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

Goes well beyond the annotations, disclosing that this is a stateful mode-switch that reroutes sibling tools to a separate replay journal, that stops/targets/liquidations are evaluated per candle, what stop does ('closes everything at the cursor'), and that an API key is required. Annotations only say it's a non-readOnly, non-destructive, idempotent write.

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?

Front-loaded with purpose, then modes in a logical order. Slightly dense and colon/semicolon-heavy with nested parentheticals, but every clause (speed meaning, cursor fields, stop behavior) carries real information.

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 stateful, no-output-schema tool this is nearly complete: it describes what status returns (cursor, price, progress, candles up to the cursor), what stop returns (summary), and the side effects on sibling tools. The only gap is that field-level detail of the returned journal/status objects is left implicit.

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 per-parameter docs already exist; the description adds relational semantics the schema lacks, clarifying that symbol and day must be supplied TOGETHER to trigger a start and that speed is market-seconds-per-real-second. It does not need to restate the enum, so 4 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?

Names a specific verb+resource ('replay one past UTC day on MarginPad's 1-minute candles') and states the core value prop ('backtest with the real routes'), which immediately distinguishes it from pure-query siblings like get_klines or paper_sim.

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 maps each invocation shape to a mode: symbol+day STARTS, no arguments reads status, act='stop' ends it. It also names the downstream siblings (paper_open/paper_positions/paper_close) and states the condition under which they switch to the replay journal - nothing is left to inference.

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