workflow_plan

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

With no annotations provided, the description carries the full burden and discloses several behaviors: it calls a decomposer model via runAgent with 429 backoff and multi-model rotation, performs structural self-checks (cycle detection, stage count limit), and sets approval_required based on complexity. It also clarifies it returns a plan and does not create tasks, which is valuable beyond what the schema provides.

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 a single dense paragraph covering many details (decomposition, self-checks, approval gates, runAgent behavior). While every sentence carries useful information, it is not front-loaded with a simple summary and may overwhelm an agent. The core purpose is stated first, but the length reduces scannability.

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?

The description explains the tool's main flow, the return of an executable plan for human review, and the approval mechanism. It even mentions the plan structure (stages with title/description/criteria/agent/depends_on), which is helpful since there is no output schema. It could mention error handling, but for a planning tool it is largely complete.

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%, so parameters are already well documented. The description adds context about the 'goal' parameter ('一句话目标') and implies the 'approve_threshold' concept through the approval gate logic, but it does not go into per-parameter semantics beyond the schema. Baseline 3 is appropriate.

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 'Orchestrator automatic decomposition (L1)' and clearly states it takes a one-sentence goal and produces a structured multi-stage DAG plan. It explicitly says it does not directly create tasks, which distinguishes it from task_create, and mentions it can connect to workflow_start, differentiating it from execution-focused tools.

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 says it is for turning a one-sentence goal into a plan, and notes that small tasks can auto-approve and connect to workflow_start, implying this tool is the planning step before execution. It also states it does not create tasks, which helps rule it out when task creation is needed. However, it does not explicitly name alternatives or state when not to use it.

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