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

The annotations only convey readOnlyHint=false and idempotentHint=false, which signal this is not a read-only operation but don't explain side effects. The description adds meaningful behavioral context: it returns immediately with an id and status, it costs the same capacity as a regular backtest, compute is identical, and it skips trade simulation. These behavioral traits are beyond what annotations provide. Minor gap: it doesn't explicitly state what happens to prior confirmation sources or whether it creates/overwrites state, but the 'returns immediately with an id and status' description covers the essential async behavior.

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 substantive yet tight: three sentences that front-load the core purpose, then explain the follow-up workflow and capacity/cost implications. Every sentence earns its place—no filler, no repetition of schema details. The concrete example at the end adds value without bloating the text.

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 tool's complexity (it has a rich DSL schema and an async workflow), the description is complete: it tells the agent the tool is not a backtest, explains what it returns (id and status), how to monitor it (get_confirmation_source), and how to use the result in submit_backtest. The output schema is not present, but the description covers return behavior ('returns immediately with an id and status') sufficiently for an agent to proceed.

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 schema already provides very thorough parameter descriptions (e.g., assetPair format, date formats, DSL example, fee semantics, indicator function list with caveats). The description adds workflow-level meaning by clarifying that this tool accepts the same strategy DSL as submit_backtest and that riskManagement is accepted but never used. That extra context—'not a backtest itself' and the caveat about riskManagement—helps agents understand parameter behavior beyond what the schema states.

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 uses a specific verb ('Runs a strategy against historical data purely to produce a signal timeline') and clearly defines the resource ('confirmation source'). It distinguishes this tool from a backtest ('not a backtest itself') and explicitly names the sibling tool it corresponds to (submit_backtest). The contrast is explicit, not implied.

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 explicitly states when to use this tool: 'to confirm OTHER backtests — not a backtest itself.' It explains the workflow ('call get_confirmation_source to check completion, then pass that id as a confirmationSources sourceId in submit_backtest') and gives a concrete example ('only count an XRP entry once a BTC-neutral confirmation source agrees'). It also states what does NOT apply here (no trade, no PnL, no WinRate/drawdown), which is a useful exclusion.

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