media_search

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

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is settled without the description. The text adds genuine value by disclosing the result shape: source-time ranges plus possible alternative attempts and preview hints. It says nothing about pagination behavior, empty-result handling, or whether the index must already be built (media_index / media_index_get exist as siblings).

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?

Three tight sentences, front-loaded with the core action, then the disambiguation, then the follow-up workflow. Every sentence carries distinct information with no restatement of the tool name. Slightly dense phrasing ("possible alternative attempts") but nothing 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 five-parameter read tool with no output schema and thin schema descriptions, the description covers what comes back and how to use it next, which is the important part. It leaves the indexId prerequisite and the limit/offset pagination model entirely unexplained, so an agent must guess how to page through a large result set or obtain a valid index.

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 only 20%, with just projectDir documented, so the description carries most of the burden. "Search words spoken" does clarify that query is a lexical phrase against transcript text rather than an image or take descriptor, which is useful. But indexId, limit and offset get no explanation anywhere, and the maximum of 50 results is invisible to the reader.

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?

"Search words spoken in an indexed source recording" names a specific verb and resource, and the description states the output (source-time ranges plus alternative attempts). It also rules out semantic image search and best-take selection, so the agent knows what class of operation this is. It never mentions the sibling transcript_search, which is the closest analogue, so full sibling differentiation is missing.

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 downstream step: use returned preview hints with source_preview to compare actual delivery before editing, which is real routing guidance. It also excludes two tempting misuses (semantic image search, automatic best-take selection). The gap is that it never says when to choose this over transcript_search, an adjacent sibling that an agent will likely confuse it with.

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