agent_reuse_recommend

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

With no annotations provided, the description carries the full behavioral burden and largely meets it. It discloses that this is a read/recommend operation (no mutation), that the default output is a COMPACT trimmed projection (not missing fields), and that the resume_hint is a ready-to-run call requiring summary. The 'does not mutate — Leader decides' framing is explicit. Minor gap: it doesn't explicitly state what happens on empty/no-match results or whether candidate data is sourced from storage vs live state.

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 a single dense paragraph followed by a focused second paragraph on availability tiers and addressing, and a closing on the default compact response. Every sentence earns its place — purpose, decision actions, availability framework, addressing semantics, fallback, and output projection. No filler, no repetition of schema content that isn't enriched.

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 tool with 6 params, an output schema, and no annotations, this description is thorough. It explains the decision model (three actions), availability tiers, addressing semantics, name-vs-id fallback, and output projection behavior. The output schema exists so return-value details needn't be in the description. It covers the behavioral and operational context an agent needs to correctly invoke and interpret results.

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. The description adds meaningful value beyond the schema by explaining the 'Address candidates by NAME' vs agentId fallback semantics, what resume_hint contains (ready-to-run call), and the compact vs all projection meaning (trimmed, NOT missing fields). The 'fields' behavior is enriched beyond its schema line, and the project_id/session_id scoping is clarified (cross-session inference). This lifts it above the baseline.

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?

The description clearly states the tool's purpose: recommend whether to reuse an existing sub-agent for follow-up work, ranking candidates by domain match and returning one of three actions. It is specific about the verb+resource (recommend reuse) and the decision framework (reuse/slim_then_reuse/spawn_new). Among siblings like agent_list, agent_template_recommend, and fleet_dispatch, this stands out as the reuse-decision tool, not just an enumeration or dispatch mechanism.

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 description explicitly states when to use it ('follow-up work: bug re-fix, deeper research, same-domain iteration') for resuming a prior sub-agent to preserve accumulated context. It clarifies the tool only recommends and the Leader decides. The availability tiers and the addressing guidance (by NAME vs agentId fallback) provide strong operational context. While it doesn't name specific sibling alternatives, the situational trigger is precise enough.

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