Start an agent on a task; return the answer or a session handle to wait on. agent from list_agents.
| Name | Required | Description | Default |
|---|---|---|---|
| task | Yes | ||
| agent | Yes | ||
| max_steps | No | ||
| max_time_s | No | ||
| idempotency_key | No |
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description bears full responsibility for behavioral disclosure. It only states the basic action and outcome, without mentioning potential side effects, resource usage, authentication needs, or error conditions.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences long, immediately stating the purpose and key outcome. Every word adds value, and it is front-loaded for quick comprehension.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Despite having an output schema, the description fails to explain when to expect an answer versus a session handle, how parameters like max_steps or max_time_s affect behavior, or how this tool interacts with others like wait_for_session.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description adds meaning only for the 'agent' parameter by referring to list_agents. The other four parameters (task, max_steps, max_time_s, idempotency_key) are not described, and schema coverage is 0%, leaving their purpose unclear.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses the verb 'start' with the resource 'agent on a task', which clearly indicates the action. It also mentions returning an answer or session handle, and references 'agent from list_agents', which distinguishes it from sibling tools like list_agents and wait_for_session.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implicitly suggests use after list_agents by referencing it, but does not explicitly state when to use this tool versus alternatives like send_message or wait_for_session. No exclusions or prerequisites are provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Each tool targets a distinct resource and action: session lifecycle (run, cancel, wait, message, share), file operations (list, read, write), and agent discovery. No two tools have overlapping purposes; even the session-related tools are clearly separated by their verb.
All tool names follow a consistent verb_noun snake_case pattern (cancel_session, list_files, run_agent, etc.). The pattern is uniform across session, file, and agent operations, making the API highly predictable.
With 9 tools, the server is well-scoped for a computer-use agent API. It covers agent discovery, session lifecycle, messaging, sharing, and file access without unnecessary bloat, fitting comfortably in the ideal 3-15 range.
The tool set covers the core workflow: list agents, run them, wait for results, send follow-ups, cancel, share, and access files. Minor gaps exist (e.g., no list_sessions or delete_file), but they do not block the primary use case.