Skip to content
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions features/agent.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,7 @@ Whether you need one data point or entire datasets at scale, Firecrawl `/agent`

**Think of `/agent` as deep research for data, wherever it is!**

Agent builds on everything great about `/extract` and takes it further:
Agent replaces the legacy `/extract` endpoint and takes it further:

- **No URLs Required**: Just describe what you need via `prompt` parameter. URLs are optional
- **Deep Web Search**: Autonomously searches and navigates deep into sites to find your data
Expand Down Expand Up @@ -469,9 +469,9 @@ In `skip` mode there's no pending approval to answer. After the terms are accept
| `exchange.decline` | object | No | `{ approvalId }`: declines the previous turn's `pendingApproval`. Needs `threadId` |
| `exchange.onTermsRequired` | string | No | What to do when an Alexandria data provider the agent would use needs terms your team has not accepted: `skip` (default) or `ask`. See [Data providers that need terms](#data-providers-that-need-terms) |

## Agent vs Extract: What's Improved
## Agent vs legacy /extract: What's Improved

| Feature | Agent (New) | Extract |
| Feature | Agent | Legacy /extract |
|---------|-------------|---------|
| URLs Required | No | Yes |
| Speed | Faster | Standard |
Expand Down
Loading