claude_agent

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

No annotations are provided, so the description carries the full burden. It discloses auth checks before session creation, injection of sub-agent self-awareness and delegation lineage, addition of launch-specific Stop hook settings, and the ownership model of aiterm. It also documents completion exit codes (0/3/4) and outcome semantics, going far beyond basic mutation traits.

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 long but information-dense; each sentence covers a distinct aspect (launch mode, config sharing, delegation, auth, completion handling, result retrieval). It is front-loaded with purpose, though a reader must parse a dense paragraph; slight structural organization could improve scannability, so 4.

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?

Given the complexity (7 optional params, output schema, multiple sibling tools), the description is exceptionally complete: it covers preconditions, behavioral traits, completion signals, timeout semantics, and alternative retrieval methods. The output schema exists, so return-value details are not required.

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 input schema has 100% parameter coverage with clear descriptions (e.g., prompt states '送信後は待たずに即返る'), and the description adds minimal parameter-specific detail beyond referencing the wait_command tied to the launch response. Thus the baseline 3 is appropriate; the schema carries the parameter documentation burden.

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 opens with a specific verb+resource: 'Claude Code (Anthropic) の対話エージェントTUIを永続端末に起動する' (launch the Claude Code conversational agent TUI in a persistent terminal). It distinguishes itself from `claude -p` and from sibling agent tools by specifying Claude Code and the persistent-session model.

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?

Clearly explains the fire-and-forget pattern: '起動して投げたら投げっぱなしでよい＝親はここで待たない' and advises not to poll or run in foreground. Provides explicit alternatives for completion notification (wait_command or aiterm-wait) and result retrieval (pty_read(agent_transcript:true) or claude_turn). Also gives when-not guidance by rejecting /login and /logout in correlated sessions.

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