Start a memory session

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

Annotations already cover the safety profile (readOnlyHint=true, idempotentHint=false, openWorldHint=false). The description adds useful non-obvious context the annotations can't convey: coverage is explicit, candidates are ranked but never establish identity, and a incomplete-context warning. It stops short of describing what 'live notes' or 'write capability' look like, so the added value is moderate rather than rich.

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?

Six dense sentences with heavy domain-specific jargon ('versioned common discipline', 'known_contract_hash', 'bounded live notes'). It is front-loaded, but the phrasing is so compressed and system-specific that it reads as under-specified rather than concise; the agent expends effort decoding terms instead of acting.

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?

An output schema exists, so return values needn't be explained, which lowers the burden. The description covers the session concept, coverage transparency, identity caution, and known_contract_hash behavior, but with 0% schema coverage and three undocumented params, plus no output or rate/pagination context, it is only marginally complete for a zero-required-param, five-parameter entry-point 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 description coverage is 0%, so the description must compensate, and it partially does: it flags known_contract_hash semantics (send only when the exact contract is already in context; a match short-circuits and returns identity) and what subject does (finds ranked candidates without repairing the index). But domain, max_tokens, and note_paths get no semantic treatment, leaving three of five params 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?

The opening imperative 'Start memory-dependent work here' states a purpose, and the title restates it as 'Start a memory session.' However, the core action is described in abstract, system-specific jargon ('versioned common discipline', 'bounded live notes') that an agent cannot concretely map to a resource. It doesn't clearly distinguish this session-initialization tool from siblings like prepare_follow_up or get_write_progress beyond the 'start here' cue.

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?

Strong routing cues are present: 'Start memory-dependent work here' gives an explicit when-to-use, and 'Read next pages before relying on incomplete context' tells the agent not to treat a single call as authoritative. It doesn't name any sibling as an alternative, but the entry-point framing is clear enough to guide invocation ordering.

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