读取任务完整产物

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

Annotations already declare readOnlyHint=true and openWorldHint=false, so safety is covered; the description adds genuinely new behavior — the disk-spill trigger, pagination semantics, and a directory-scope restriction that tells the agent it cannot read outside the task sandbox. It stops short of describing the returned payload shape (plain character slices vs. structured records), which matters since there is no output schema.

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?

A single clause-chained sentence with the precondition front-loaded, followed by the action and the scope restriction. No filler, no repetition of the title or parameter names.

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 read-only tool with full schema coverage and no output schema, the definition covers trigger, scope, and the two selectable sources. The only gap is that paging semantics across successive calls (how to advance offset, when reading is complete) are left implicit, which an agent must infer.

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% — offset, source, max_chars, and task_id all carry their own descriptions including defaults and enum meanings. The description's 'paginated read' and 'result or event log' phrasing conceptually motivates offset/max_chars and source, but adds no syntax, limits, or interaction detail beyond the schema, so the baseline of 3 applies.

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 names a specific action (paginated full read), the two resources it can read (final result file or event-stream JSONL), and the exact precondition that distinguishes it from its sibling dsh_result (spilled-to-disk output returning artifacts). An agent can separate it from dsh_result, dsh_status, and dsh_audit without opening any schema.

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 gives an explicit trigger (only when dsh_result returns artifacts because output was too long), names the alternative it complements rather than replaces, and states a hard boundary (only within the task directory). Both the when-to-use and the when-not-to-use are inferable directly from the text.

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