Agent and Tools
Use search, files, terminal, notes, and task collaboration tools.
How to Make Hackcat Search the Web?
Specify the topic, source, and time requirements clearly so that the search results can be used for the current task.
What is the difference between cloud environment and local environment?
Choose based on files, tools, network location, and execution boundaries.
How to Make Hackcat Read a Specific Webpage?
Provide the complete public URL and specify the content that needs to be verified.
How to Search for Recent Information or Specific Version Details?
Limit searches by both time range and applicable software version.
Are search results with source links necessarily reliable?
Assess evidence quality based on link relevance, main content, and date.
The website opens in the browser, so why can't Hackcat read it?
Distinguishing login status, access verification, network location, and page structure.
How to Assign Large Tasks to an Agent Without Letting It Go Off Track?
Define the scope, deliverables, and milestone checkpoints.
How to Choose the Agent's Permission Mode?
Decide which actions require approval based on the task environment and operational risks.
What Should Be Checked When Seeing an Operation Approval Card?
Confirm the actual action and scope before allowing or rejecting.
Does "Acting on Behalf Approval" Automatically Allow All Actions?
Understand the role of automatic review and situations where your decision is required.
What Does Full Access Allow the Agent to Do?
Confirm the execution environment and acceptable modification scope before reducing step-by-step approvals.
How to Confirm Which Computer the Agent Is Using When Multiple Computers Are Connected?
Verify the environment name before sending, and confirm with the actual path and output after execution.
How to Make Agent Read a Specified File?
Provide the correct file and question to make the reading scope easy to confirm.
How to Have Agent Modify Files and Confirm Correct Changes?
Define the changes first, then verify the results; check existing content before retrying after failure.
How to Have Agent Analyze Large Files in Segments?
Narrow down materials by structure, keywords, and range while preserving traceable evidence.
How to View Screenshots or Images in the Agent Environment?
Enable tools to truly read images and distinguish between preview and analysis failures.
How to Download Reports and Result Files Generated by Agent?
Clearly request delivery of attachments and confirm file generation completion before downloading.
What to Do If Report Is Generated but Attachment Download or Delivery Fails
Check the generation status, actual path, and connection, then redo the delivery.
How to Make Agent Execute Commands in the Environment?
Clearly define the target environment, working location, and the results you need to see.
Command Still Running: Should You Keep Waiting or Restart?
Waiting for a return does not mean the process has ended; first check the original execution session.
Can the Agent Continue Operating While the Terminal Program Waits for Input?
Distinguishing between interactive sessions and run sessions that only allow viewing output.
What is the difference between background running and ordinary long tasks?
Background processes do not necessarily provide a terminal session for continuous interaction.
How to Determine if a Tool Has Truly Executed Successfully?
Check the exit status, actual output, and generated artifacts simultaneously.
Can Agent Open Webpages, Click, and Take Screenshots?
Perform explicit page operations in environments that support browser tools.
How to Make Hackcat Save Notes Usable Across Tasks?
Organize stable information into titled personal notes.
How to Find, Update, or Delete Existing Notes?
First locate the original record, then modify it to avoid duplicates and contradictions.
Where to Disable Notes if You Don't Want Tasks to Use Them?
Control subsequent saving and usage, distinguished from deleting existing notes.
How Does the Agent's To-Do List Help Track Tasks?
Use status to check progress; do not equate list completion with result verification.
Can Agents Delegate Work to Subtasks?
When the feature is available, use clearly bounded subtasks to handle independent issues in parallel.
How to Check Subtask Status and Use Its Results?
Distinguish between queued, working, completed, failed, and canceled statuses, then integrate by the main task.