Plan or execute an adaptive verified task

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

Goes well beyond the annotations (destructiveHint/readOnlyHint/idempotentHint) by disclosing that mutations require allowMutations=true, that an identical mutating action is never retried, that verification is delegated to read-only assert-state and only counts on explicit boolean truth, that recovery runs only through named recoverWith branches, and that no LLM guesses omitted args. This is exactly the mutation/retry/permission context the annotations cannot carry.

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?

Purpose is front-loaded and the flow is logical, but the definition is very dense and includes a full Signature block that restates the input schema, plus metadata lines (Phase:, cost=, Produces:, Safety:) that partly echo structured fields. Every sentence is informative, yet the size exceeds what is needed and hurts 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?

For a highly complex nested-input tool with no output schema, it covers modes, budgets, recovery, mutation gating, and even summarizes the return payload (evidence timeline, consumed budgets, confidence, continuation plan). Near complete, though the absence of an output schema means return-shape detail is only sketched rather than specified.

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 coverage is 100%, so the schema already carries field-level semantics, but the description adds real value beyond it: $steps.<id>.<path> reference syntax, the no-guessing omission behavior, and the caller-supplied recovery branch model. The embedded signature largely duplicates the schema, limiting it below 5.

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 precise verb-set: turn a goal into a plan, preview an explicit typed workflow, or execute it one tool at a time. It names the resource (schema-aware workflow) and its three modes, letting an agent distinguish it from raw executors like execute or batch-execute.

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?

It states the mode semantics ('Plan ranks tools, preview validates explicit steps, execute invokes tools') and cross-references assert-state for verification and explain-failure for diagnosis. However it never explicitly says when to prefer this over siblings such as run-loop, execute-and-wait, or playbook-run, so routing guidance is implied rather than excluded.

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