connectSource

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 the trust model (ingested content is UNTRUSTED; facts land proposed and capped, never superseding your own) and the return shape (itemsApplied/Skipped plus a jobUid). It omits auth/permission requirements beyond the granola key hint and says nothing about idempotency or repeat-run behavior.

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?

Opens with the action and outcome in one clause, then packs connector selection, optional knobs, trust model, and return value into three dense sentences with zero filler. Parenthetical asides are used efficiently rather than sprawling.

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 6-parameter mutating ingestion tool, the description covers the safety model, refusals, per-connector prerequisites, and the return report, and an output schema exists to carry return detail. A gap remains around the jobUid's lifecycle (polling via `listJobs`/`waitForIdle`) and how this differs from `ingestSource`.

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 100%, so the baseline is 3, but the description adds real meaning: it groups which parameters apply to which connector, restates defaults (personal), and characterizes maxItems/maxExtractionCalls as run bounds and dryRun as a count-only preview. The enum for scope is reinforced with the project-refusal rationale.

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 ('Connect') and resource ('a source') plus the key scope ('bounded first backfill into the graph'), which is far more informative than a bare name. It implicitly separates itself from the ongoing-ingest sibling (`ingestSource`) via 'first backfill', but never names that sibling outright, so an agent must infer the distinction.

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?

Gives per-connector guidance (repoPath for file-based, stored API key for granola, `maz source add granola` to set it) and explains dryRun as the preview/consent gate. It also states a hard exclusion ('Refuses project scope'). It stops short of saying when to prefer this over `ingestSource` or `listSources`.

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