agent_start

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does deliver real behavioral context: the agent is persistent, production persistence depends on the separately managed Herdr service, and startup trust/auth prompts are reported but never auto-approved. It still omits failure behavior on the timeoutMs window and the lifetime/cleanup model, so it stops short of full disclosure.

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?

Three tight sentences: purpose first, routing second, operational caveat last. Every sentence earns its place, though the final caveat sentence is dense and could be split for scanability.

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?

For a resource-creating tool with no output schema, the description never indicates what 'start' yields (e.g., an agent identifier for use with agent_get/agent_prompt/agent_send_keys), which the agent needs to continue the workflow. Persistence and auth caveats are covered, but the handoff-to-other-tools information is missing.

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?

Schema description coverage is 100% and both enums (harness, effort) are documented in the schema, so the baseline is 3. The description adds no meaning beyond the schema for harness, cwd, model, effort, or timeoutMs.

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?

States a specific verb+resource ('Start a persistent interactive coding agent in a dedicated Herdr workspace') and immediately differentiates itself from the nearest siblings (exec, process_start). An agent can tell what this does without opening any schema.

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?

Explicitly routes usage: 'Use this, not exec or process_start, for coding-harness work' and names the conditions that select it (long-running, parallel, cross-turn Codex/agy/Claude tasks). When-not guidance is stated, not inferred.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.