Resources · Claude connector

Claude Productised connectorWhat Claude can do in your Productised workspace.

The Claude Productised connector exposes 73 tools: 34 read your workspace, 38 change it, 1 is undocumented. Here: what Claude can build, edit and publish, what it reads about your leads, what it asks first, what it cannot touch, and who can switch it off.

Verified Trustpilot reviews · AI, automation & growth agency

Overview

What changes when Claude works inside Productised

Productised hosts AI products such as assessments and lead magnets: a respondent answers questions, gets an AI-generated result and becomes a lead in your pipeline. With this connector you build, edit and publish them by talking to Claude. Its authorization screen says Claude asks to read products, leads and brand and to write changes, and every call is logged.

A product from a description. You describe the product you want: plan_product recommends the outcome type and funnel stage, create_product opens an empty record, generate_product builds the conversation flow and workflow, and publish_product returns the public link.

Leads worked from the chat. get_morning_briefing sums up new leads and hot prospects, get_hot_leads surfaces the most promising ones, and update_response_status moves a lead along your pipeline, from new to contacted and beyond.

Result pages reworked without the canvas. get_page pulls the current HTML of a result page, you discuss a new version with Claude, and update_page replaces it.

Now what the directory sheet does not say. Productised documents what the connector cannot do: delete a lead, change billing or subscription state, read or send personal email, or reach another workspace. Deleting a product is possible but needs an explicit confirm flag. And nothing runs on its own: no tool fires when a new submission arrives. For that kind of background reaction you need an automation tool, a different job; the Integrations hub gathers that side.

Vocabulary

The vocabulary in one minute

Five words you will meet while connecting Productised.

Connector
The link you set up once between Claude and an account you already own, so Claude can work in it while it answers you.
Tool
One named thing a connector lets Claude do in that account. Claude picks the ones it needs; the directory sheet lists them by name.
Authorization
Productised's own sign-in screen, where you hand Claude the access it will use. Granted once per person, and it can be taken back.
Approval
The confirmation Claude waits for before it goes ahead with something that changes your workspace, shown in the chat when it matters.
MCP
The shared standard connectors are built on: it is what lets an assistant such as Claude talk to a service outside it.
Connect

Setting up the Claude Productised connector in three steps

  1. 01

    Find Productised in Claude

    Open Claude's settings, go to Customize, then Connectors, and look for Productised in the list. On a Team or Enterprise workspace, an Owner or Primary Owner enables the connector before each member can sign in on their own.

  2. 02

    Connect from its row

    Click Connect on the Productised row, then sign in to Productised in the window the service opens itself. If the link breaks later, Disconnect and connect again.

  3. 03

    Read the authorization screen

    Read the authorization screen before you accept. It belongs to Productised, not to Claude, and it is what decides how far the access reaches. Productised says that access can be withdrawn from its own settings.

Tools

The 73 tools, split by what they do

Productised gives Claude 73 tools: 34 that read your account, 38 that change something in it, and 1 no official source describes.

Three groups: what Claude reads in your workspace, what it builds or changes there, and the one tool no source classifies. Names stay exactly as Claude displays them.

  • 34 read
  • 38 write
  • 1 not documented

What Claude reads (34)

34 tools

Thirty-four tools that look at products, leads, pages, plans and Brain, and leave them untouched.

brain_ask

Puts a strategic question to Brain, Productised's advisor, which loads your workspace context automatically before it answers. Claude relays the question and brings Brain's answer back into the conversation.

When it helps
you want Brain's view on which audience a next product should target, based on your workspace.

Sourceproductised.ai · October 1, 2026 ↗

generate_outbound_url

Produces one link to a product with any key and value parameters you choose, and warns when a parameter is not wired into the product. Productised says it replaces the older get_product_url for outbound use.

When it helps
you need one personalised link for a single prospect.

Sourceproductised.ai · October 1, 2026 ↗

generate_outbound_urls_from_csv

Takes a list as CSV text and returns it enriched with a new column of outbound links, one personalised link per row; an optional key mapping lets a header such as First Name match the token firstName.

When it helps
you have a prospect list and need a distinct link for each row.

Sourceproductised.ai · October 1, 2026 ↗

get_analytics

Returns a product's response count, last submission, 14-day trend and completion rate, which Productised offers for any product. The trend covers the last 14 days of submissions.

When it helps
you want to know whether respondents finish a product they started, and how many came in lately.

Sourceproductised.ai · October 1, 2026 ↗

get_credits

Shows a visual credit gauge: plan title, balance, usage this period and all time, per-product breakdown and BYOK status, meaning whether you bring your own AI key. Productised says Claude uses it for questions on credits, usage or cost.

When it helps
you want to see how much credit is left before running generation again.

Sourceproductised.ai · October 1, 2026 ↗

get_design_system

Reads the workspace design-system tokens, brand colours, fonts and corner radius, with platform defaults filled in for anything left unset.

When it helps
you want to know which brand values the products of the workspace inherit before touching their look.

Sourceproductised.ai · October 1, 2026 ↗

get_domain_status

Checks whether DNS has propagated for a product's custom domain, so Claude can tell you whether the address is ready to use.

When it helps
you set up the CNAME record at your registrar and the branded address does not load yet.

Sourceproductised.ai · October 1, 2026 ↗

get_form_questions

Lists the current intake questions of a product, with their names, labels and the flag that says whether each one is required. Productised returns the set as it stands today, required fields marked.

When it helps
you want to know what respondents must fill in before the AI conversation starts.

Sourceproductised.ai · October 1, 2026 ↗

get_hot_leads

Returns your most promising leads across all products, those tagged hot-lead or carrying a high AI score, with optional filters on product, score threshold and lookback period.

When it helps
you want to know which respondents to follow up first.

Sourceproductised.ai · October 1, 2026 ↗

get_landing_page_template

Fetches the designer-built default landing page, in grid or signature style, already themed to your product's brand colour. Productised presents it as the starting point for a custom welcome page.

When it helps
you want a solid starting point for a product's welcome page before Claude adapts the wording.

Sourceproductised.ai · October 1, 2026 ↗

get_morning_briefing

Delivers a full daily briefing: new leads today against yesterday, hot prospects needing attention and a per-product activity breakdown, with no parameters to give. Productised suggests starting each session with it.

When it helps
you open a session and want the state of the workspace before anything else.

Sourceproductised.ai · October 1, 2026 ↗

get_page

Retrieves the current HTML and tokens of the output pages of a product's Page node, the result screens respondents reach. The HTML comes back with the tokens the page uses.

When it helps
you want Claude to read an existing result page before discussing a new version with you.

Sourceproductised.ai · October 1, 2026 ↗

get_plan_status

Reports your current Productised plan, usage of every gated feature such as live products and AI credits, and the links to upgrade or bring your own key.

When it helps
you are about to launch a fourth product and wonder whether your Productised subscription allows it.

Sourceproductised.ai · October 1, 2026 ↗

get_product

Returns the full details of one product: its form questions, its workflow structure and the identifiers of its nodes, which the node tools then need.

When it helps
you need to understand how an existing product is put together before touching it.

Sourceproductised.ai · October 1, 2026 ↗

get_product_insights

Brings back AI-generated insights across all the submissions of a product: patterns, surprises and opportunities. Productised generates them; Claude retrieves them. They are built across all submissions of the product, not response by response.

When it helps
you want the main takeaways from a product's answers without opening each response.

Sourceproductised.ai · October 1, 2026 ↗

get_product_structure

Shows a product's node graph: which nodes exist, how they connect and what each one does in the flow.

When it helps
respondents skip a step you expected them to see and you want to trace the path they take.

Sourceproductised.ai · October 1, 2026 ↗

get_product_url

Returns a shareable link to a product, optionally pre-filled with a contact's name and email so it is ready to paste.

When it helps
after a discovery call, you want to give one lead the assessment link with her details already in.

Sourceproductised.ai · October 1, 2026 ↗

get_recent_activity

Returns a feed of recent submissions across all products, with contact details, product, segment, pipeline status and tags. The feed spans all the products of the workspace at once.

When it helps
you want one view of who came in recently, and through which product.

Sourceproductised.ai · October 1, 2026 ↗

get_response

Opens one response in full detail: all form answers, the AI outputs, tags, status and contact information. It is the full record behind one line of the activity feed or of a search.

When it helps
you want to read exactly what one respondent answered before talking to them.

Sourceproductised.ai · October 1, 2026 ↗

get_segment_breakdown

Shows how respondents segment by the persona or tier the AI determined for them, across products. The persona or tier comes from the product's AI, not from a manual label.

When it helps
you want the actual distribution of respondents between tiers rather than an impression.

Sourceproductised.ai · October 1, 2026 ↗

get_started

Displays the personalised welcome and getting-started panel, which Productised describes as useful on a user's first interaction. It shows a panel and leaves products and leads as they are.

When it helps
you just connected the account and want Claude to show you where to begin.

Sourceproductised.ai · October 1, 2026 ↗

get_talking_points

Produces ready-to-use conversation starters for a sales follow-up, generated from one end user's data. They are meant for a sales follow-up conversation, ready to use as they come.

When it helps
you are about to talk to someone who took one of your products and want an opening grounded in their answers.

Sourceproductised.ai · October 1, 2026 ↗

get_variables

Lists every token available in a product, form fields, node outputs, objects and custom values, with labels and syntax ready to copy.

When it helps
you want a result page to greet each respondent by first name and need the right token.

Sourceproductised.ai · October 1, 2026 ↗

get_welcome_html

Fetches a product's current custom landing page, which Productised says is so Claude can iterate on it. It returns the page currently attached to the product, code included.

When it helps
you want Claude to read the welcome page before proposing another wording.

Sourceproductised.ai · October 1, 2026 ↗

get_workspace_stats

Returns account-wide totals, products created, live products, responses and leads, plus ready-made links to settings, billing and subscription management, members and variables. Productised pairs it with get_plan_status to show value next to usage.

When it helps
you need the overall numbers of the workspace in one place.

Sourceproductised.ai · October 1, 2026 ↗

list_brain_sources

Lists every source Brain currently has in its knowledge base, so Claude can tell you what Brain is drawing on. Each source is a URL or a text, as Productised describes them.

When it helps
Brain's answers seem off and you want to check its sources.

Sourceproductised.ai · October 1, 2026 ↗

list_marketplace_templates

Browses public marketplace templates inside Claude as a visual gallery with filter chips for industry, GROW stage and search. Install and Preview buttons on each card lead to the marketplace page, opened in a new tab.

When it helps
you want to see what templates exist for a given industry.

Sourceproductised.ai · October 1, 2026 ↗

list_products

Lists all your AI products with their names, response counts, visibility and public links. The public URLs it returns are the ones to share for each product.

When it helps
you want to see which products exist in the workspace and which ones are public.

Sourceproductised.ai · October 1, 2026 ↗

list_result_templates

Shows the gallery of result-page designs, such as Dimensions, Radar, Persona or Narrative Report, to preview before choosing.

When it helps
you want the end screen of a quiz to look less like a report and more like a profile.

Sourceproductised.ai · October 1, 2026 ↗

list_themes

Lists all 20 visual theme presets with their name, description, primary colour and font. The names listed are the ones apply_theme expects when a preset goes on a product.

When it helps
you want to compare the presets before one is applied to a product.

Sourceproductised.ai · October 1, 2026 ↗

plan_product

Analyses a product idea in plain words and recommends an outcome type and funnel stage, shown as a blueprint card.

When it helps
you are torn between a scorecard and a diagnostic for the same audience and want a reasoned recommendation.

Sourceproductised.ai · October 1, 2026 ↗

preview_outbound_personalisation

Renders the welcome text, subtitle, custom HTML and starters twice, with sample details and with none, so you see both versions before anything is saved.

When it helps
you want to be sure a link without details still reads naturally.

Sourceproductised.ai · October 1, 2026 ↗

search_responses

Filters responses by date, segment, status, tags, keyword or contact details. Productised calls it the general-purpose find tool. Contact details count among the filters, so a name or an email works too.

When it helps
you remember something about a respondent but not their name.

Sourceproductised.ai · October 1, 2026 ↗

What Claude changes (38)

38 tools

Thirty-eight tools that build, edit, publish or delete. No source spells out their confirmation, so the general rule below applies.

add_brain_memory

Approval: see the rule

Stores a fact or preference in Brain's long-term memory, so its later answers take it into account.

What Claude asks for
no source describes a confirmation specific to this tool, so the general approval rule below is what covers it.
When it helps
you want Brain to keep a preference about your audience in mind in later answers.

Sourceproductised.ai · October 1, 2026 ↗

add_brain_source

Approval: see the rule

Feeds Brain's knowledge base with a new source, a URL or a block of text.

What Claude asks for
nothing specific to this tool is documented; the default approval rule applies.
When it helps
you published a long article on your method and want Brain to reason from it.

Sourceproductised.ai · October 1, 2026 ↗

add_mcp_connection

Approval: see the rule

Connects a product to an outside service, such as a CRM or a note-taking tool, so the product's AI can use that service's tools live during conversations with respondents.

What Claude asks for
the sources say nothing about a prompt for this tool; rely on the approvals section.
When it helps
a product's AI should use the tools of a service such as HubSpot or Notion during its conversations.

Sourceproductised.ai · October 1, 2026 ↗

add_node

Approval: see the rule

Inserts a node into a product's canvas, an AI step, a document builder, a design page or an advisor chat, and wires it into the flow.

What Claude asks for
no confirmation specific to this tool is published; the general rule holds.
When it helps
you want a downloadable action plan generated at the end of an assessment.

Sourceproductised.ai · October 1, 2026 ↗

apply_theme

Approval: see the rule

Applies one of the named theme presets to a product in a single call.

What Claude asks for
no source documents an approval for this tool; the section below holds the rule.
When it helps
a product needs another look and one of the presets fits it.

Sourceproductised.ai · October 1, 2026 ↗

clone_product

Approval: see the rule

Copies an existing product so it can serve as a template for a new niche or audience.

What Claude asks for
nothing tool-specific is published, so the default confirmation rule applies here.
When it helps
a product works for one audience and you want a version for another without starting over.

Sourceproductised.ai · October 1, 2026 ↗

create_page

Approval: see the rule

Adds a new HTML page to a product's design node, one more personalised screen at the end of the flow.

What Claude asks for
no source describes a prompt for new pages; the general approval rule covers it.
When it helps
you pasted a design exported from another tool and want it to become a result page.

Sourceproductised.ai · October 1, 2026 ↗

create_product

Approval: see the rule

Opens a new, empty product record; its conversation flow and prompts stay blank until generation runs.

What Claude asks for
nothing specific to product creation is documented, so rely on the section below.
When it helps
the plan for a new lead magnet is agreed and you want the record in place.
Watch out
Productised says a plan limit can block it and returns an upgrade link instead.

Sourceproductised.ai · October 1, 2026 ↗

delete_brain_item

Approval: see the rule

Removes a memory or a knowledge source from Brain.

What Claude asks for
the sources describe no confirmation specific to this removal; the default rule applies.
When it helps
Brain keeps relying on a source that is no longer valid.

Sourceproductised.ai · October 1, 2026 ↗

delete_node

Approval: see the rule

Takes a node off the canvas and cleans up the connections around it.

What Claude asks for
no source sets a prompt for this tool, so the general approval rule stands.
When it helps
an extra AI step slows the experience down and adds nothing respondents notice.

Sourceproductised.ai · October 1, 2026 ↗

delete_page

Approval: see the rule

Removes a single output page from a product's design node.

What Claude asks for
nothing specific to this deletion is documented; see the approvals section.
When it helps
you tested three result layouts and want to keep only the winning one.

Sourceproductised.ai · October 1, 2026 ↗

delete_product

Approval: see the rule

Permanently deletes a product along with all its responses, pages and digital card links.

What Claude asks for
Productised requires an explicit confirmation flag as a second step for this tool, and logs the call; Claude's general approval rule applies on top.
When it helps
a test product cluttering the workspace should go for good.
Watch out
its leads disappear with it.

Sourceproductised.ai · October 1, 2026 ↗

enable_outbound_personalisation

Approval: see the rule

Wires address-based personalisation into a product, adding tokens with fallback text to its content and recording which fields it accepts.

What Claude asks for
no confirmation specific to this tool is described; Productised advises previewing first, and the default rule applies.
When it helps
you want the welcome text to mention each prospect's company.

Sourceproductised.ai · October 1, 2026 ↗

generate_product

Approval: see the rule

Runs AI generation on a freshly created product to build its conversation flow, collection goals, system prompt and canvas workflow.

What Claude asks for
no source describes a prompt for generation; the general rule covers it.
When it helps
the empty record exists and you want a working first version to react to.
Watch out
a plan limit can stop it, Productised says.

Sourceproductised.ai · October 1, 2026 ↗

publish_product

Approval: see the rule

Sets a product to public and returns its live shareable link.

What Claude asks for
nothing specific to publishing is documented; the approvals section holds the rule.
When it helps
the product is ready and you need its public link.

Sourceproductised.ai · October 1, 2026 ↗

revise_product

Approval: see the rule

Runs generation again on an existing product from a revised prompt, replacing the whole workflow.

What Claude asks for
no tool-specific confirmation appears in the sources; the default rule applies.
When it helps
the first version missed the point and you would rather start the flow over than patch it.
Watch out
Productised calls it lossy and points to update_node for targeted edits.

Sourceproductised.ai · October 1, 2026 ↗

save_idea

Approval: see the rule

Saves a product idea straight to the Ideas tab of Brain.

What Claude asks for
no source documents an approval for saving ideas; the general rule covers it.
When it helps
during a brainstorm with Claude, a promising concept comes up and you do not want to lose it.

Sourceproductised.ai · October 1, 2026 ↗

set_agent_fields

Approval: see the rule

Configures the structured outputs of the product's AI agent, the score, tier, dimensions and insights it produces for each respondent; this is how scoring is set up.

What Claude asks for
nothing specific is published for this tool; see the section below.
When it helps
you want each respondent rated on four dimensions instead of one overall score.

Sourceproductised.ai · October 1, 2026 ↗

set_custom_branding

Approval: see the rule

Replaces the Productised footer badge with your own logo and click-through link, distinct from the chat-header logo.

What Claude asks for
no source sets a prompt for this tool; the default approval rule stands.
When it helps
an agency delivers assessments to clients and does not want another brand in the footer.

Sourceproductised.ai · October 1, 2026 ↗

set_custom_domain

Approval: see the rule

Configures a custom hostname for one product and returns the exact CNAME record to add at your registrar.

What Claude asks for
no confirmation specific to domains is documented; the general rule holds.
When it helps
you want the quiz to live at an address on your own site rather than on Productised's.

Sourceproductised.ai · October 1, 2026 ↗

set_design_system

Approval: see the rule

Saves workspace-wide design tokens that every brand-aware surface inherits, from new themes to result-page chrome, or clears them.

What Claude asks for
nothing tool-specific is described in the sources; the approvals section applies.
When it helps
a rebrand changed your colours and every product should follow at once.

Sourceproductised.ai · October 1, 2026 ↗

set_product_slug

Approval: see the rule

Names the path of a product under your workspace domain, so the address reads like a real page of your site.

What Claude asks for
no source describes a prompt here; the default approval rule covers it.
When it helps
a link ending in a random string looks odd on a printed brochure.

Sourceproductised.ai · October 1, 2026 ↗

set_submission_cap

Approval: see the rule

Caps how many submissions a product accepts; new sessions are blocked once the cap is reached, while conversations already under way finish.

What Claude asks for
nothing specific to caps is documented, so rely on the section below.
When it helps
you offer free audits to the first fifty respondents only.

Sourceproductised.ai · October 1, 2026 ↗

set_variable

Approval: see the rule

Adds or edits a workspace variable that every product can then use as a custom token.

What Claude asks for
no source documents a confirmation for variables; the general rule applies.
When it helps
your booking link changed and every result page should show the new one.

Sourceproductised.ai · October 1, 2026 ↗

set_webhook

Approval: see the rule

Adds, replaces or removes an outgoing webhook that pushes full response data to your URL on each submission.

What Claude asks for
nothing specific to webhooks is published; see the approvals section.
When it helps
each new lead should reach your own system the moment it comes in.
Watch out
Productised has no tool to read the current webhook; it shows in the product's settings.

Sourceproductised.ai · October 1, 2026 ↗

set_welcome_html

Approval: see the rule

Saves a custom HTML landing page on a product, the first screen respondents see.

What Claude asks for
no source sets a prompt for this tool; the default rule stands.
When it helps
Claude just drafted a sharper welcome page with you and it should go live on the product.

Sourceproductised.ai · October 1, 2026 ↗

set_workspace_domain

Approval: see the rule

Configures a hostname for the whole workspace, under which every product is served unless it has its own override.

What Claude asks for
nothing tool-specific is documented; the approvals section holds the rule.
When it helps
you want all assessments under one subdomain of your brand.
Watch out
checking status and unlinking happen in the Productised app, not here.

Sourceproductised.ai · October 1, 2026 ↗

tag_response

Approval: see the rule

Adds, removes or replaces tags on a response, such as hot lead, booked call or converted.

What Claude asks for
no confirmation specific to tagging appears in the sources; the general rule covers it.
When it helps
after a call, you want a respondent tagged booked-call or converted.

Sourceproductised.ai · October 1, 2026 ↗

update_branding

Approval: see the rule

Sets a custom logo and an optional click-through link for a product's chat header.

What Claude asks for
no source describes a prompt for branding; the default approval rule applies.
When it helps
a client wants their own logo at the top of the assessment you built for them.

Sourceproductised.ai · October 1, 2026 ↗

update_form_questions

Approval: see the rule

Replaces the whole set of intake questions of a product.

What Claude asks for
nothing specific to this tool is documented; the general rule holds.
When it helps
you no longer need the phone number and want a shorter form before the conversation.

Sourceproductised.ai · October 1, 2026 ↗

update_mcp_connection

Approval: see the rule

Adjusts an existing connection node: its server address, its token, or which of its tools are on or off.

What Claude asks for
no source documents an approval here; see the section below.
When it helps
your CRM token expired and the product's AI can no longer log leads there.

Sourceproductised.ai · October 1, 2026 ↗

update_node

Approval: see the rule

Patches the configuration of any canvas node, by its identifier or its type, for targeted edits.

What Claude asks for
nothing tool-specific is published, so the default confirmation rule applies.
When it helps
one AI step sounds too formal and only its instructions need softening.

Sourceproductised.ai · October 1, 2026 ↗

update_page

Approval: see the rule

Replaces the HTML, and optionally the CSS, of an existing result page.

What Claude asks for
no confirmation specific to page edits is described; the approvals section holds the rule.
When it helps
Claude reworked a result page with you and the new layout should replace the old one.

Sourceproductised.ai · October 1, 2026 ↗

update_product

Approval: see the rule

Patches a product's top-level settings: name, visibility and AI chat settings such as welcome message, button text, button colour and system prompt.

What Claude asks for
no source sets a prompt for this tool; the general rule applies.
When it helps
the button reads Start, but your audience reacts better to a label that promises their score.

Sourceproductised.ai · October 1, 2026 ↗

update_product_settings

Approval: see the rule

Locks submissions after a set number of responses and switches email notifications on or off.

What Claude asks for
nothing specific to these settings is documented; the default rule stands.
When it helps
notifications flood your inbox and you would rather read the morning briefing.

Sourceproductised.ai · October 1, 2026 ↗

update_response_status

Approval: see the rule

Moves a lead through your pipeline, from new to contacted, in progress, converted or lost.

What Claude asks for
no source describes a confirmation for status changes; the general rule covers it.
When it helps
you signed a client who came through an assessment and the pipeline should say so.

Sourceproductised.ai · October 1, 2026 ↗

update_seo

Approval: see the rule

Sets the page title, meta description, share card, favicon and keywords of a product's public page.

What Claude asks for
nothing specific to SEO settings appears in the sources; see the approvals section.
When it helps
the link looks bare when shared on a professional network and needs a proper card.

Sourceproductised.ai · October 1, 2026 ↗

upload_image

Approval: see the rule

Re-hosts an image, from a URL or encoded data, on Productised's own delivery network and returns a permanent link.

What Claude asks for
no source documents a prompt for uploads; the default approval rule applies.
When it helps
your logo sits on your website and should be reused on the product without downloading it yourself.

Sourceproductised.ai · October 1, 2026 ↗

Unclassified (1)

1 tool

Productised describes this tool, but no source says whether it saves anything. We leave it unclassified rather than guess.

generate_plan

Productised describes this tool as producing a full-funnel journey plan from Brain, with product recommendations at every stage. No source says whether that plan is stored in the workspace or only shown, so we do not classify it as reading or changing. The general approval rule covers it.

Approvals

What Claude asks before it acts

By default, Claude stops and asks for your go-ahead before any action it takes on an account on your behalf. The request shows up in the conversation at the moment it matters.

No source says whether Claude asks before a given Productised tool, so that default rule covers the thirty-eight changing ones. Productised adds a second step of its own only for deleting a product, which requires an explicit confirmation flag. Every call Claude makes is recorded in the workspace's API logs with the tool, parameters, result and origin. On Team and Enterprise workspaces, owners can restrict what a connector may do for the whole organization.

Plans

Which plans include it

None of the 819 sheets in the official directory shows availability by plan. The answer, connector by connector, is not published anywhere: a real gap in the catalogue.

The general rule is published: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile. On Team and Enterprise, an Owner or Primary Owner enables a connector for the organization before members can sign in. For the current state, check the Productised sheet in the official directory.

Limits

Where this connector stops

A connector is not an automation. Claude calls these tools while it answers you: nothing starts when a respondent submits, a lead turns hot or a cap is reached.

This connector has no Claude help article of its own; what we know comes from the directory sheet and from Productised's documentation of its Claude connector, including its pages on tools and on privacy and security. The tool list is an observed floor, not a promise: an administrator can open actions that no public sheet shows. The directory's verified badge is not a security audit, and Anthropic writes on every sheet that it does not control which tools a publisher exposes.

Need help

Need help connecting Productised to Claude?

A person reads every message.

FAQ

Questions about the Claude Productised connector

01What can Claude do with the Productised connector?
Claude can work in your Productised workspace from a conversation. It reads products, analytics, leads, result pages, plan status, credits and Brain, and it can plan, create, generate, edit, theme and publish products, personalise outbound links, set domains and branding, tag leads and move them through your pipeline. Productised's reference lists 73 tools for the connector, and one of them, generate_plan, stays unclassified here because no source says whether the plan it produces is saved.
02Can Claude create, publish or delete things in Productised?
Yes, thirty-eight tools change your workspace. Claude can create, clone, generate, revise and publish products, edit pages, nodes, forms and settings, set domains, webhooks and variables, and save to Brain. Deleting a product, with all its responses and pages, is possible but Productised requires confirm:true as a second step. Deleting a lead is not: Productised says that happens only in its app, deliberately. Every successful write returns a preview link and a message describing what changed.
03Does Claude ask me before it changes my Productised workspace?
By default, Claude stops and asks for a confirmation before any action it takes on an account on your behalf, and no source describes a rule for one Productised tool in particular. On its side, Productised adds a confirmation flag to product deletion only, and logs every call with its tool, parameters and result. On Team or Enterprise, owners decide whether members may skip some prompts, and can shut all changing tools.
04Which Claude plans include the Productised connector?
No official source publishes plan availability connector by connector, and none of the 819 directory sheets displays it. The general rule says remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile, with an Owner or Primary Owner enabling them first on Team and Enterprise. Separately, some Productised tools can be blocked by your Productised subscription's limits. The Productised sheet in the directory shows the current state for your account.
05Does Claude see all my leads and products?
It sees your workspace through your own access, and only that workspace. Productised says its screen asks for access to read products, leads and brand, that each API key is scoped to a single workspace, and that the connector cannot bypass workspace permissions: what your account cannot reach, it cannot reach either. What Claude reads enters the conversation and is processed by Anthropic under its privacy policy. On Team or Enterprise in Claude, an owner can also restrict the connector.
06How do I see what Claude did in Productised, or cut it off?
Productised records every MCP call in Settings, API, Logs, with the timestamp, the API key name, the tool, its parameters, the result and the origin of the call, such as Claude.ai, Claude Desktop or Claude Code. To end the access, Productised says the token can be revoked at any time from Settings, API, Connected apps. In Claude, the Disconnect button on the connector's row removes the connection on Claude's side.
07Should I use Claude or an automation tool for Productised?
They do different jobs, so it depends on the use. Claude with this connector suits work done in the moment: build this product, rework that page, show me the hot leads. An automation tool suits what must happen without anyone asking, such as pushing each new submission somewhere, because a connector never starts on its own. Productised's set_webhook tool, for its part, sends full response data to your URL on each submission.