Capture, rollback, and clean up bounded live state transactions

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

Far exceeds the destructive=true annotation: it discloses that capture stores raw client references in getgenv until commit/rollback/cleanup/expiry/shutdown, that rollback writes in reverse order, that connection disconnect cleanup is irreversible and reported separately, and that commit intentionally discards snapshots without cleaning registered resources.

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?

Front-loaded and organized into labeled sections, but bulky; the inline { action: ... } signature duplicates information already fully covered by the schema, and the Safety sentence restates points made earlier about rollback ordering and disconnect irreversibility.

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 a 12-param mutating tool with no output schema, the description covers lifecycle semantics, best-effort limitations, failure handling ('inspect tool-schema'), and verification path. An agent has everything needed to invoke it 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 coverage is 100%, so the schema already documents all 12 params (baseline 3). The description adds real meaning beyond field text by tying params to actions, e.g. 'capture it may only lower the call cap' for maxItems and the per-action semantics of cleanupMode/rollbackAction, though the inline signature largely echoes 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?

Opens with a specific verb+resource framing ('STATE TRANSACTION JOURNAL') and enumerates the six lifecycle actions (begin/capture/status/commit/rollback/cleanup) with what each does. This is unmistakably distinct from siblings like assert-state or world-delta, which it explicitly cross-references.

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?

States when to use it ('begin a transaction before mutating reversible client state, capture every property/attribute before changing it'), when not to ('not a general undo system') with a concrete exclusion list, and routes verification to 'assert-state'. Explicit alternatives and prerequisites are all present.

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