rsi_impact

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

With no annotations, the description carries the full burden and does well: it explicitly warns that RSI cannot see real host token usage and that a savings number is only produced when ab is supplied, and it discloses that export=true writes a markdown file to ~/.rsi/exports/. It omits one notable side effect — the snapshot flag persisting a daily aggregate — which is only disclosed in the schema, not the description.

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 content is front-loaded and the honesty constraint is well placed, but the entire text is duplicated verbatim in Chinese, roughly doubling length for an English-reading agent without adding information. Redundancy costs it points even though no individual sentence 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 4-parameter, dependency-free reporting tool with no output schema, the definition covers the main inputs and the key limitation. The remaining gap is that the return payload's shape (what the 'panel' actually looks like) is only gestured at, and snapshot behavior is delegated entirely to the schema.

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 baseline is 3; the description goes beyond it by spelling out the nested ab shape (baseline/treatment with tokens and tasks) and, more importantly, the precondition that numbers must come from the host's own reported usage rather than being estimated. period and snapshot are left to the schema.

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?

It names a concrete deliverable — a value/impact panel composed of cost, recall value, behavioral gains, a time curve, and an A/B calculator — so an agent knows exactly what it produces. It does not, however, distinguish itself from the sibling rsi_stats, which plausibly also returns statistics, leaving overlap ambiguous.

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 phrase 'for evangelism' and 'a one-page markdown' imply the presentation/reporting scenario, and the ab condition is stated precisely. But there is no explicit when-not-to-use or routing versus rsi_stats / rsi_memory_health, so selection among siblings 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.