decision_record

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

With no annotations, the description carries the full burden. It usefully discloses that the tool is deterministic ('Ноль LLM, чистая агрегация') and invariant-preserving (tiers copied as-is from session.advisors[].opinions[].marker; never raises or invents 🔵). But it omits whether it persists/writes a record (the name 'decision_record' implies a side effect), auth requirements, and return shape.

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 the identity and trigger condition, then the output contents, then the key invariant — each clause earns its place despite dense jargon and capitalization. No filler or repetition.

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?

Covers purpose, trigger, output contents, and a core determinism/tier invariant, which is substantial for a nested-object tool with no output schema. Gaps remain on the optional 'surface' param, whether the tool persists an artifact, and its return value, so it is adequate but not complete.

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 0% for two parameters, so the description should compensate. It does add meaning for the required 'session' object by referencing its nested structure (advisors[].opinions[].marker, disagreement), which is genuine value beyond the schema. It says nothing about the 'surface' parameter or its md-only enum, leaving that half 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?

States a specific verb+resource: assembles a meeting protocol/minutes from the council's canonical output, itemizing what it collects (advisor positions, dissent, decision+status, review triggers, marker provenance). The purpose is concrete and distinguishable from a generic renderer. However, it never names or contrasts against the many nearby siblings (council_verdict, render_session, save_decision_card, close_decision_card), so the agent must infer differentiation.

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 an explicit activation condition — only on an EXPLICIT user request, with trigger phrases quoted ('протокол', 'минуты', 'оформи решение'). This is a clear when-to-use cue. It stops short of naming alternatives or stating when NOT to use it relative to sibling tools that also produce session artifacts.

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