Executes a selected write or action tool through the token-reduction proxy and returns a compressed result within a token budget, with full output available on request.
Run a tool that changes something ([w] in atr_find_tools). Result is compressed.
| Name | Required | Description | Default |
|---|---|---|---|
| tool | Yes | ||
| budget | No | Max tokens for the result (default from config). | |
| arguments | No |
Changes observed during successful MCP inspections.
v1.0.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint=true and openWorldHint=true, so safety is covered structurally. The description adds the non-obvious fact that results are compressed (and that a token budget applies), but says nothing about irreversibility, confirmation needs, or what happens when the target tool fails.
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?
Two short sentences with no padding, but the cryptic bracketed notation and unexplained 'compressed' term trade clarity for brevity rather than earning it. Front-loading the mutation semantics is correct.
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 destructive open-world dispatcher with a nested arguments object and no output schema, the description omits argument forwarding, error behavior, and compression/budget semantics. An agent cannot reliably call this without inspecting atr_find_tools or guessing.
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 only 33%, and only the budget parameter is documented in the schema. The description never explains that 'tool' expects a name from atr_find_tools or that the nested 'arguments' object is forwarded as the target tool's arguments — the single most important thing to know for this dispatcher.
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 the tool dispatches another tool that mutates state, which distinguishes it from atr_read in principle. But 'Run a tool that changes something' is loose, and the '[w] in atr_find_tools' notation is unexplained jargon an agent must guess at.
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 hints that mutation targets come from atr_find_tools marked [w], but never states when to use this versus atr_read, atr_run, or atr_query. There are no explicit exclusions or decision criteria despite nine siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.