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

The description asserts "(READ-ONLY: mutations refused)" while annotations declare readOnlyHint=false and the name/title say "Save". This directly contradicts the structured safety signal and gives an agent no reliable basis for deciding whether calling this mutates state. It also omits any behavioral detail about what a checkpoint write affects or what constitutes a refused mutation.

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?

Two short sentences with the purpose front-loaded, which is the right shape. The trailing parenthetical, however, is pure cost: it is both misleading and appended as an afterthought rather than integrated into a coherent statement of effect.

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 ten-parameter write-shaped tool with no output schema, the description should say what comes back (a checkpoint ID? a path?) and whether any fields are effectively required for a usable handoff. Those gaps would matter less if the annotations were trusted, but the description has actively muddied the side-effect profile instead of clarifying it.

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 all ten parameters including session_id, crystallize, and tests_green are already documented in the schema and nothing needs compensating. The description restates a subset of the payload fields but adds no semantics beyond the schema text, so baseline 3 applies.

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 and resource ("Save a structured checkpoint") and enumerates the payload contents (summary, git state, next queue, open flags, track/lease metadata), which makes the intent unambiguous. It never names which sibling it competes with (session.record, session.continuity, session.digest, memory.create), so an agent cannot differentiate it from those without opening the schemas.

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?

"for lossless handoffs" implies the situation that selects this tool, but no explicit when-to-use rule, prerequisite, or exclusion is given. With five closely related session.* siblings in the list, the agent is left to infer the boundary.

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