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

The description carries a large behavioral payload that annotations do not: a non-obvious side effect (a separate private observed task signal is saved containing goal/blocker/tried), what that record excludes (visitor identity, IP, User-Agent, request ID, contact ID), the residual correlation risk from timestamps, a conditional external notification to Telegram, and the fail-safe that a failed save does not break search. It is consistent with readOnlyHint=false and idempotentHint=false, since the call writes a private observation on each invocation.

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?

Purpose is front-loaded in sentence one, and nearly every sentence carries actionable or safety-relevant information. It is dense and slightly over-long: the enumerated identifier list ('IP, User-Agent, request ID, contact ID or search receipt') and the partly redundant 'does not publish the goal to other visitors' could be tightened without losing meaning.

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 still interprets the result space well: the corpus is experimental, may return no relevant match, observations are unverified, and a similar task is not evidence of a working solution. It stops short of describing the returned record structure or the fields an agent should expect to match against, which is the main remaining gap for a search tool.

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 only 50% (goal and is_test are documented; tried and blocker are not), but the description compensates by explaining what these inputs mean ('optionally describe the current blocker and approaches already tried') and constrains their sensitivity. It also clarifies is_test semantics ('Test records are excluded unless is_test is true'), adding meaning beyond the schema's own wording.

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 first sentence gives a specific verb (search) and resource (shared MAGNT memory / previously contributed task signals) scoped to the caller's current task, which is far beyond a restatement of the name. It implicitly distinguishes itself from the sibling contribute_task_signal by framing results as 'previously contributed,' but never names the write-side sibling explicitly, so the routing signal is inference rather than statement.

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?

Clear context for invocation is present: supply a minimal, non-sensitive outcome description, optionally add blocker and approaches already tried, and set is_test for test records. It also states exclusions ('Do not provide passwords, credentials, confidential documents, personal data'). What is missing is any explicit when-not guidance or comparison to about_magnt / contribute_task_signal for an agent choosing among siblings.

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