Claude AutoRFP.ai connectorWhat Claude can do in your AutoRFP.ai account.
The Claude AutoRFP.ai connector exposes 7 tools, all 7 for reading. 0 write to your workspace and 0 are undocumented. Here: how Claude searches your approved answers, checks where each RFP stands, and why drafting back into AutoRFP.ai is not on the menu yet.
Verified Trustpilot reviews · AI, automation & growth agency
What changes when Claude can read your RFP library
With this connector, Claude reads your AutoRFP.ai library and projects directly, so the draft it writes in the chat starts from content your team already reviewed. It reaches your live RFP, DDQ and security questionnaire projects as well as the AI library, and it needs an AutoRFP.ai account to work at all.
Answer from approved wording. Ask for your approved statement on single sign-on and search_content runs a semantic search across the library, filtered by tag if you want, so Claude starts from wording your team already approved.
See what is still open. list_projects shows your RFP, DDQ and questionnaire projects by status or due date, and list_requirements pulls the questions inside one project, for instance only those without an answer yet.
Make a faster bid or no-bid call. AutoRFP.ai describes pasting an incoming RFP into the chat and asking how much your library already covers; Claude estimates the coverage and flags the questions with no good match.
The other side of the ledger: the published tools do not write. No project gets created, no answer gets pushed back, nothing is edited or deleted in AutoRFP.ai. Claude sees what your own role allows. And nothing runs by itself, since no tool reacts when a new RFP lands. Event-driven flows are the job of automation platforms, covered from the Integrations hub.
Five words before you plug it in
The vocabulary around this connector, quickly.
- Connector
- The link you set up once between Claude and an account you already have, so Claude can work in it while it answers you.
- Tool
- One named action a connector opens to Claude. Claude picks which tools to call by itself, and the directory lists every one by name.
- Authorization
- The service's own sign-in and consent screen, where you hand Claude the access it will use. Granted per person and revocable.
- Approval
- The confirmation Claude waits for before an action that changes an account, shown right in the conversation when it applies.
- MCP
- The common standard connectors are built on: it lets an assistant like Claude talk to an outside service such as AutoRFP.ai.
Connect AutoRFP.ai to Claude in three steps
- 01
Look it up in Claude
Open Claude's settings, go to Customize, then Connectors, and find AutoRFP.ai in the list. On a Team or Enterprise workspace, an Owner or Primary Owner has to enable it before each member can sign in.
- 02
Sign in with your own login
Click Connect on its row and log in to AutoRFP.ai in the window the service opens. Each person signs in with their own credentials. If the link breaks later, Disconnect and connect again.
- 03
Read the consent screen
Before accepting, check the scopes AutoRFP.ai requests: tags:read, projects:read and content:read, with no write scope. That screen belongs to the service, not to Claude, and it sets what the access covers.
The 7 tools, every one of them read-only
AutoRFP.ai gives Claude 7 tools: 7 that read your account, 0 that change something in it.
Library, projects, questions, tags: Claude can look at all four and change none. Names stay as Claude displays them.
- 7 read
What Claude reads (7)
7 toolsSeven tools that query your AutoRFP.ai workspace and leave it as it was.
get_project
Opens one project in full: its header, how many requirements sit in each status, and the tags applied to it. Claude gets the snapshot it needs to tell you how far along the response is.
list_content
Walks through your content library and returns records filtered by tag, file type or date range. It is a browse, not a semantic match, so you see what exists in a slice of the library.
list_content_usage
Works backwards from one piece of content to the projects and requirements that reused it, and tells whether it was inserted automatically or by hand.
list_projects
Brings back your RFP, DDQ and questionnaire projects, narrowed by status, due date or tag. Claude can then sort out what is urgent and what can wait.
list_requirements
Pages through the questions and answers inside a single project. Filters cover status, tags, who updated an answer last, and whether an answer exists at all.
list_tags
Returns your organisation's full tag vocabulary. AutoRFP.ai exposes it so Claude can use the right tags when it filters the other calls, rather than guessing at labels.
search_content
Runs a semantic search across your approved content library, optionally filtered by tag, so a question finds the right answer even when the wording differs.
What Claude asks before it acts
By default, Claude pauses and asks for confirmation before each action it takes on an account for someone, inside the conversation.
On Team and Enterprise, workspace owners decide whether a member may let some actions through without a prompt each time. They can also cap what a connector may do for the whole organization, for example keeping reading and closing writing, and that choice binds everyone. Claude also works with the rights of the person who connected and no more. AutoRFP.ai adds its own layer: each user signs in with their own credentials, and the assistant inherits that user's role-based permissions.
Is it included in your plan?
None of the 819 sheets in the official directory shows plan availability. Nobody publishes that answer connector by connector.
The public rule: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile, and on Team and Enterprise an Owner or Primary Owner enables a connector before members connect. AutoRFP.ai once published a different line, compared in the limits below. For the live state, check the AutoRFP.ai sheet in the official directory.
Where the connector stops
A connector is not an automation. Claude calls these tools while it answers you, so nothing starts when an RFP arrives or a due date gets close.
By default the sheet exposes reading only. That is an observed floor, not a promise: an administrator could open actions no public sheet lists, and AutoRFP.ai says write tools are on its roadmap. The partner badge is not a security audit, and Anthropic notes on every sheet that it neither chooses a publisher's tools nor guarantees how they behave. Only AutoRFP.ai documents this connector. Similar rules apply on the Claude affinity connector page and across the directory.
Which Claude plans can use AutoRFP.ai?
- Claude help center, general connector rule ↗September 2026Remote connectors are open to all users; on Team and Enterprise, an Owner enables each one first.
- AutoRFP.ai launch article ↗May 2026Out of dateThe publisher said custom connectors need Pro, Max, Team or Enterprise, and that the free plan would gain access once AutoRFP.ai joined the directory.
What this page followsAutoRFP.ai's line predates the connector's arrival in the directory in July 2026 and itself announced the change, so it is outdated on this point. Anthropic's general rule is the current one, but neither text speaks for your account: check the directory sheet while signed in.
Need help connecting AutoRFP.ai to Claude?
A person reads every message.
