hebbrix_choose_action

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

With annotations only declaring false hints, the description carries full responsibility. It reveals behavioral traits: records the action, recommends conservatively, never grants permissions, censors missing outcomes (not failures), and limits exploration to ≤0.2. These are meaningful and go beyond the schema. No contradiction with annotations.

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?

The description is well-structured into paragraphs that flow logically: purpose, usage modes, exploration constraints, and follow-up workflow. It front-loads the main purpose and avoids redundancy, though it is longer than minimal due to the complexity. Every sentence adds value without unnecessary fluff.

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?

The description covers the complete decision-recording loop: choose, get decision_id, perform action, report outcome later. It explains the main parameters and the safe baseline. Given that an output schema exists, return values are not required. Missing parameters like idempotency_key and collection_id would benefit from brief notes, but overall it is nearly complete for the tool's intended use.

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?

The description explains key parameters (policy_key, actions, context, baseline_action, chosen_action, action_probability, exploration_rate) with semantic context (e.g., 'actions are stable machine keys', 'context contains only factors that may change which action works'). However, it omits details for user_id, collection_id, and idempotency_key, which are left entirely to the schema. Since schema description coverage is 0%, the description compensates for most but not all.

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?

Clearly states the tool's purpose: 'Choose and RECORD an action before its result is known.' It lists specific use cases (reply strategy, workflow, etc.) and differentiates from siblings like hebbrix_log_decision and hebbrix_report_outcome by focusing on the decision recording before outcome.

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?

Provides explicit guidance: normal use (omit chosen_action) vs logging an external choice (pass chosen_action and action_probability), sets an exploration_rate cap (≤0.2), and instructs to call hebbrix_report_outcome later. It also clarifies that outcomes are censored when missing, giving clear when-to-use vs when-not-to-use information.

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