Refine Job Search
REMOTE · MCP.GETREFINE.CO
Job search management: applications, interviews, offers, recruiter email, and company openings.
Available components
Last scanned (11 hours ago)
Scores higher than 91% of remote servers
Reachable. Last reached 5 Nov 2026.
Category Project Management
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. How we score → Why this is hard to score →
Endpoint Security81
- The endpoint's TLS certificate is valid, in date, and uses a strong key. View diagnostics → Pass
- 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. View diagnostics → Pass
- HTTPS enforcement could not be verified: the plaintext port answered with HTTP 401, which proves neither a plaintext path nor enforcement. View diagnostics → Unverified
- HSTS check failed: the Strict-Transport-Security header is absent. See how to fix → View diagnostics → Fail
- DNSSEC check failed: this domain isn't protected by DNSSEC. See how to fix → View diagnostics → Fail
- The authorisation server supports Client ID Metadata Documents, the current MCP client-registration mechanism. View diagnostics → Pass
Transport & Reachability100
- Verified streamable-http transport via a live MCP handshake. View diagnostics → Pass
Schema Quality & AI Usability85
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (good).Pass
- Tool/resource definitions use about 4811 tokens (~72/item across 66 items; 62 tools + 4 resources), lean.Pass
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management67
- Stability check failed: schema churn in the 30 days we've observed: 18 tool removals, 0 breaking changes, 0 auth/transport breaks, 25 additions. See how to fix → Fail
Tool Coverage96
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 86% of tool parameters carry a description.Partial
- Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- All 5 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
- An AI judge read all 63 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Checks we ran (12)
- TLS certificate
- Authorization
- HSTS header
- DNSSEC
- Transport
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
Checks we could not complete (1)
These are gaps on our side, not findings about this hosted endpoint. We only credit what we can confirm, so a category made up entirely of them scores 0.
- HTTPS in Endpoint SecurityHTTPS enforcement could not be verified: the plaintext port answered with HTTP 401, which proves neither a plaintext path nor enforcement.
Does not apply to this hosted endpoint (2)
- Supply Chain SecurityScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
- Provenance & TransparencyScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
The score does not measure everything. What the score does not measure →
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 22 Oct 26 −5
- Stability: pass → fail ▼ security
- Tool “signals.list” was removed ▼ security
- Tool “signals.list_applied_jobs” was removed ▼ security
- Tool “signals.openings.list” was removed ▼ security
- Tool “signals.openings.mark_all_read” was removed ▼ security
- Tool “signals.openings.mark_read” was removed ▼ security
- Tool “signals.openings.save” was removed ▼ security
- Tool “signals.pause” was removed ▼ security
- Tool “signals.recommendations.block” was removed ▼ security
- Tool “signals.recommendations.dismiss” was removed ▼ security
- Tool “signals.recommendations.feedback” was removed ▼ security
- Tool “signals.recommendations.get” was removed ▼ security
- Tool “signals.recommendations.list” was removed ▼ security
- Tool “signals.recommendations.restore” was removed ▼ security
- Tool “signals.recommendations.watch” was removed ▼ security
- Tool “signals.resume” was removed ▼ security
- Tool “signals.track” was removed ▼ security
- Tool “signals.untrack” was removed ▼ security
- Tool “signals.update” was removed ▼ security
- New tool “following.unfollow”, which the server declares destructive security
- Tool coverage: 100% → 86% ▼ functional
- Schema quality: 4003 → 4811 ▼ functional
- Schema quality: excellent → good functional
- Server version: 0.5.2 → 0.6.0 functional
- New tool “companies.submissions.create” functional
- New tool “companies.submissions.get” functional
- New tool “following.digest.get” functional
- New tool “following.digest.update” functional
- New tool “following.follow” functional
- New tool “following.list” functional
- New tool “following.openings.list” functional
- New tool “following.openings.mark_all_read” functional
- New tool “following.openings.mark_read” functional
- New tool “following.pause” functional
- New tool “following.resume” functional
- New tool “following.update” functional
- New tool “openings.add_to_board” functional
- New tool “openings.get” functional
- New tool “recommendations.block” functional
- New tool “recommendations.dismiss” functional
- New tool “recommendations.feedback” functional
- New tool “recommendations.get” functional
- New tool “recommendations.list” functional
- New tool “recommendations.restore” functional
- New tool “search.company_openings” functional
- New tool “search.create” functional
- New tool “search.get” functional
- New tool “search.update” functional
- 28 Sept 26 0
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
- 25 Sept 26 0
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
- 27 Aug 26 0
- Server version: 0.5.1 → 0.5.2 functional
- 26 Aug 26 0
- Stability: 0.07 → fail ▼ security
- Tool “interviews.add” was removed ▼ security
- Tool “interviews.delete” was removed ▼ security
- Tool “interviews.list” was removed ▼ security
- Tool “interviews.update” was removed ▼ security
- Tool “persona.remove_trait” was removed ▼ security
- Tool “recommendations.block” was removed ▼ security
- Tool “inbox.accept” was removed ▼ security
- Tool “inbox.status” was removed ▼ security
- Tool “jobs.update_notes” was removed ▼ security
- Tool “journey.update” was removed ▼ security
- Tool “notifications.list” was removed ▼ security
- Tool “notifications.read” was removed ▼ security
- Tool “notifications.read_all” was removed ▼ security
- Tool “offers.accept” was removed ▼ security
- Tool “offers.add_negotiation” was removed ▼ security
- Tool “offers.create” was removed ▼ security
- Tool “offers.decline” was removed ▼ security
- Tool “offers.delete” was removed ▼ security
- Tool “offers.delete_negotiation” was removed ▼ security
- Tool “offers.get” was removed ▼ security
- Tool “offers.update” was removed ▼ security
- Tool “persona.confirm_trait” was removed ▼ security
- Tool “persona.correct_trait” was removed ▼ security
- Tool “recommendations.dismiss” was removed ▼ security
- Tool “recommendations.feedback” was removed ▼ security
- Tool “recommendations.get” was removed ▼ security
- Tool “recommendations.list” was removed ▼ security
- Tool “recommendations.restore” was removed ▼ security
- Tool “recommendations.watch” was removed ▼ security
- Tool “signals.save_opening” was removed ▼ security
- First check of Injection markers: pass security
- First check of Judged manipulation: pass security
- New tool “inbox.apply_suggestion”, which the server declares destructive security
- New tool “jobs.interviews.delete”, which the server declares destructive security
- New tool “jobs.offer.accept”, which the server declares destructive security
- New tool “jobs.offer.decline”, which the server declares destructive security
- New tool “jobs.offer.delete”, which the server declares destructive security
- New tool “jobs.offer.delete_negotiation”, which the server declares destructive security
- Tool “inbox.acknowledge” rewrote its description, which is the text the model reads security
- Tool “inbox.undo” rewrote its description, which is the text the model reads security
- Tool “integrations.connect_google” rewrote its description, which is the text the model reads security
- Tool “jobs.archive” rewrote its description, which is the text the model reads security
- Tool “jobs.delete” rewrote its description, which is the text the model reads security
- Tool “jobs.refresh” rewrote its description, which is the text the model reads security
- Tool “jobs.update” no longer declares itself destructive security
- Tool “signals.update” no longer declares itself destructive security
- Tool “inbox.acknowledge” is now declared destructive security
- Schema quality: 59 → 67 ▼ functional
- “inbox.acknowledge” added a required parameter “confirmed”, so existing callers break ▼ functional
- “inbox.dismiss” added a required parameter “confirmed”, so existing callers break ▼ functional
- “inbox.undo” added a required parameter “confirmed”, so existing callers break ▼ functional
- “inbox.unsubscribe” added a required parameter “confirmed”, so existing callers break ▼ functional
- “jobs.delete” added a required parameter “confirmed”, so existing callers break ▼ functional
- “signals.untrack” added a required parameter “confirmed”, so existing callers break ▼ functional
- Schema quality: 78% → 100% ▲ functional
- First check of Destructive annotations: 100 functional
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
- Server version: 0.4.1 → 0.5.1 functional
- New tool “inbox.sync_status” functional
- New tool “jobs.interviews.add” functional
- New tool “jobs.interviews.update” functional
- New tool “jobs.offer.add_negotiation” functional
- New tool “jobs.offer.create” functional
- New tool “jobs.offer.update” functional
- New tool “signals.openings.list” functional
- New tool “signals.openings.mark_all_read” functional
- New tool “signals.openings.mark_read” functional
- New tool “signals.openings.save” functional
- New tool “signals.recommendations.block” functional
- New tool “signals.recommendations.dismiss” functional
- New tool “signals.recommendations.feedback” functional
- New tool “signals.recommendations.get” functional
- New tool “signals.recommendations.list” functional
- New tool “signals.recommendations.restore” functional
- New tool “signals.recommendations.watch” functional
- “inbox.list” added an optional parameter “limit” cosmetic
- “inbox.list” added an optional parameter “page” cosmetic
- “jobs.update” added an optional parameter “notes” cosmetic
- “signals.list_applied_jobs” added an optional parameter “limit” cosmetic
- “signals.list_applied_jobs” added an optional parameter “page” cosmetic
- “jobs.add” dropped the optional parameter “captured_at” cosmetic
- “jobs.add” dropped the optional parameter “journey_id” cosmetic
- “jobs.add” dropped the optional parameter “state” cosmetic
- “jobs.update” dropped the optional parameter “captured_at” cosmetic
- “jobs.update” dropped the optional parameter “journey_id” cosmetic
- “jobs.update” dropped the optional parameter “state” cosmetic
- 24 Aug 26 0
- Stability: unverified → 0.03 ▲ functional
- Server version: 0.4.0 → 0.4.1 functional
- 23 Aug 26 0
First indexed and scored.
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 5 Nov 2026 · Probed https://mcp.getrefine.co/mcp
TLS valid
Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .
| Subject | Issuer | Valid from | Valid until | Key | Signature | Serial |
|---|---|---|---|---|---|---|
| CN=getrefine.co | CN=WE1,O=Google Trust Services,C=US | 7 Oct 2026 | 5 Jan 2027 | ECDSA 256 | ECDSA-SHA256 | 5d8dd664195403d60e796d257f0978a9 |
| SANs: getrefine.co, mcp.getrefine.co, *.mcp.getrefine.co | ||||||
| CN=WE1,O=Google Trust Services,C=US (CA) | CN=GTS Root R4,O=Google Trust Services LLC,C=US | 13 Dec 2023 | 20 Feb 2029 | ECDSA 256 | ECDSA-SHA384 | 7ff31977972c224a76155d13b6d685e3 |
| CN=GTS Root R4,O=Google Trust Services LLC,C=US (CA) | CN=GlobalSign Root CA,OU=Root CA,O=GlobalSign nv-sa,C=BE | 15 Nov 2023 | 28 Jan 2028 | ECDSA 384 | SHA256-RSA | 7fe530bf331343bedd821610493d8a1b |
Background: What to check on a remote MCP endpoint →
DNSSEC insecure
Validation of mcp.getrefine.co. — Not signed
| Zone | DS | Keys | Algorithms | Outcome |
|---|---|---|---|---|
| . | Trust anchor | 20326, 38696 | 8, 8 | Verified |
| co. | DS published | 7786 | 8 | Verified |
| getrefine.co. | No DS | Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation |
Authentication Enforced and verified
The endpoint asked for a token and published valid RFC 9728 metadata describing how to get one.
| Result | Enforced and verified |
|---|---|
| Enforced | On tool calls |
| HTTP status | 200 |
WWW-Authenticate challenge Bearer resource_metadata="https://mcp.getrefine.co/.well-known/oauth-protected-resource", scope="openid read_user read_jobs write_jobs read_companies write_companies read_inbox write_inbox read_integrations write_integrations"
Bearer resource_metadata="https://mcp.getrefine.co/.well-known/oauth-protected-resource", scope="openid read_user read_jobs write_jobs read_companies write_companies read_inbox write_inbox read_integrations write_integrations"Protected resource metadata
| Document | https://mcp.getrefine.co/.well-known/oauth-protected-resource |
|---|---|
| Retrieved | Yes |
| Resource | https://mcp.getrefine.co/mcp |
| Authorisation server | https://api.getrefine.co |
Background: How OAuth 2.1 works in the 2026 MCP spec →
Transports 2 probes
| Transport | URL | Outcome | Status | Location |
|---|---|---|---|---|
| streamable-http | https://mcp.getrefine.co/mcp | Verified | 200 | |
| http (plaintext) | http://mcp.getrefine.co/mcp | Inconclusive | 401 |
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
openings.get ~43
Read an indexed opening, including its full description and qualifications. Use opening.id from Search or Following, not a follow-opening row ID.
| Name | Type | Req | Description |
|---|---|---|---|
| opening_id | string | yes | – |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
No examples provided.
persona.get ~32
Get the user's persona: inferred job-search preferences and traits with an understanding score. Use it to tailor advice and searches.
Input schema present but exposes no named parameters.
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
| meta | object | – | – |
No examples provided.
recommendations.block ~30
Dismiss a recommended company and block it from future recommendations
| Name | Type | Req | Description |
|---|---|---|---|
| recommendation_id | string | yes | Recommendation id |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
| meta | object | – | – |
No examples provided.
recommendations.dismiss ~40
Dismiss a recommended company
| Name | Type | Req | Description |
|---|---|---|---|
| reason | string | – | Why it was dismissed; sharpens future recommendations |
| recommendation_id | string | yes | Recommendation id |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
| meta | object | – | – |
No examples provided.
recommendations.feedback ~96
Record recommendation feedback to improve future suggestions. The historical feedback tag "watched" records that a company was followed; it does not create a follow. Use following.follow first. Dismiss/block/restore actions are separate tools.
| Name | Type | Req | Description |
|---|---|---|---|
| feedback_types | array | yes | One or more supported feedback tags |
| interaction_source | string | – | Where the feedback came from in the UI or agent flow |
| recommendation_id | string | yes | Recommendation id |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
| meta | object | – | – |
No examples provided.
recommendations.get ~28
Get one recommended company with sources and persona fit
| Name | Type | Req | Description |
|---|---|---|---|
| recommendation_id | string | yes | Recommendation id |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
| meta | object | – | – |
No examples provided.
recommendations.list ~102
Discover companies recommended from the companies you follow. Optionally filter by source_company_id (company.id from following.list, not a follow ID). Returns one flat list plus per-source counts. To follow a recommendation, use following.follow with candidate_company.id and your chosen filters.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | Maximum records to return |
| page | string | – | Pagination cursor returned as meta.next_page |
| source_company_id | string | – | Company ID of a followed company. |
| Name | Type | Req | Description |
|---|---|---|---|
| data | array | yes | – |
| meta | object | – | – |
No examples provided.
recommendations.restore ~25
Restore a dismissed or blocked recommendation
| Name | Type | Req | Description |
|---|---|---|---|
| recommendation_id | string | yes | Recommendation id |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
| meta | object | – | – |
No examples provided.
search.company_openings ~86
Read all matching jobs for a company in a ready search, not just its three preview jobs. Continue with data.next_page until null. Use company.id from the search results.
| Name | Type | Req | Description |
|---|---|---|---|
| company_id | string | yes | – |
| page | integer | – | Numeric results page from results.next_page; not a cursor. |
| search_id | string | yes | – |
| sort | string | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
No examples provided.
search.create ~139
Search indexed jobs or companies. Supply input OR query, not both. Natural-language searches return a search ID and processing state: use search.get later, without rapid polling. Keep the ID to paginate or update this same search. Empty query {} is All openings, currently sorted by newest without persona personalization.
| Name | Type | Req | Description |
|---|---|---|---|
| input | string | – | Natural-language search, processed asynchronously. Omit when supplying query. |
| page | integer | – | Numeric results page from results.next_page; not a cursor. |
| query | object | – | Structured filters. Use {} for All openings; results are immediately available. |
| sort | string | – | – |
| view | string | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
No examples provided.
search.get ~117
Read a search and its results, switch jobs/companies view, or fetch results.next_page. Pending/interpreting/processing means it is still working: do not create another search or poll rapidly. If missing_company_name is set, ask for its careers URL and use companies.submissions.create. Company rows contain samples, not all matching jobs.
| Name | Type | Req | Description |
|---|---|---|---|
| page | integer | – | Numeric results page from results.next_page; not a cursor. |
| search_id | string | yes | – |
| sort | string | – | – |
| view | string | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
No examples provided.
search.update ~122
Update the existing search instead of creating another. Supply input OR the complete replacement query; omitted filters are cleared. Wait until the current search is ready or failed before updating it.
| Name | Type | Req | Description |
|---|---|---|---|
| input | string | – | Natural-language search, processed asynchronously. Omit when supplying query. |
| page | integer | – | Numeric results page from results.next_page; not a cursor. |
| query | object | – | Structured filters. Use {} for All openings; results are immediately available. |
| search_id | string | yes | – |
| sort | string | – | – |
| view | string | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| data | object | yes | – |
No examples provided.
How do I install the Refine Job Search MCP server?
Refine Job Search is a hosted endpoint at https://mcp.getrefine.co/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.
remote · mcp.getrefine.co
claude mcp add --transport http co-getrefine-refine 'https://mcp.getrefine.co/mcp'
What is the Refine Job Search MCP server?
Refine Job Search is an MCP server listed in the public MCP registry as co.getrefine/refine. Job search management: applications, interviews, offers, recruiter email, and company openings. This page covers its hosted endpoint (https://mcp.getrefine.co/mcp).
Is the Refine Job Search MCP server safe to use?
Refine Job Search scores 85 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 Refine Job Search MCP server expose?
Refine Job Search exposes 62 tools: jobs.list, jobs.get, jobs.add, jobs.update, jobs.delete, and 57 more. Their descriptions and schemas cost roughly 4,730 tokens of context every time the server is loaded.
Does the Refine Job Search MCP server require authentication?
Yes. Refine Job Search 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 Refine Job Search MCP server still maintained?
Refine Job Search is still listed as active in the MCP registry. We last reached this channel on 5 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 Refine Job Search MCP server?
VerifyMCP ran 12 checks on Refine Job Search, covering TLS certificate, Authorization, HSTS header, DNSSEC and 8 more. One item could not be completed, a gap on our side rather than a finding about Refine Job Search: HTTPS. 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.