watch_log_start

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

Annotations already declare readOnly/idempotent/non-destructive, so the bar is lower, yet the description adds real behavioral color: it survives log recreation at game start, maps events back to draft or cached mod source, and emits code excerpts. It does not clarify what happens if it is called while a watcher is already running, which matters given the idempotentHint claim on an inherently stateful action.

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 critical information (target file, start point, output routing) is front-loaded, and the closing pointer to watch_log_poll is well placed. The middle clause stacking parse behavior, source mapping, and log-recreation survival is dense, but each clause carries distinct information.

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?

With no output schema, the description does the work of explaining what the watcher yields and where to retrieve it, which is the key thing an agent needs. It is slightly incomplete on lifecycle edges — double-start behavior and the relationship to watch_log_stop — but is otherwise self-sufficient for correct invocation.

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 carry both parameters. It documents from_start well (inclusive of the semantic effect of the flag), but the 'warnings' toggle is only obliquely implied by 'engine warnings' — an agent cannot tell from the text what warnings=false suppresses.

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 and resource — 'Start a background watcher on gen/log.txt' — and immediately characterizes the payload it produces (parsed Lua errors, engine warnings, mapped mod source with excerpts). It is clearly distinguishable from read-only siblings like read_log and log_tail, and it names watch_log_poll as the consumer.

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 clear operational context: start the watcher, then 'Read events with watch_log_poll', and it explains the from_start choice as a usage decision (tail from current end vs. replay from the beginning). It stops short of explicit when-not guidance (e.g. when to prefer read_log/log_tail over a live watcher, or when watch_log_stop is required).

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