Record a decision the principal made

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

Annotations already declare the safety profile (write, non-destructive, not idempotent, not open-world), so the description's burden is lighter. It adds one useful behavioral fact beyond annotations — 'Returns follow-up questions' — but says nothing about what recording does to downstream state, whether escalation triggers fire anything, or how duplicates are handled.

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?

One dense sentence plus a short return-value fragment, front-loaded with the core verb and the field inventory. Every clause maps to a real parameter, though the comma list is compressed enough that an agent must parse it carefully.

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 14-parameter mutation tool with 21% schema coverage and no output schema, the description is thin: it covers roughly the decision narrative but omits required-parameter emphasis (workflow and situation are required), provenance via sources, outcome, missing_information, and idempotency. The single-line return note does not compensate for the missing parameter guidance.

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 only 21% across 14 parameters, so the description has to carry weight; it does map several natural-language phrases to parameters (situation, options/choice, reasons for rejected options, cues, 'what would change it' → reversal_conditions, 'when to escalate' → escalation_triggers). However, it leaves workflow, goal, rationale, sources, outcome, and missing_information unexplained, and gives no format or coupling guidance.

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 ('Capture a real decision') and enumerates the substance being captured (situation, options, choice, rejected-option reasons, cues, escalation triggers), so an agent knows exactly what payload this tool wants. It does not, however, explicitly differentiate itself from twin_amend_decision or twin_record_correction, so it falls short of a 5.

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?

The word 'real' hints that this records actual (not hypothetical) decisions, but there is no stated when-to-use, no prerequisite, and no routing against the 11 siblings such as twin_amend_decision (existing decision) or twin_record_correction. An agent must infer that this is the creation path.

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