Claude Code + Web search
Claude Code runs the work loop and Web search performs the selected operation. The Composer below preselects Claude Code, Claude Opus 5, and this tool so you can test one real task with the same billing and recovery path as any AgentSky run.
How the combination works
Claude Code decides when to call Web search.
Claude Code runs the coding loop, reads and edits repository files, invokes tools, and reports the result from its cloud workspace.
Use Exa search to find public URLs for a question, and Exa contents to retrieve the text of a known page. A useful research task combines both: discover relevant sources, read the strongest ones, and produce a comparison whose claims link back to those sources.
Claude Code
Owns the task loop, context, model calls, tool choice, cloud computer, progress and final result.
Web search
Performs the operation exposed by the current Web search capability. Its input limits and output shape come from the live tool guide and adapter.
AgentSky
Keeps the selected agent, model, capability grant, task state and usage record together. Tool charges and model or runtime charges remain itemized separately.
Setup and first task
Start with one bounded web search task.
The Composer carries Claude Code, Claude Opus 5, and Web search. Begin with the smallest input that lets you inspect the output before expanding the workflow.
Open the Claude Code Composer
The Composer above preselects Claude Code, Claude Opus 5, and Web search. Add the target, required inputs and a concrete expected result; the draft stays in place through sign-in.
State the Web search boundary
Find the official documentation for browser-based AI agents. Compare three approaches and cite the pages supporting each claim.
Run and verify the returned result
Check the output against the source or brief, confirm that the result has the expected format, and inspect the task usage for model, runtime and tool charges.
Inputs, cost and limits
What Claude Code + Web search establishes
Repository work where Claude Code's coding workflow and supported Claude models fit the team's existing review process.
Configure and verify
- Claude Code and Claude Opus 5 are preselected for the browser evaluation.
- Web search is enabled from the current capability catalog with its live input and output guidance.
- The first task can be checked from the returned artifact, text or provider result before a larger workflow is attempted.
Still depends on the task
- A tool guide does not guarantee a provider result for malformed, blocked or unsupported Web search inputs.
- A successful tool call does not establish that the result is correct for your business decision; inspect the output at its source boundary.
- Total task cost can include model tokens, active computer time and the selected tool's billable units.
Current Web search rates
exa.search
$0.01 per run.
exa.contents
$0.002 per page.
Tool-specific details
Use the current Web search contract.
Use Exa search to find public URLs for a question, and Exa contents to retrieve the text of a known page. A useful research task combines both: discover relevant sources, read the strongest ones, and produce a comparison whose claims link back to those sources.
Search input
query is required. limit defaults to 8; the agent catalog accepts 1–20 results.
Page retrieval
Provide one public URL. The current Exa adapter requests page text and returns url, content and provider.
Task example
Research three alternatives to a named product. Return a table of feature differences, supporting page URLs and unresolved questions.
Common questions
Related resources