ClearCo agents operate under a human-in-the-loop model, meaning the agent only drafts ideas and gathers information while you make the final decisions and approve all changes.
Depending on your platform, configuring custom agents requires IT Admin, User Admin, or Super Admin permissions. To use an agent, users must be assigned to the right ClearCo security group or ClearCo Learning group.
Approve all changes before they happen
The system operates under a human-in-the-loop model where the AI drafts information and you make the final decisions.
- Before making any change, the agent will show you a clear summary of what it wants to update.
- If you are changing many files at once, the agent will show you exactly how many files will be affected.
- For major changes that cannot be undone, the system will ask you to confirm your choice.
Protect hiring decisions and candidate data
Recruiting agents follow strict rules to prevent bias and accidental candidate communication.
When the AI agent shows you past notes from teammates, it shows the exact notes as written without altering them.
The agent does not send rejection emails. However, if you instruct the agent to move a candidate to a disposition stage, any automated emails set up in your system for that stage will trigger as usual.
The agent can discuss job roles and candidate records across your whole system, allowing you to easily search and compare talent.
NOTE
System prompts include built-in bias protections to help keep recommendations fair and objective.
Keep performance reviews fair and accurate
Performance agents keep your data safe while you work.
- The agent cannot save review forms for you. You must manually click Save as Draft or Submit Review to save or finalize your work.
- When the agent looks at employee goals, notes, or feedback, it can only read them. It cannot change them.
- You have step-by-step approval. The agent shows you the exact text and the question it answers before filling out any part of a review form.
Keep company knowledge private and accurate
Custom agents answer questions using only the files you upload. They do not connect to the main ClearCo databases or view live employee records. When the agent gives an answer, it points to the specific page of the document it used so employees can double-check the source.
Chats are saved in Threads for as long as your account is active so you can return to past answers whenever you need them.
Enforce strict behavioral boundaries
For custom ClearCo agents, Policy settings give your organization strict control over what topics the AI can discuss so your employees receive safe and objective guidance.
- Set knowledge limits that force the agent to admit when it lacks an answer instead of guessing.
- Turn on safety rules that automatically route sensitive compliance questions directly to human experts.
- Communication tools scan the text before anyone sees it to ensure the agent uses neutral, unbiased language.
Use your company voice
Personality settings give you direct control over how custom agents represent your organization. You can choose a strict tone for an agent handling compliance rules or a friendly tone for an onboarding assistant. This ensures the agent always matches your corporate values and speaks exactly how you want it to.
Comments
Please sign in to leave a comment.