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

Annotations only declare destructive/not-read-only; the description adds substantial context beyond them: the conflict failure mode on settled decisions, the null-clears-prediction semantics, that facet stays editable post-settlement, that kind/lesson are immutable, and that a needs_confirmation envelope may be returned. This is the behavioral detail an agent needs for a destructive in-place 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?

Front-loads what is corrected, then constraints, then failure modes, then the id source — a logical order of decreasing immediate relevance. Every clause carries a distinct rule (editable set, invariants, failure conditions, confirmation flow); no filler sentences.

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 7-param destructive mutation with a nested prediction object and no output schema, the description covers the editable surface, immutability rules, failure/conflict behavior, id provenance, and the confirmation envelope. An agent has everything needed to invoke it correctly and to anticipate refusals.

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 and the schema already documents each field. The description still adds meaning beyond it by mapping summary→title, rationale→body, explaining null clears the prediction, and noting facet is editable even after settlement. Modest added value over an already-complete schema, hence 4 rather than 5.

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 ('Corrects') plus resource ('an entry') and enumerates exactly which fields are editable (summary, rationale, prediction, facet). It also draws hard boundaries against siblings by declaring what it never edits (kind, lesson) and what it will not touch (settled, superseded, retracted entries). An agent can distinguish it from ledger_entries_settle/retract/create without opening a schema.

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 explicit when-to-use ('fix a typo, swapped fields, or a misrecorded detail') and explicit when-not (settled decisions, superseded/retracted entries), with the reason for the refusal. It also routes the agent to ledger_entries_list for ids, removing the guesswork about where entryId comes from.

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