ideas

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

With no annotations, the description does real work: it discloses side effects (writes ideas.json/.md into the run folder and appends a 'Test ideas' section to the dashboard) and the shape of the return (prioritised experiments each citing evidence plus a simulation config). It omits permission/auth needs, what happens when no run folder exists, and whether writes are additive or overwrite.

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 input contract is front-loaded before the outputs and persistence behavior, and both sentences carry real information with no filler. The opening clause is long and comma-heavy, which slightly slows parsing but no sentence 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 six-parameter tool with zero schema coverage, no annotations, and no output schema, the description covers the data inputs, the run context, and the returned artifacts adequately. It still leaves three parameters undefined and gives no guidance on defaults or failure modes, which is a meaningful gap given the empty schema.

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 for six bare parameters. It does well for rows (enumerating name, campaign, spend, impressions, clicks, conversions, country), ads_file (export file path), and run (prior run's objections, audiences, decision), but n, brand, and brief receive no explanation at all, leaving half the surface undocumented.

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 action (generate next test ideas) on a specific resource (real ad data / a previous run) and names what comes out (prioritised experiments with evidence plus a MarketMirror config to simulate). It implicitly separates itself from market_test/run_test by positioning ideas as the pre-simulation step, though the verb phrasing 'Next test ideas' is slightly awkward.

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?

The description implies when it applies ('pass ad rows you fetched... and/or an export file path, plus a previous run'), which tells the agent the expected preconditions. It never states when NOT to use it or names an explicit alternative sibling, so the agent must infer the boundary with market_test, research, or campaign_plan.

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