Codex + Browser use
Codex runs the work loop and Browser use performs the selected operation. The Composer below preselects Codex, GPT-5.6 Sol, 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
Codex decides when to call Browser use.
Codex owns the coding-agent loop: it explores the repository, changes files, runs commands, and returns evidence from the cloud workspace.
Use browser-backed retrieval for a JavaScript-heavy directory, product listing or application page that a plain fetch cannot read. Give the agent the starting URL and the fields you need, such as product name, visible availability and source link. These capabilities are classified as read operations in the agent catalog.
Codex
Owns the task loop, context, model calls, tool choice, cloud computer, progress and final result.
Browser use
Performs the operation exposed by the current Browser use 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 browser use task.
The Composer carries Codex, GPT-5.6 Sol, and Browser use. Begin with the smallest input that lets you inspect the output before expanding the workflow.
Open the Codex Composer
The Composer above preselects Codex, GPT-5.6 Sol, and Browser use. Add the target, required inputs and a concrete expected result; the draft stays in place through sign-in.
State the Browser use boundary
Open the public directory URL below. Extract the names, descriptions, and links into a table without submitting any forms.
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 Codex + Browser use establishes
Software engineering tasks that benefit from Codex's repository navigation, command execution, and iterative implementation loop.
Configure and verify
- Codex and GPT-5.6 Sol are preselected for the browser evaluation.
- Browser use 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 Browser use 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 Browser use rates
tinyfish.fetch
$0.004 per page.
tinyfish.browser
$0.03 per reserved session.
Tool-specific details
Use the current Browser use contract.
Use browser-backed retrieval for a JavaScript-heavy directory, product listing or application page that a plain fetch cannot read. Give the agent the starting URL and the fields you need, such as product name, visible availability and source link. These capabilities are classified as read operations in the agent catalog.
Two operations
tinyfish.fetch retrieves a page; tinyfish.browser accepts a URL plus a task, defaulting to extracting readable content.
Browser session
The adapter reserves 60 seconds and returns session connection details. It does not forward the task string or complete the browser task by itself.
Page retrieval
The fetch operation returns page content. The browser operation creates a session that needs further browser interaction; they are distinct results.
Common questions