万忆反事实之镜

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

With no annotations supplied, the description carries the full disclosure burden and does it well: it reveals automatic triggers (auto-open on interception), deferred behaviors (auto-settle when due, auto-verdict determination), and cross-tool side effects (counter_won auto-sinks into the error notebook and reinforces confidence interception). It omits output/return behavior and error handling, but the core behavioral traits are disclosed.

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 core mechanism is front-loaded, and each subsequent clause conveys distinct information: trigger condition, manual actions, verdict values, and downstream effects. The opening self-reference ('护城河#2「反事实之镜」') is mild filler, and the single dense run-on sentence is harder for an agent to parse than structured sentences, but nothing is wasted.

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 tool with 13 parameters and 6 actions but no annotations and no output schema, the description covers the open→settle→verdict lifecycle and downstream effects well. However, it omits the 'get' action that appears in the schema's action enum, says nothing about return values or list output shapes for list_open/list_settled, and provides no error or edge-case guidance—gaps an agent cannot fill from structured data.

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 every parameter is already documented in structured form and the baseline applies. The description adds workflow-level semantics (risk_level drives auto-generation of counter_path and settlement days; settle auto-judges verdict from outcomes), but it largely restates what the schema already says rather than adding meaning beyond it.

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 states a concrete mechanism: opening parallel fact/counterfactual branches at decision points and auto-settling them for comparison. It enumerates the sub-actions (open/settle/list_open/list_settled/auto_check_due), the verdict outcomes (fact_won/counter_won/neutral), and the workflow linkage, which clearly distinguishes it from the sibling memory/experience/confidence 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 gives clear trigger conditions: automatic branch opening on every BLOCK/CAUTION interception, plus manual open at key decision points, and explicit purposes for list_open/list_settled/auto_check_due. It does not name sibling alternatives or state when not to use the tool, which would push this to a 5.

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