null): The posting's title, when the ledger has it.
- `uncertain` (boolean): True when the send went out without a confirmed receipt; ApplySent never sends such a job again.
- `waiting` (string|null): Set while the background worker is on this application right now.

### `cancel_application` (~87 tokens)

Cancel an application

Stops an application whose send has not started; it can then never be sent and its open questions are withdrawn. Refused once sending started.

Input parameters:

- `idempotency_key` (string): Repeating a call with the same key and arguments returns its first result instead of acting again.
- `job_id` (string, required): A job id from search_jobs, get_job_from_link or list_applications.

Output parameters:

- `answer_batches` (array): The open question batches this application waits on.
- `answers` (array): The answers this application sends.
- `apply_url` (string): The employer's posting URL.
- `company` (string|null): The employer's name, when the ledger has it.
- `hold` (object|null): What the employer waits on before this application can go on, when it waits on something ApplySent recognizes.
- `id` (string): The job id of this application.
- `missing_prerequisites` (array): What still stops this application from being sent, in words.
- `not_sent_reason` (string|null): Why a failed application was not sent, when ApplySent refused it.
- `poll_after_seconds` (integer|null): Seconds to wait before reading this application again; null when nothing will change by itself.
- `receipt` (object|null): The employer's confirmation of this application, or null when none was recorded.
- `recorded_at` (string|null): When the user started this application, ISO 8601.
- `retry_at` (string|null): When ApplySent retries a blocked application, ISO 8601.
- `review` (object): How far the review of this application has come.
- `review_reasons` (array): Why the application waits for the user's review.
- `stage` (string|null): The hiring stage the employer's messages established (applied, interview, offer, rejected), when known.
- `status` (string): The application's ledger status, e.g. queued, submitted, uncertain or failed.
- `step` (string): Where the application stands in ApplySent's pipeline, e.g. preparing_application, needs_answer or sent.
- `suggested_next_tool` (string): The tool that answers what this application needs next.
- `title` (string|null): The posting's title, when the ledger has it.
- `uncertain` (boolean): True when the send went out without a confirmed receipt; ApplySent never sends such a job again.
- `waiting` (string|null): Set while the background worker is on this application right now.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/com-applysent-applysent/api-mcp#diagnostics

## Score history

- 2026-11-21: 83
- 2026-11-20: 82
- 2026-11-19: 82
- 2026-11-18: 81
- 2026-11-17: 81
- 2026-11-16: 81
- 2026-11-15: 81
- 2026-11-14: 80
- 2026-11-13: 79
- 2026-11-12: 79
- 2026-11-11: 39

## Common questions

### What is the ApplySent MCP server?

ApplySent is an MCP server listed in the public MCP registry as com.applysent/applysent. Search jobs, preview each application and apply with your OK on Workday, Greenhouse, Ashby and more. This page covers its hosted endpoint (https://applysent.com/api/mcp).

### Is the ApplySent MCP server safe to use?

ApplySent scores 83 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the ApplySent MCP server expose?

ApplySent exposes 35 tools: search_jobs, list_saved_jobs, set_job_saved, set_job_feedback, list_blocked_companies, and 30 more. Their descriptions and schemas cost roughly 3,457 tokens of context every time the server is loaded.

### Does the ApplySent MCP server require authentication?

Yes. ApplySent asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the ApplySent MCP server still maintained?

ApplySent is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the ApplySent MCP server?

VerifyMCP ran 13 checks on ApplySent, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://applysent.com/api/mcp
- Website: https://applysent.com/
- Changelog RSS feed: https://verifymcp.io/servers/com-applysent-applysent/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-applysent-applysent/api-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/com-applysent-applysent/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/com-applysent-applysent/api-mcp
