shell_job_output

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 discloses that the call is idempotent/non-destructive ('does not consume the buffer'), that it is safe to call repeatedly, and that exit_code is null while running. It omits permissions/auth and any rate-limit or buffer-eviction caveats, keeping it just short of a 5.

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?

Front-loaded one-line purpose followed by Args/Returns blocks; every section earns its place. Slightly formulaic but no wasted sentences, and the behavioral caveat appears before the parameter detail.

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?

No output schema exists, so the Returns block (running, exit_code, stdout, stderr, elapsed_s) is genuinely needed and supplied. Combined with the non-consuming guarantee, an agent has enough to poll and interpret results; only cross-tool routing to shell_job_wait is missing.

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 0%, so the description must compensate and largely does: job_id is explained as the id returned by shell_start, and tail_lines is explained as 'last N lines of each stream, 0 means all.' It fails to mention the schema's default of 200, so an agent cannot fully predict default behavior, but the added semantic meaning is substantial.

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?

States a specific verb+resource ('Read the output captured so far from a background job') and clarifies the snapshot semantics. It names shell_start as the source of job_id, which implicitly differentiates it from siblings, but never explicitly contrasts it with shell_job_wait, the closest alternative.

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?

'Safe to call repeatedly while the job runs' implies a polling usage pattern, and the non-consuming behavior hints at when this is preferred. However, it never explicitly says when to use this versus shell_job_wait (blocking) or how to know the job is finished beyond the returned 'running' field. Usage is implied rather than stated.

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