AI-Adjacent Issue
Agent Keeps Adding Tool Dependencies
An agent keeps adding required tools, connectors, searches, files, APIs, or integrations instead of completing the task with the available surface.
What This Looks Like
An agent starts a task but keeps adding new required tools, connectors, searches, files, APIs, or integrations before it will complete the work. The user may ask for a bounded result, but each pass introduces another dependency the agent says it needs.
Why Users Blame AI
The user experiences the agent as refusing to finish or making the task bigger. The underlying cause may be task planning, tool-routing behavior, insufficient input boundaries, unavailable capabilities, or an agent loop that treats every uncertainty as a reason to add another tool dependency.
What to Check First
- Whether the task actually requires the added tool
- Whether the agent has a clear done condition
- Whether the prompt separates required tools from optional checks
- Whether unavailable information has a declared fallback
- Whether the agent is overusing search, retrieval, or connector calls
- Whether tool availability differs by workspace, mode, or runtime
When This Is AI-Adjacent
Use this AI-Adjacent Issue when the visible problem is tool dependency growth around an agent task. If the agent keeps expanding the task itself, use the related Workbench Issue for task expansion. If the agent repeatedly calls tools beyond what the task permits, use the related Issue for too many tool calls.
Related Workbench Issues
Agent Keeps Expanding the Task
The agent repeatedly expands the task, plan, scope, or next-step list instead of completing the declared work.
Too Many Tool Calls
The AI or agent makes more tool calls, searches, retrievals, API calls, or integration actions than the task requires or permits.
Task Has No Clear Limit
The task does not declare where the AI should stop, what is out of scope, or what counts as enough work.
Tool Exists but Required Inputs Are Missing
A usable tool or integration exists, but the AI or agent does not have the required inputs, permissions, fields, identifiers, or context needed to call it correctly.
Agent Never Settles on Final Answer
The agent keeps revising, rechecking, planning, or branching instead of converging on a final answer or completed result.
Common Ways People Describe This
- agent keeps adding tool dependencies
- AI agent says it needs more tools
- agent keeps asking for integrations
- AI task keeps requiring more tools
- agent will not finish without another tool
- tool dependency keeps growing