API & MCP

Available on Pro and Business.

API keys

Create a key in Settings → API keys. Keys start with fb_live_ and have scopes, for example conversations:read, issues:read and issues:write. Send the key as a bearer token:

curl https://api.feedbotai.com/v1/bots \
  -H "Authorization: Bearer $FEEDBOT_API_KEY"

REST endpoints

Method	Path	Description
GET	/v1/bots	Your bots.
GET	/v1/conversations	Search conversations. Filters: botId, from, to, tag, category, minPriority, q.
GET	/v1/conversations/{id}	Messages and analysis.
GET	/v1/insights	Insight report for a period.
POST	/v1/insights	Generate a new report.
GET	/v1/issues	Product issues grouped from conversations.
GET	/v1/issues/{id}	One issue with quotes and linked conversations.
PATCH	/v1/issues/{id}	Update status (open, in_progress, fixed), note, external ref.
GET	/v1/reports/agent.md	Agent-ready report.
GET	/v1/stats	Counts for the dashboard overview.

MCP server

The MCP server uses Streamable HTTP and the same API keys.

Claude Code

claude mcp add --transport http feedbot https://mcp.feedbotai.com/mcp \
  --header "Authorization: Bearer $FEEDBOT_API_KEY"

Codex

~/.codex/config.toml

[mcp_servers.feedbot]
url = "https://mcp.feedbotai.com/mcp"
bearer_token_env_var = "FEEDBOT_API_KEY"

Other MCP clients work the same way: HTTP transport, the URL above and the Authorization header.

Tools

Tool	What it does
list_bots	List your bots.
search_conversations	Search conversations by text, date, tag or priority.
get_conversation	Messages and analysis for one conversation.
list_product_issues	Open issues, sorted by frequency.
get_issue	One issue with quotes, pages and acceptance criteria.
mark_issue_fixed	Mark an issue as fixed, with a note.
get_insight_report / generate_insight_report	Read or generate an insight report.
get_stats	Conversation, lead and issue counts.

Example prompts

Show me the top 5 open Feedbot issues this week and which pages they affect.

Fix issue #42, run the tests, then mark it fixed in Feedbot with a short note.

Closing the loop

When an issue is marked fixed (by the API, MCP, the CLI or in the dashboard):

• the bot stops working around the problem in its answers;
• if you enable it, users who reported the issue and left a contact get a “we fixed it” message.