AutoRFP.ai — Claude Code MCP server: what it does, how to install
Provides AI assistants with read-only access to AutoRFP.ai projects, requirements, and content libraries. Helpful for drafting proposal answers directly inside chat.
Open on GitHubWhat AutoRFP.ai does
AutoRFP.ai hosts the server on its own infrastructure. You do not run any local processes. Connection endpoints depend on your data storage region: APAC (https://api.autorfp.ai/mcp), EU (https://api.eu.autorfp.ai/mcp), or US (https://api.us.autorfp.ai/mcp). Once configured, the client accesses workspace tags and project requirements alongside reference entries.
The connector operates strictly in read-only mode. It cannot add or alter records. Access controls match each person's existing account permissions. If a project is hidden from your user profile, the model cannot inspect it either.
Connecting requires an active AutoRFP.ai account and a paid subscription tier such as Claude Pro, Team, or Enterprise. Without valid workspace credentials, the OAuth authentication step will fail.
What it helps with
- search verified response libraries using semantic queries
- review pending proposal deadlines and unaddressed questionnaire items
- filter requirements by organizational compliance tags
- inspect workspace tag taxonomies without opening a web browser
Example prompts for the agent
- "List open RFP submissions that are due within the next fourteen days."
- "Search our verified answers for paragraphs explaining our GDPR data protection policies."
- "Show in-progress projects labeled Security that still have unassigned questions."
How to install AutoRFP.ai
1Template address
The server address is a template: replace {…} or […] with your own values
2Platform CLI: remote server
claude mcp add --transport http mcp https://{api_host}/mcp1Template address
The server address is a template: replace {…} or […] with your own values
2Platform CLI: remote server
codex mcp add autorfpai --url https://{api_host}/mcp1Template address
The server address is a template: replace {…} or […] with your own values
2Config file
{
"mcpServers": {
"autorfpai": {
"url": "https://{api_host}/mcp"
}
}
}1Template address
The server address is a template: replace {…} or […] with your own values
2Config file
{
"servers": {
"autorfpai": {
"type": "http",
"url": "https://{api_host}/mcp"
}
}
}1Template address
The server address is a template: replace {…} or […] with your own values
2Platform CLI: remote server
gemini mcp add -t http autorfpai https://{api_host}/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about AutoRFP.ai
How do I install AutoRFP.ai in Claude Code?
Run claude mcp add --transport http autorfpai https://{api_host}/mcp, substituting {api_host} with your region endpoint (api.autorfp.ai, api.eu.autorfp.ai, or api.us.autorfp.ai). After adding it, sign in to your AutoRFP.ai account to authorize permissions.
What access permissions does AutoRFP.ai request?
The server requests three read-only scopes: tags:read for organizational tags, projects:read for RFP projects and questionnaire items, and content:read for library search.
Can the assistant edit or delete items in AutoRFP.ai?
No, the connection is read-only. It cannot create, edit, or delete projects, requirements, or library content in your workspace.
