# com.keywordise/keywordise (remote · keywordise.com)

Auto-apply to jobs: matches your CV, tailors a fresh CV per posting, and applies for you.

- Trust score: 70/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-28

## Components

- remote · `keywordise.com`: 70/100 (this document), [markdown](https://verifymcp.io/servers/com-keywordise-keywordise/keywordise.md), [page](https://verifymcp.io/servers/com-keywordise-keywordise/keywordise)

## Channel facts

- Endpoint: `https://keywordise.com/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `2.0.0`

## 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 2026-09-28.

- **Endpoint Security**: 46/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (cancel_subscription).
  - HTTPS enforcement could not be verified: the plaintext port answered with HTTP 405, which proves neither a plaintext path nor enforcement.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 79/100
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 3574 tokens (~72/item across 49 items; 49 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 90/100
  - Stability observed for 27 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 80/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 39% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 4 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 50 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 60/100
  - Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28.

## Install

### How do I install the com.keywordise/keywordise MCP server?

com.keywordise/keywordise is a hosted endpoint at https://keywordise.com/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-keywordise-keywordise 'https://keywordise.com/mcp'
```

### Cursor

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

### VS Code

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

### Codex

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

### opencode

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

### OpenClaw

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

### Hermes

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

### Netclaw

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

### Vellum

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

### Other

```json
{
  "mcpServers": {
    "com-keywordise-keywordise": {
      "type": "http",
      "url": "https://keywordise.com/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.

### 2026-09-28 (score 70, 0)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-27 (score 70, +1)

No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-26 (score 69, 0)

- [security] Tool “describe_what_i_want” rewrote its description, which is the text the model reads
- [security] Tool “explain_job_match” rewrote its description, which is the text the model reads
- [security] Tool “get_match_status” rewrote its description, which is the text the model reads
- [security] Tool “get_my_profile” rewrote its description, which is the text the model reads
- [security] Tool “list_matching_jobs” rewrote its description, which is the text the model reads
- [security] Tool “list_my_applications” rewrote its description, which is the text the model reads
- [security] Tool “preview_tailored_cv” rewrote its description, which is the text the model reads
- [security] Tool “refresh_matches” rewrote its description, which is the text the model reads
- [security] Tool “start_apply_run” rewrote its description, which is the text the model reads
- [security] Tool “stop_apply_run” rewrote its description, which is the text the model reads
- [security] Tool “upload_cv” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 65 → 72
- [cosmetic] “refresh_matches” added an optional parameter “rebuild”

### 2026-09-25 (score 69, +1)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-23 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-21 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-18 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 53 to 57. That category is still filling its 30-day observation window: 16 days of observed history at the previous scan, 17 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-16 (score 65, +1)

No change was recorded against any check on this day. Stability & Change Management went from 47 to 50. That category is still filling its 30-day observation window: 14 days of observed history at the previous scan, 15 at this one. The score rises as the window fills, whether or not the server changes.

## MCP tools (49)

### `create_account` (~111 tokens)

Create a Keywordise account

Register a new Keywordise account from the terminal and receive an API key for it. Requires accept_terms=true: the person accepts the Terms and Privacy Policy (https://keywordise.com/terms) and authorises Keywordise to apply for jobs on their behalf. A confirmation email is sent; applying opens once the address is confirmed. Trial: 7 days and 27 applications, then $49/month.

Input parameters:

- `accept_terms` (boolean, required)
- `email` (string, required)
- `password` (string, required)

### `login` (~94 tokens)

Log in and get an API key

Exchange the account's email and password for a new API key, so a terminal or an assistant can act on the account. The key is shown once; keep it. Up to five keys may be active per account (revoke_api_key frees a slot).

Input parameters:

- `email` (string, required)
- `label` (string): A name for this key, e.g. the machine it lives on.
- `password` (string, required)

### `get_account` (~43 tokens)

My account and plan

The account behind this key: email and whether it is confirmed, plan state (trial days left, applications used and remaining, subscription), and how many API keys are active.

### `resend_verification_email` (~36 tokens)

Resend the confirmation email

Send the email-confirmation link again to this account's address. Applying is blocked until the address is confirmed; nothing else is.

### `get_upgrade_link` (~51 tokens)

Upgrade to Pro (checkout link)

A checkout link for the Pro plan ($49/month, 200 applications a month) from the active payment provider. The person opens it in a browser to pay; the account upgrades automatically when the payment lands.

### `get_billing_portal_link` (~37 tokens)

Manage my subscription (portal link)

A link to the payment provider's customer portal for this account's subscription, where invoices, the payment method and cancellation live.

### `cancel_subscription` (~40 tokens)

Cancel my subscription at period end

Cancel the Pro subscription at the end of the current billing period. Access stays until then. Irreversible from here; the person can resubscribe later.

### `list_api_keys` (~31 tokens)

My API keys

The active API keys on this account: prefix, label, created and last used. Never the key itself.

### `create_api_key` (~40 tokens)

Mint another API key

Mint an additional API key for this account, for another machine or assistant. Shown once. Five active keys maximum.

Input parameters:

- `label` (string)

### `revoke_api_key` (~55 tokens)

Revoke an API key

Revoke one of this account's API keys immediately. Any terminal or assistant using it stops working at once. Revoking the key you are calling with ends this session.

Input parameters:

- `key_id` (string, required): From list_api_keys.

### `export_my_data` (~54 tokens)

Export everything Keywordise holds about me

The account's complete data as one JSON file (profile, targeting, matched jobs, applications, saved answers, run history, audit trail). Returned as a file; large accounts are told to download it in the app instead.

### `delete_my_account` (~53 tokens)

Delete my account permanently

Erase the account and everything in it: profile, CV files, matches, applications, answers, keys. Irreversible. Requires confirm_email equal to the account's address.

Input parameters:

- `confirm_email` (string, required)

### `get_my_profile` (~84 tokens)

Who am I acting for

Use when the person asks what is on their profile or CV. Not needed before searching or applying: those tools read the saved CV themselves. The CV profile Keywordise holds for this account (name, contact, summary, skills, roles, education), the job-targeting rules (countries, remote preference, keywords) and the plan status (applications remaining). Call this first.

### `upload_cv` (~140 tokens)

Upload my CV (PDF or DOCX)

Use when the person shares or mentions their resume/CV ('here's my resume', 'use my CV'). Upload a CV file and turn it into the account's profile: the same parser the app uses reads the document into name, contact, summary, roles, skills, languages and education. With save=true (default) the profile is saved and the job feed is built from it; with save=false the parsed profile is returned for review only. Max 8 MB.

Input parameters:

- `content_base64` (string, required): The file's bytes, base64-encoded.
- `filename` (string, required): e.g. cv.pdf
- `save` (boolean)

### `update_profile` (~96 tokens)

Edit my profile

Change fields of the saved profile without re-uploading the CV. Send only the fields to change (full_name, contact_email, phone, location, city, address, postcode, professional_summary, skills, pinned_skills, languages, experience, education, projects, certifications, links, …); everything else is untouched. The feed is re-ranked against the new CV.

Input parameters:

- `patch` (object, required): Profile fields to change, by name.

### `get_profile_gaps` (~49 tokens)

What my CV is missing

The gaps in the saved profile that weaken applications (missing dates on a role, no contact phone, empty summary, …), each with a severity, so they can be fixed with update_profile.

### `suggest_keywords` (~25 tokens)

Keywords my CV suggests

Search keywords derived from the saved profile, useful as include_keywords in update_targeting.

### `list_cv_designs` (~28 tokens)

Available CV designs

The CV designs a tailored CV can be rendered in, with the account's current pick.

### `set_cv_design` (~34 tokens)

Choose my CV design

Set the design every tailored CV is rendered in (a key from list_cv_designs).

Input parameters:

- `design` (string, required)

### `get_my_cv_pdf` (~56 tokens)

Render my CV as a PDF

The account's base CV (the saved profile, untailored) rendered as a PDF in the chosen design, returned as a file. No model call.

Input parameters:

- `design` (string): Optional; defaults to the saved design.

### `get_targeting` (~70 tokens)

My job targeting

Every targeting rule the feed and the apply run obey: remote mode, geo scope and countries, locations, include/exclude keywords, title allow/block lists, seniority, salary, languages, work authorisation, company allow/block, daily and per-company caps, cover-letter and CV settings, screening defaults.

### `update_targeting` (~134 tokens)

Change my job targeting

Change targeting rules; send only the keys to change (remote_mode any|remote|hybrid|onsite, geo_scope worldwide|region|country, geo_countries [ISO-2], locations, include_keywords, exclude_keywords, title_allow, title_block, seniority, salary_min, currency, languages, company_block, max_apps_per_day, max_apps_per_company, cover_letter_enabled, cv_tone, tailor_instructions, years_experience, salary_expectation, notice_period, …). The feed re-matches in the background.

Input parameters:

- `patch` (object, required): Targeting fields to change, by name.

### `describe_what_i_want` (~121 tokens)

Set targeting from one sentence

Use when the person says what they are looking for in their own words ('senior remote data roles in the US, no agencies'). Turn a plain sentence ('remote data roles in Germany or the Netherlands, senior, not agencies') plus the saved CV into concrete targeting rules. Returns the inferred rules and the chips a person would confirm; with apply=true they are saved and the feed re-matches. preset: balanced | broad | focused.

Input parameters:

- `apply` (boolean)
- `preset` (string)
- `sentence` (string, required)

### `get_feed_reach` (~53 tokens)

What my targeting is costing me

How many jobs the current targeting shows versus how many each single rule is hiding, and which one change would widen the feed most. Every number comes from re-asking the feed's own predicate with one control relaxed.

### `list_countries_with_jobs` (~40 tokens)

Countries with live jobs

The countries the live pool actually has inventory in (ISO-2, name, job count), for choosing geo_countries in update_targeting.

### `list_matching_jobs` (~207 tokens)

Jobs that match my CV right now

Use when the person wants jobs or their best matches: 'find me jobs', 'my top matches', 'remote marketing jobs that fit my resume', 'what's new for me this week'. It already uses the saved CV, so there is no need to read the profile first. Live vacancies matched to this account's CV, ranked. Each job carries a calibrated match percentage, a recruiter-pass verdict with a one-line reason, and whether an application was already sent. Filter by match tier and freshness; page with offset.

Input parameters:

- `applied` (string): hide = leave out jobs already applied to.
- `fresh` (string): new = posted in the last two weeks; all = the last month.
- `limit` (integer)
- `match` (string): Match tier. 'all' includes weak matches the recruiter pass rejected.
- `offset` (integer)
- `sort` (string): fit = best match first; fresh = newest day first, best match within it.

### `get_job_posting` (~56 tokens)

Read one job posting

The full description of one matched vacancy, with a short summary and highlights, plus the link that opens the posting from the app.

Input parameters:

- `job_id` (string, required): A job_id from list_matching_jobs or list_my_applications.

### `explain_job_match` (~108 tokens)

Why this job does or does not fit me

Use for 'is this job a good fit', 'why this match', 'what am I missing for this role'. For one vacancy: the match percentage, the requirements this account's CV covers and the ones it is missing. With narrative=true, also a short written assessment with strengths and gaps (a paid model call; limited to 60 per hour).

Input parameters:

- `job_id` (string, required): A job_id from list_matching_jobs or list_my_applications.
- `narrative` (boolean)

### `preview_tailored_cv` (~121 tokens)

Tailor my CV for one job (preview only)

Use for 'tailor my resume for this job' or 'what would you send for this one'. Rewrite this account's CV for one vacancy the way Keywordise would send it: headline, summary, the roles selected for this job with rewritten bullets, curated skills and a cover letter, every segment labelled tailored, verbatim or pinned. Nothing is sent. A paid model call, limited to 20 per hour; the same billing gate as applying.

Input parameters:

- `job_id` (string, required): A job_id from list_matching_jobs or list_my_applications.

### `rate_job` (~76 tokens)

Hide, dislike or mark a job interesting

Feedback on one job: hide or dislike removes it from the feed and teaches the matcher what this person rejects; interested is a positive signal. Undo with unhide_job.

Input parameters:

- `action` (string, required)
- `job_id` (string, required): A job_id from list_matching_jobs or list_my_applications.
- `reason` (string)

### `unhide_job` (~47 tokens)

Put a hidden job back

Undo hide or dislike: the job returns to the feed and the negative signal is dropped.

Input parameters:

- `job_id` (string, required): A job_id from list_matching_jobs or list_my_applications.

### `list_hidden_jobs` (~27 tokens)

Jobs I hid

The jobs this account hid or disliked, newest first, so one can be put back.

### `refresh_matches` (~124 tokens)

Refresh my matches

Match the account's CV against the vacancies added since it last looked, in the background. Idempotent and cheap when nothing is new. force=true runs a full re-match regardless. rebuild=true rebuilds the feed from scratch through the same build as the app's Rebuild button (use it when the feed looks wrong). Every answer carries `feed_build`: poll get_match_status until its state is `ready` before reading list_matching_jobs, because a feed that is still building is not an empty feed.

Input parameters:

- `force` (boolean)
- `rebuild` (boolean)

### `get_match_status` (~113 tokens)

Is a match run in progress

Only for checking on a match or feed rebuild that is already running (after upload_cv, update_targeting or refresh_matches); to find or show jobs use list_matching_jobs. Live status of the latest match run (scanned, inserted, elapsed), the feed counts it produced, and `feed_build`: where the feed is in its build (queued, building, ready, failed). Poll after upload_cv, update_targeting or refresh_matches; the feed is final only when feed_build.feed_state.state is `ready`.

### `start_apply_run` (~245 tokens)

Apply to my matching jobs

Use when the person says 'apply for me', 'apply to the best matches' or 'apply to this job' (pass job_id). Show them the jobs first. Start an application run: Keywordise takes up to `limit` of the matching jobs (same filters as list_matching_jobs, best fit first, never one already applied to), tailors the CV and cover letter for each, answers the employer's screening questions from the saved answers, and SUBMITS inside the employer's own system. Real applications leave in the person's name; needs a confirmed email and an active plan; counts against the plan. dry_run=true does everything except submit. Poll get_apply_run_status.

Input parameters:

- `dry_run` (boolean)
- `fresh` (string): Freshness window; defaults to the account's saved filter.
- `job_id` (string): Apply to exactly this ONE job (an id from list_matching_jobs) instead of a filtered batch — the same run, scoped to it. limit/match/fresh are ignored when set.
- `limit` (integer)
- `match` (string): Match tier to draw from; defaults to the account's saved filter.

### `get_apply_run_status` (~50 tokens)

How my application run is going

Live progress of the latest run: running or not (from a heartbeat, not a status claim), processed of total, the job being worked on now, confirmed, needs review, skipped, closed.

### `stop_apply_run` (~49 tokens)

Stop my application run

Use for 'stop applying', 'pause', 'hold off'. Ask the running application run to stop after the job it is on. Nothing is left half-done; applications already sent stay sent.

### `list_my_applications` (~102 tokens)

Applications Keywordise sent for me

Use for 'what did I apply to', 'any replies', 'status of my applications'. Every application sent for this account: company, title, stage, outcome, when it went out, how many screening questions it still waits on, and the employer's acknowledgement where one was received.

Input parameters:

- `include_cover_letter` (boolean)
- `limit` (integer)
- `only_waiting_on_me` (boolean): Only applications stalled on unanswered screening questions.

### `list_blocking_questions` (~61 tokens)

Screening questions waiting on me

Questions employers asked on applications that are stalled until this account answers them, grouped by application, each with a suggested answer drawn from what the person answered before. Answer with answer_screening_question.

Input parameters:

- `limit` (integer): Maximum number of applications to expand.

### `answer_screening_question` (~95 tokens)

Answer an employer's screening question

Record the person's answer to one screening question on one application. The answer is written in the person's name and saved to their answer book, so the same question is never asked again. When it is the last required answer, the application is SUBMITTED to the employer. Only give an answer the person actually stated.

Input parameters:

- `answer` (string, required)
- `application_id` (string, required)
- `question_id` (string, required)

### `get_sent_cv_pdf` (~36 tokens)

The exact CV that was sent

The tailored CV PDF that went to the employer for one application, returned as a file.

Input parameters:

- `application_id` (string, required)

### `list_run_history` (~27 tokens)

My recent match and application runs

The last five match runs and the last five application runs, with their counts and status.

### `list_saved_answers` (~35 tokens)

My saved screening answers

The answer book: every screening answer this account has given or taught, which future applications reuse automatically in any phrasing or language.

### `save_answer` (~89 tokens)

Teach an answer once, for every application

Add or replace a screening answer in the answer book ('What is your notice period?' → '1 month'). Applications already waiting on that fact are completed and sent in the background. Only save what the person actually stated.

Input parameters:

- `answer` (string, required)
- `kind` (string)
- `options` (array): For select questions: the choices the form offers.
- `question` (string, required)

### `delete_saved_answer` (~33 tokens)

Forget a saved answer

Remove one answer from the answer book (an answer_id from list_saved_answers).

Input parameters:

- `answer_id` (string, required)

### `list_facts_to_provide` (~52 tokens)

Facts that would unblock the most applications

The universal screening facts this account has not answered yet (right to work, notice period, salary expectation, …), ranked by how many waiting applications each one unblocks. Answer them with save_answer.

### `get_tracker` (~36 tokens)

My application tracker

The tracker board: applications grouped by the stage the person set (saved, applied, interview, offer, rejected) with their own notes.

### `move_tracker_card` (~42 tokens)

Move an application to another stage

Set the tracker stage of one application: saved, applied, interview, offer or rejected.

Input parameters:

- `application_id` (string, required)
- `stage` (string, required)

### `set_tracker_note` (~43 tokens)

Write a note on an application

Replace the person's own note on one tracked application (interview date, contact, next step).

Input parameters:

- `application_id` (string, required)
- `notes` (string, required)

## Diagnostics

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

## Score history

- 2026-09-28: 70
- 2026-09-27: 70
- 2026-09-26: 69
- 2026-09-25: 69
- 2026-09-24: 68
- 2026-09-23: 68
- 2026-09-22: 67
- 2026-09-21: 67
- 2026-09-20: 66
- 2026-09-19: 66
- 2026-09-18: 66
- 2026-09-17: 65
- 2026-09-16: 65
- 2026-09-15: 64
- 2026-09-14: 64
- 2026-09-13: 63
- 2026-09-12: 63
- 2026-09-11: 62
- 2026-09-10: 62
- 2026-09-09: 61
- 2026-09-08: 61
- 2026-09-07: 60
- 2026-09-06: 60
- 2026-09-05: 59
- 2026-09-04: 59
- 2026-09-03: 59
- 2026-09-02: 58
- 2026-09-01: 58

## Common questions

### What is the com.keywordise/keywordise MCP server?

com.keywordise/keywordise is an MCP server listed in the public MCP registry as com.keywordise/keywordise. Auto-apply to jobs: matches your CV, tailors a fresh CV per posting, and applies for you. This page covers its hosted endpoint (https://keywordise.com/mcp).

### Is the com.keywordise/keywordise MCP server safe to use?

com.keywordise/keywordise scores 70 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 com.keywordise/keywordise MCP server expose?

com.keywordise/keywordise exposes 49 tools: create_account, login, get_account, resend_verification_email, get_upgrade_link, and 44 more. Their descriptions and schemas cost roughly 3,449 tokens of context every time the server is loaded.

### Does the com.keywordise/keywordise MCP server require authentication?

No. We connected to com.keywordise/keywordise without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the com.keywordise/keywordise MCP server still maintained?

com.keywordise/keywordise is still listed as active in the MCP registry. We last reached this channel on 28 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://keywordise.com/mcp
- Website: https://keywordise.com/
- Changelog RSS feed: https://verifymcp.io/servers/com-keywordise-keywordise/keywordise.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-keywordise-keywordise/keywordise.json
- HTML version of this page: https://verifymcp.io/servers/com-keywordise-keywordise/keywordise