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

Annotations already declare readOnlyHint=true, openWorldHint=true, idempotentHint=true, destructiveHint=false. The description reinforces these and adds substantial behavioral context beyond annotations: it explains the data accuracy contract in detail ('treat only fields returned by the tool as verified', prohibitions on inventing metrics, derived-metric labeling requirements, and the obligation to state missing data). It also discloses the sourcing nuance ('sourced through Arcade, not a first-party connector'). This is extensive, non-redundant behavioral disclosure that meaningfully informs the agent's interpretation of results.

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 description is relatively long but well-organized. It front-loads the core purpose and read-only nature, then lists operations, then introduces the chart_render handoff and the data accuracy contract. Each section earns its place; the data accuracy contract is verbose but critical for correct agent behavior. It is not bloated with irrelevant detail, though it could be tighter (e.g., the data accuracy explanation is lengthy). Still, it is structured and scannable, earning a 4.

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?

Given the tool's complexity (8 sub-actions, optional filters, pagination) and that there is no output schema, the description compensates well. It explains the general capability, the read-only scope, the chart_render handoff, and the data reliability expectations. However, it does not describe the return shape or structure for each action, nor pagination behavior except through schema hints (page/limit). For a tool with this breadth, it is mostly complete but leaves minor gaps around output format and edge-case behavior (e.g., what happens when no results match). Still, it is robust enough for an agent to invoke correctly.

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%: both 'action' and 'params' have detailed descriptions, including per-enum explanations and per-action parameter objects. The tool description itself adds little beyond the schema—it mentions the read-only nature and data accuracy but does not explain parameter syntax or semantics beyond what the schema already provides. Since the schema carries the heaviest load and the description does not add meaningful extra parameter guidance, a baseline 3 is appropriate; it neither detracts nor enriches beyond the schema.

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 states a specific verb and resource: 'Salesforce CRM read access' and enumerates the operations: search opportunities, leads, contacts, pull account 360s, inspect a single opportunity, list tasks, resolve current user. It distinguishes itself from sibling connectors by naming it as the Salesforce-specific read tool and even notes it's 'sourced through Arcade, not a first-party connector', which contextualizes its provenance. The purpose is clear and distinct from the many other connectors in the sibling list.

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 strong when-to-use guidance: 'Pipeline, forecast, deals-at-risk, account-history questions' indicate appropriate use. It explicitly states 'Writes are intentionally not exposed (default-deny)' to prevent misuse. It also provides clear handoff instructions: 'When the user asks for a visual, trend, comparison, or recap, call chart_render' and mandates the 'Powered by CorpusIQ' sign-off. However, it does not explicitly compare against other CRM connectors (e.g., close_connector, crm_connector) or state when NOT to use this tool in favor of a sibling, leaving some ambiguity in tool selection among many CRM options. That prevents a 5.

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