Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

Flux LinkedIn MCP

REMOTE · WWW.FLUXGRAPH.APP

LinkedIn intelligence: post analytics and research for public profiles and company pages.

+6 this week 88 Trust /100

Last scanned (17 hours ago)

Scores higher than 95% of remote servers

Reachable. Last reached 21 Nov 2026.

Category Marketing and SEO

Trust breakdown (7 categories)

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 Security94
Transport & Reachability100
Schema Quality & AI Usability78
  • 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 14631 tokens (~203/item across 72 items; 72 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management63
  • Stability observed for 19 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • All 3 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
  • An AI judge read all 73 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
What we tested

Checks we ran (13)

  • TLS certificate
  • Authorization
  • HTTPS
  • HSTS header
  • DNSSEC
  • Transport
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

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 →

Changelog

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.

  • 19 Nov 26 +1
    • Tool “log_outcome” rewrote its description, which is the text the model reads security
    • New tool “get_inbound_attribution” functional
    • “log_outcome” added an optional parameter “message_text” cosmetic
    • “log_outcome” added an optional parameter “occurred_at” cosmetic
    • “log_outcome” added an optional parameter “outcome_id” cosmetic
    • “log_outcome” added an optional parameter “post_url” cosmetic
    • “log_outcome” added an optional parameter “profile_id” cosmetic
    • “log_outcome” reworded the description of “action” cosmetic
    • “log_outcome” reworded the description of “campaign_id” cosmetic
    • “log_outcome” reworded the description of “linkedin_url” cosmetic
    • “log_outcome” reworded the description of “person_id” cosmetic
    • “log_outcome” made “campaign_id” optional cosmetic
    • Tool “log_outcome” changed its title: Log Outreach Outcome → Log Outcome cosmetic
  • 17 Nov 26 +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.

  • 15 Nov 26 +4
    • The server rewrote its instructions, which are the text every model session reads security
    • Tool “add_campaign_accounts” rewrote its description, which is the text the model reads security
    • Tool “add_to_collection” rewrote its description, which is the text the model reads security
    • Tool “add_to_segment” rewrote its description, which is the text the model reads security
    • Tool “claim_preview” rewrote its description, which is the text the model reads security
    • Tool “create_campaign” rewrote its description, which is the text the model reads security
    • Tool “create_collection” rewrote its description, which is the text the model reads security
    • Tool “create_segment” rewrote its description, which is the text the model reads security
    • Tool “create_stream” rewrote its description, which is the text the model reads security
    • Tool “delete_stream” rewrote its description, which is the text the model reads security
    • Tool “enrich_people” rewrote its description, which is the text the model reads security
    • Tool “find_similar_posts” rewrote its description, which is the text the model reads security
    • Tool “get_alerts” rewrote its description, which is the text the model reads security
    • Tool “get_alumni_engagers” rewrote its description, which is the text the model reads security
    • Tool “get_campaign_stats” rewrote its description, which is the text the model reads security
    • Tool “get_company_engagement_overlap” rewrote its description, which is the text the model reads security
    • Tool “get_cost_estimate” rewrote its description, which is the text the model reads security
    • Tool “get_credit_balance” rewrote its description, which is the text the model reads security
    • Tool “get_engagement_brief” rewrote its description, which is the text the model reads security
    • Tool “get_engagement_industry_breakdown” rewrote its description, which is the text the model reads security
    • Tool “get_engagement_trends” rewrote its description, which is the text the model reads security
    • Tool “get_engager_position_history” rewrote its description, which is the text the model reads security
    • Tool “get_outreach_path” rewrote its description, which is the text the model reads security
    • Tool “get_outreach_queue” rewrote its description, which is the text the model reads security
    • Tool “get_post_engagers_by_company” rewrote its description, which is the text the model reads security
    • Tool “get_post_set_engagers” rewrote its description, which is the text the model reads security
    • Tool “get_posting_insights” rewrote its description, which is the text the model reads security
    • Tool “get_posts” rewrote its description, which is the text the model reads security
    • Tool “get_profile_summary” rewrote its description, which is the text the model reads security
    • Tool “get_stream_feed” rewrote its description, which is the text the model reads security
    • Tool “get_target_warmth_brief” rewrote its description, which is the text the model reads security
    • Tool “get_topic_insights” rewrote its description, which is the text the model reads security
    • Tool “get_usage_history” rewrote its description, which is the text the model reads security
    • Tool “linkedin_ad_search” rewrote its description, which is the text the model reads security
    • Tool “linkedin_company_info” rewrote its description, which is the text the model reads security
    • Tool “linkedin_company_posts” rewrote its description, which is the text the model reads security
    • Tool “linkedin_company_search” rewrote its description, which is the text the model reads security
    • Tool “linkedin_job_search” rewrote its description, which is the text the model reads security
    • Tool “linkedin_people_search” rewrote its description, which is the text the model reads security
    • Tool “linkedin_post_comments” rewrote its description, which is the text the model reads security
    • Tool “linkedin_post_reactions” rewrote its description, which is the text the model reads security
    • Tool “linkedin_post_search” rewrote its description, which is the text the model reads security
    • Tool “linkedin_profile_comments” rewrote its description, which is the text the model reads security
    • Tool “linkedin_profile_info” rewrote its description, which is the text the model reads security
    • Tool “linkedin_profile_posts” rewrote its description, which is the text the model reads security
    • Tool “linkedin_profile_reactions” rewrote its description, which is the text the model reads security
    • Tool “list_collections” rewrote its description, which is the text the model reads security
    • Tool “list_profiles” rewrote its description, which is the text the model reads security
    • Tool “list_segment_templates” rewrote its description, which is the text the model reads security
    • Tool “list_segments” rewrote its description, which is the text the model reads security
    • Tool “list_streams” rewrote its description, which is the text the model reads security
    • Tool “log_outcome” rewrote its description, which is the text the model reads security
    • Tool “map_constellation” rewrote its description, which is the text the model reads security
    • Tool “pause_stream” rewrote its description, which is the text the model reads security
    • Tool “preview_posting_insights” rewrote its description, which is the text the model reads security
    • Tool “preview_profile” rewrote its description, which is the text the model reads security
    • Tool “preview_search_posts” rewrote its description, which is the text the model reads security
    • Tool “remove_from_collection” rewrote its description, which is the text the model reads security
    • Tool “remove_from_segment” rewrote its description, which is the text the model reads security
    • Tool “resolve_collection” rewrote its description, which is the text the model reads security
    • Tool “resolve_segment” rewrote its description, which is the text the model reads security
    • Tool “resume_stream” rewrote its description, which is the text the model reads security
    • Tool “run_stream_now” rewrote its description, which is the text the model reads security
    • Tool “save_post” rewrote its description, which is the text the model reads security
    • Tool “score_connections” rewrote its description, which is the text the model reads security
    • Tool “score_draft” rewrote its description, which is the text the model reads security
    • Tool “search_posts” rewrote its description, which is the text the model reads security
    • Tool “suggest_angles” rewrote its description, which is the text the model reads security
    • Tool “traverse_graph” rewrote its description, which is the text the model reads security
    • Tool “traverse_graph_help” rewrote its description, which is the text the model reads security
    • Tool “claim_preview” is now declared destructive security
    • Tool “delete_stream” is now declared destructive security
    • Tool “remove_from_collection” is now declared destructive security
    • Tool “remove_from_segment” is now declared destructive security
    • Schema quality: 158 → 198 ▼ functional
    • The server now declares the “prompts” capability functional
    • First check of Schema quality: 100 functional
    • Destructive annotations: All 3 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation. functional
    • New prompt “monthly-client-report” functional
    • New prompt “research-a-prospect” functional
    • New prompt “triage-linkedin-urls” functional
    • New prompt “warm-path-to” functional
    • New prompt “warm-way-into-account” functional
    • New prompt “what-to-post-more-of” functional
    • New prompt “who-at-company-engages” functional
    • New prompt “who-engaged” functional
    • New tool “resolve_company” functional
    • New tool “resolve_people” functional
    • “get_alumni_engagers” added an optional parameter “company_name” cosmetic
    • “get_alumni_engagers” added an optional parameter “linkedin_company_id” cosmetic
    • “get_company_engagement_overlap” added an optional parameter “company_a_linkedin_id” cosmetic
    • “get_company_engagement_overlap” added an optional parameter “company_a_name” cosmetic
    • “get_company_engagement_overlap” added an optional parameter “company_b_linkedin_id” cosmetic
    • “get_company_engagement_overlap” added an optional parameter “company_b_name” cosmetic
    • “get_post_set_engagers” added an optional parameter “post_ids” cosmetic
    • “get_post_set_engagers” added an optional parameter “post_urls” cosmetic
    • “log_outcome” added an optional parameter “linkedin_url” cosmetic
    • “create_segment” reworded the description of “config” cosmetic
    • “get_cost_estimate” reworded the description of “params” cosmetic
    • “get_cost_estimate” reworded the description of “tool_name” cosmetic
    • “get_post_set_engagers” reworded the description of “max_posts” cosmetic
    • “log_outcome” reworded the description of “person_id” cosmetic
    • “log_outcome” made “person_id” optional cosmetic
  • 14 Nov 26 0
    • Tool “add_campaign_accounts” rewrote its description, which is the text the model reads security
    • Tool “add_to_segment” rewrote its description, which is the text the model reads security
    • Tool “create_segment” rewrote its description, which is the text the model reads security
    • Tool “get_alumni_engagers” rewrote its description, which is the text the model reads security
    • Tool “get_campaign_stats” rewrote its description, which is the text the model reads security
    • Tool “get_credit_balance” rewrote its description, which is the text the model reads security
    • Tool “get_outreach_path” rewrote its description, which is the text the model reads security
    • Tool “linkedin_company_search” rewrote its description, which is the text the model reads security
    • Tool “linkedin_people_search” rewrote its description, which is the text the model reads security
    • Tool “list_segment_templates” rewrote its description, which is the text the model reads security
    • Tool “log_outcome” rewrote its description, which is the text the model reads security
    • Tool “map_constellation” rewrote its description, which is the text the model reads security
    • Tool “remove_from_segment” rewrote its description, which is the text the model reads security
    • Tool “resolve_segment” rewrote its description, which is the text the model reads security
    • Schema quality: 143 → 158 ▼ functional
    • New tool “enrich_people” functional
    • New tool “linkedin_company_info” functional
    • “linkedin_company_search” added an optional parameter “page” cosmetic
    • “linkedin_people_search” added an optional parameter “page” cosmetic
    • “create_segment” reworded the description of “config” cosmetic
    • “get_cost_estimate” reworded the description of “params” cosmetic
    • “linkedin_job_search” reworded the description of “companyId” cosmetic
    • “linkedin_people_search” reworded the description of “company” cosmetic
    • “linkedin_people_search” reworded the description of “title” cosmetic
    • “map_constellation” reworded the description of “max_targets” cosmetic
  • 12 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 30 to 33. That category is still filling its 30-day observation window: 9 days of observed history at the previous scan, 10 at this one. The score rises as the window fills, whether or not the server changes.

  • 11 Nov 26 0
    • Tool “get_engagement_industry_breakdown” rewrote its description, which is the text the model reads security
    • Tool “get_post_set_engagers” rewrote its description, which is the text the model reads security
    • Tool “linkedin_post_comments” rewrote its description, which is the text the model reads security
    • Tool “linkedin_post_reactions” rewrote its description, which is the text the model reads security
    • “get_post_set_engagers” added an optional parameter “max_pages_per_post” cosmetic
    • “linkedin_post_comments” added an optional parameter “max_pages” cosmetic
    • “linkedin_post_reactions” added an optional parameter “max_pages” cosmetic
    • “linkedin_post_comments” reworded the description of “refresh” cosmetic
    • “linkedin_post_reactions” reworded the description of “refresh” cosmetic
  • 10 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

  • 8 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes.

Diagnostics

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 21 Nov 2026 · Probed https://www.fluxgraph.app/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=*.fluxgraph.app CN=YR2,O=Let's Encrypt,C=US 10 Oct 2026 8 Jan 2027 RSA 2048 SHA256-RSA 5295c50efc4a846063868806d0cf31bc630
SANs: *.fluxgraph.app, fluxgraph.app
CN=YR2,O=Let's Encrypt,C=US (CA) CN=Root YR,O=ISRG,C=US 3 Sept 2025 2 Sept 2028 RSA 2048 SHA256-RSA 4ebd24947e24d394802d84a52fd5b319
CN=Root YR,O=ISRG,C=US (CA) CN=ISRG Root X1,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 RSA 4096 SHA256-RSA f24b6d17f9d9ad7cb1c9fea78782699f

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of www.fluxgraph.app. — Not signed

Zone DS Keys Algorithms Outcome
. Trust anchor 20326, 38696 8, 8 Verified
app. DS published 23684 8 Verified
fluxgraph.app. 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 error="invalid_token", error_description="Sign in to Flux to use this tool", resource_metadata="https://www.fluxgraph.app/.well-known/oauth-protected-resource/mcp"

Bearer error="invalid_token", error_description="Sign in to Flux to use this tool", resource_metadata="https://www.fluxgraph.app/.well-known/oauth-protected-resource/mcp"
Header Value
strict-transport-security max-age=63072000

Protected resource metadata

Document https://www.fluxgraph.app/.well-known/oauth-protected-resource/mcp
Retrieved Yes
Resource https://www.fluxgraph.app/mcp
Authorisation server https://flqwttpbzetagwisyoly.supabase.co/auth/v1

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://www.fluxgraph.app/mcp Verified 200
http (plaintext) http://www.fluxgraph.app/mcp HTTPS enforced 308 https://www.fluxgraph.app/mcp
MCP tools · 72 exposed · ~13,766 tokens

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 →

Tool Tokens
list_streams ~99

[Read · free] List all research streams for the authenticated user (external profiles, companies, topics, person interests, hiring signals, ad watches, and own_profile). Pass `type` to filter. Replaces list_profiles for the streams-first world, but list_profiles continues to work for V1-compatible callers.

NameTypeReqDescription
include_inactiveboolean–Include paused streams (active = false)
typestring–Filter by stream type

No output schema declared.

No examples provided.

log_outcome ~622

[Act · writes] Record an outcome by hand. Two kinds. • Outbound (campaign outreach): action sent_request | accepted | replied | no_response with campaign_id and exactly one of person_id or linkedin_url (any profile URL form). The person must be connected to you (in one of your campaigns or segments, engaged with posts you can see in Flux, or pulled by you); anyone else returns not-found with a next_step to pull their profile via linkedin_profile_info. A person outside this campaign's mapped constellation is still logged (with a `warning`) and get_campaign_stats counts it under `unattributed`. • Inbound (results your own posting brought in): action inbound_dm | lead | client with profile_id (your tracked profile from list_profiles, whose posts it counts against). No campaign. When the user pastes a LinkedIn DM conversation, pass the whole paste as message_text and set occurred_at to the date of its first message (ISO 8601; omit only if the paste has no date). The sender's linkedin_url is optional and never triggers a paid lookup. Flux matches the outcome to the post that most likely prompted it: post_url if given, else for a lead or client the post already credited for this person's earlier inbound outcome, else the conversation's opening text and timing against your posts from the 30 days before. The result says the method (explicit | inherited | matched | recency | none) and lists up to three candidate posts. When nothing was matched and candidates exist, show them to the user and ask which post it was; then call again with outcome_id and that post_url to correct it. get_inbound_attribution reports which posts bring in DMs, leads and clients. Free — costs 0 credits.

NameTypeReqDescription
actionstringyesOutbound: sent_request | accepted | replied | no_response. Inbound: inbound_dm | lead | client
campaign_idstring–Outbound only, required there: campaign id
linkedin_urlstring–LinkedIn profile URL of the person. Outbound: provide this or person_id. Inbound: optional (the sender)
message_textstring–Inbound: the DM or message text; paste the whole conversation if you have it
notestring–Optional free-text note
occurred_atstring–Inbound: when it happened, ISO 8601 (for a pasted DM, the first message's date). Defaults to now
outcome_idstring–Inbound correction: the id of an inbound outcome you logged, with post_url
person_idstring–people.id of the person. Outbound: provide this or linkedin_url. Inbound: optional
post_urlstring–Inbound: URL of your post that prompted it, when the user knows. With outcome_id, corrects an earlier inbound outcome
profile_idstring–Inbound only, required there: your tracked profile id from list_profiles

No output schema declared.

No examples provided.

map_constellation ~315

[Acquire · 1 + 3×targets credits] Map the professional graph around a target company. Finds senior decision-makers at the account, pulls each target's recent posts/reactions/comments to identify engagers, then scores every connection and produces a prioritized outreach queue. The response lists the selected `targets` (target_id, person_id, full_name, current_title, linkedin_url) and `_meta.next_step(s)`: get_outreach_queue for the scored queue, then get_outreach_path (free) to check warm paths to the top targets. Pools: coworker = other senior people at the same account from the lead search (no relationship with you is implied); engager = someone a target engages with; mutual = someone several targets engage with. Costs at most 1 + 3 × max_targets credits (7 at max_targets=2, 16 at the default 5, 61 at 20): the lead search is a bundled 1 credit, a deliberate discount on linkedin_people_search's 5 credits per page; the balance must cover that worst case, and only what is actually spent is charged and reported in the response.

NameTypeReqDescription
campaign_account_idstringyescampaign_accounts.id — the company to map
max_targetsinteger–Max targets to deep-map per account (1-20, default 5). Each target adds up to 3 credits.
title_keywordsarray–Title keywords to filter targets. Default: ['VP','Director','Head of','Chief']

No output schema declared.

No examples provided.

pause_stream ~66

[Act · writes] Pause a stream so it stops running on its cron schedule. Existing posts remain in the library. Use resume_stream to re-enable it. own_profile streams are billing-managed and cannot be paused here.

NameTypeReqDescription
stream_idstringyesStream ID from list_streams

No output schema declared.

No examples provided.

preview_posting_insights ~96

[Read · free] Full breakdown for a previewed profile (from preview_profile): best day, hour, format and length, the top and bottom 5 posts with measured reasons, and engagement trend by format. Requires a free Flux account; calling it asks the person to sign in where the client supports it, otherwise use continue_url from preview_profile.

NameTypeReqDescription
preview_idstringyespreview_id returned by preview_profile

No output schema declared.

No examples provided.

preview_profile ~182

[Acquire · free] Analyze a public LinkedIn profile or company page without a Flux account: pulls the latest ~50 posts (once per profile) and returns a teaser: post count, average engagement, the format that beats the baseline, the top post and the recent trend, plus a preview_id. The full breakdown (best day and hour, top and bottom posts with reasons, trends by format) and keyword search need a free Flux account: call preview_posting_insights or preview_search_posts with the preview_id, which asks the person to sign in where the client supports it, or give the person the continue_url if that call fails. Call claim_preview after sign-in only when the person says the profile is theirs.

NameTypeReqDescription
linkedin_urlstringyesPublic LinkedIn profile or company URL, e.g. https://www.linkedin.com/in/janedoe

No output schema declared.

No examples provided.

preview_search_posts ~108

[Read · free] Keyword search over the latest 50 posts of a previewed profile (from preview_profile). Case-insensitive substring match; returns matching posts ordered by engagement. Requires a free Flux account; calling it asks the person to sign in where the client supports it, otherwise use continue_url from preview_profile.

NameTypeReqDescription
preview_idstringyespreview_id returned by preview_profile
querystringyesText to look for in the posts, 2 to 100 characters

No output schema declared.

No examples provided.

remove_from_collection ~77

[Act · writes · destructive] Remove one or more posts (by post_id) from a collection. Missing members are silently ignored. Free — costs 0 credits.

NameTypeReqDescription
collection_idstringyesTarget collection id from list_collections
post_idsarrayyespost_ids to remove (1-1000 per call)

No output schema declared.

No examples provided.

remove_from_segment ~95

[Act · writes · destructive] Remove one or more people (by person_id) from a manual audience segment. Missing members are silently ignored. Smart segments are rejected: their membership is computed from `config` by resolve_segment. Free — costs 0 credits.

NameTypeReqDescription
person_idsarrayyesperson_ids to remove (1-1000 per call)
segment_idstringyesTarget segment id from list_segments

No output schema declared.

No examples provided.

resolve_collection ~98

[Read · free] Resolve a post collection into a concrete list of post_ids. For manual / auto_stream collections, returns stored members. For smart collections, evaluates the collection's `config` predicate against the user's posts and returns matches. Free — costs 0 credits.

NameTypeReqDescription
collection_idstringyesCollection id (from list_collections / create_collection)
limitinteger–Max post_ids to return (default 1000)

No output schema declared.

No examples provided.

resolve_company ~264

[Read · free] Look up a company saved in Flux by exactly one of name (case-insensitive exact match), linkedin_url, linkedin_company_id (LinkedIn's numeric id) or company_id, free. Returns company_id, linkedin_company_id, name, linkedin_url, page_type and coverage for you: connected_people_current (people connected to you, meaning they engaged with posts you can see, are in your campaigns or segments, or you pulled or enriched them, who hold a current position there) and engagers_current (those of them who engaged with your posts, with their engagement total). A name shared by several saved companies returns `candidates` to pick from. A company that is not saved returns _meta.next_step = linkedin_company_info (1 credit), which saves it. Generic placeholders (Self-employed, Freelance, Stealth...) are refused. Pure DB read, no LinkedIn pull, 0 credits.

NameTypeReqDescription
company_idstring–companies.id UUID
linkedin_company_idstring|number–LinkedIn's numeric company id, as linkedin_company_info returns it
linkedin_urlstring–Company LinkedIn URL (linkedin.com/company/…)
namestring–Company name, exact and case-insensitive

No output schema declared.

No examples provided.

resolve_people ~279

[Read · free] Resolve up to 200 LinkedIn profile URLs and/or person_ids to people in your Flux graph, free. Each result has person_id, the stored LinkedIn URL, matched_via (exact / canonical / member_id / public_identifier / none), name, current title and company, and coverage: has_positions (job history on file), enriched_at, engaged_with_you (engagements on your posts), in_campaigns (how many of your campaigns include them) and connected_via (engaged / campaign / segment / pulled). Only people connected to you resolve: someone who engaged with posts you can see, is in your campaigns or segments, or whom you pulled or enriched. Anyone else comes back as matched_via "none", the same as an unknown URL. _meta.next_step is the enrich_people call (with its credit cost) that saves the unmatched URLs and fills job history for matched people who lack it. Use it to triage a list of URLs before spending credits. Pure DB read, no LinkedIn pull, 0 credits.

NameTypeReqDescription
linkedin_urlsarray–LinkedIn profile URLs, any form (vanity, /in/ACo…, trailing slash, utm params). 1-200.
person_idsarray–people.id UUIDs. 1-200.

No output schema declared.

No examples provided.

resolve_segment ~151

[Read · free] Resolve an audience segment into a concrete list of person_ids. For manual / auto_stream segments, returns stored members. For smart segments, evaluates the segment's `config` predicate (title_contains / company_contains) and returns matches ordered by engagement with your posts. Smart segments only match people connected to you: people who engaged with posts you can see in Flux (your own profile, tracked/monitored profiles, and posts you collected) and people in your campaigns' constellations. Free — costs 0 credits.

NameTypeReqDescription
limitinteger–Max person_ids to return (default 1000)
segment_idstringyesSegment id (from list_segments / create_segment)

No output schema declared.

No examples provided.

resume_stream ~63

[Act · writes] Resume a paused stream. Sets active=true and queues it to run on the next cron tick by clearing next_run_at. own_profile streams are billing-managed and cannot be resumed here.

NameTypeReqDescription
stream_idstringyesStream ID from list_streams

No output schema declared.

No examples provided.

run_stream_now ~416

[Acquire · stream price/page] Trigger an immediate run of an active stream, bypassing its schedule. The run is fire-and-forget — the response confirms the dispatch but post collection happens asynchronously. Credits are charged for pages actually walked, not pages requested: one page is taken upfront and the rest settled when the run finishes, so a run that stops early (or fails) costs less than max_pages implies. Paused streams must be resumed first. Use list_streams or get_stream_feed to see results once the run completes.

NameTypeReqDescription
history_daysinteger–How far back this run should reach, in days. Bounds the walk by TIME where max_pages bounds it by COST — it stops at whichever limit it hits first, so pair a deep window with enough pages to reach it…
max_pagesinteger–Upper bound on pages for this run (1-5, default 1) — a budget, not a price. Honoured for external_profile, company and person_interests. You are billed only for pages actually walked; a person_intere…
resumeboolean–Continue a person_interests backfill from where the deepest previous run stopped, instead of re-walking the newest pages. Use this for 'pull more history' — without it you re-fetch and re-pay for con…
stream_idstringyesStream ID from list_streams

No output schema declared.

No examples provided.

save_post ~150

[Acquire · 1 credit · writes] Save a LinkedIn post to your library by its URL. Resolves the post and stores it with source_type='saved'. Optionally pins the post into any collection you own (manual or smart — pinning into a smart collection adds it as an extra alongside the rule's matches). Costs 1 credit per call. Failed lookups don't charge.

NameTypeReqDescription
collection_idstring–Optional collection UUID to pin the post into. Must be a collection you own — any type.
urlstringyesPublic LinkedIn post URL. Accepts linkedin.com/posts/... permalinks and linkedin.com/feed/update/urn:li:activity:... feed URLs.

No output schema declared.

No examples provided.

score_connections ~145

[Act · writes] Re-score all constellation entries in a campaign and rewrite queue positions from highest to lowest composite score. Use after new engagers or targets are added. Free — costs 0 credits. Note: this tool requires a `campaign_id` because it writes back to the `constellation_entries` table. Segment- and url-list-scoped scoring is not yet supported — use `get_outreach_queue({segment_id})` or `get_outreach_queue({linkedin_urls})` for engagement-count ordering on those scopes; full composite scoring will land alongside the unified person-positions schema in a later phase.

NameTypeReqDescription
campaign_idstringyesCampaign id to re-score

No output schema declared.

No examples provided.

score_draft ~225

[Read · free] Predict how a draft post will perform against the user's historical data. Uses an OLS regression model with detrending, ridge regularization, and recency weighting — the same model used on the /score page. Returns a predicted engagement percentile, projected engagement range, factor breakdown, similar past posts, and improvement suggestions. Free. Pass `profile_id` to scope to a tracked profile, `collection_id` to scope to a collection, or neither to score against the user's full library.

NameTypeReqDescription
collection_idstring–Optional post_collections.id (scopes the model to this collection)
day_of_weeknumberyesPlanned day of week (0=Sunday, 6=Saturday, UTC)
formatstring–Post format: text, image, video, document, article, or poll
hournumberyesPlanned hour (0-23, UTC)
profile_idstring–Optional tracked profile ID (scopes the model to this profile's posts)
textstringyesDraft post text

No output schema declared.

No examples provided.

search_posts ~185

[Read · free] Keyword + semantic search over persisted posts in the local library — full-text + pgvector cosine, no LinkedIn pull, free. Use this when the caller has a query string and wants ranked persisted-post results. Pass `scope` (a post_collections.id) to limit to a specific collection. For fresh LinkedIn keyword pulls, use `linkedin_post_search` instead (5 credits).

NameTypeReqDescription
limitnumber–Max results to return
modestring–Search mode: hybrid (keyword + semantic), keyword (FTS only), or semantic (embedding similarity only)
profile_idstringyesTracked profile ID from list_profiles
querystringyesSearch query (e.g. 'cybersecurity tips', 'leadership')
scopestring–Optional post_collections.id to restrict results to a collection

No output schema declared.

No examples provided.

suggest_angles ~159

[Read · 5 credits] Given a topic, suggest distinct content angles informed by what performs well in the user's scoped data. Pulls top-performing historical posts as context and calls OpenAI to generate 5 concrete angles. Costs 5 credits per call. Pass `profile_id` to scope to a tracked profile, `collection_id` to scope to a collection, or neither to use the user's full library.

NameTypeReqDescription
collection_idstring–Optional post_collections.id (scopes to this collection's members)
profile_idstring–Optional tracked profile ID (scopes to this profile's posts)
topicstringyesTopic or theme to generate angles for (e.g. 'onboarding new hires', 'pricing strategy')

No output schema declared.

No examples provided.

traverse_graph ~126

[Read · free] Executes a TS-validated SELECT-only SQL query against the engagement graph and returns rows. Needs the schema and the caller_user_ids / caller_tracked_lp_ids that traverse_graph_help returns (it also has canonical examples). Required: SQL must start with SELECT or WITH, contain no DML/DDL keywords, and reference one of your caller_user_ids or caller_tracked_lp_ids verbatim. Cap: 1000 rows, 5s statement timeout. Pure DB read — 0 credits.

NameTypeReqDescription
sqlstringyesThe SELECT/WITH SQL to execute

No output schema declared.

No examples provided.

traverse_graph_help ~81

[Meta · free] Returns the schema, FK edges, and 5 canonical SQL examples needed to write your own SELECT against the engagement graph via traverse_graph. Also returns YOUR resolvable identities (caller_user_ids, caller_tracked_lp_ids) — substitute these into example SQL when you write your query. Pure DB read — 0 credits. Free.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

Install

How do I install the Flux LinkedIn MCP server?

Flux LinkedIn MCP is a hosted endpoint at https://www.fluxgraph.app/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 · www.fluxgraph.app

# add to Claude Code
claude mcp add --transport http app-fluxgraph-linkedin 'https://www.fluxgraph.app/mcp'
Common questions

What is the Flux LinkedIn MCP server?

Flux LinkedIn MCP is listed in the public MCP registry as app.fluxgraph/linkedin. LinkedIn intelligence: post analytics and research for public profiles and company pages. This page covers its hosted endpoint (https://www.fluxgraph.app/mcp).

Is the Flux LinkedIn MCP server safe to use?

Flux LinkedIn MCP scores 88 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 Flux LinkedIn MCP server expose?

Flux LinkedIn MCP exposes 72 tools: list_profiles, preview_profile, preview_posting_insights, preview_search_posts, claim_preview, and 67 more. Their descriptions and schemas cost roughly 13,766 tokens of context every time the server is loaded.

Does the Flux LinkedIn MCP server require authentication?

Yes. Flux LinkedIn MCP 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 Flux LinkedIn MCP server still maintained?

Flux LinkedIn MCP 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 Flux LinkedIn MCP server?

VerifyMCP ran 13 checks on Flux LinkedIn MCP, 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.