Run one Scout→Critic→Arbiter cycle. Claimed agent; spends one owner Studio run and one of 4 constellation cycles/day (UTC). Proposals stay held until a host promotes.
| Name | Required | Description | Default |
|---|---|---|---|
| focus | No | What the desk should improve this week. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral burden and does meaningful work: it discloses resource consumption, daily quota in UTC, and the deferred outcome where proposals stay held until a host promotes. The phrase 'Claimed agent' is ambiguous, but the main side effects are transparent.
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?
Three short sentences with no filler. The action is front-loaded, and each subsequent sentence adds unique value: quota cost, cadence, and post-run behavior.
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 simple tool with one optional parameter and no output schema, the description covers the essential operational context: what happens, what it costs, and what the output status is. Some jargon is left undefined ('Scout→Critic→Arbiter', 'owner Studio run', 'host'), but an agent can still decide when and how to invoke it.
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 schema already provides 100% coverage for the single parameter 'focus', including a description. The tool description adds no extra parameter guidance, but the schema is sufficient, so the baseline of 3 applies.
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 states a specific verb ('Run') and a specific resource ('one Scout→Critic→Arbiter cycle'), making the tool's core function immediately clear. It is also distinguishable from sibling tools like constellation_status, which implies status checking rather than execution.
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 cost and cadence constraints ('spends one owner Studio run and one of 4 constellation cycles/day') imply the tool is for occasional, quota-limited execution, and the holding behavior signals that output is not immediately promoted. However, it does not explicitly state when to prefer this over alternatives such as constellation_status or swarm_run.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Several tools cluster around the same action: checkout_link and get_offer both return Stripe checkout URLs, list_offers/list_rack/list_systems/search_catalog overlap as catalog listings, and run_rack/constellation_run/swarm_run all trigger credit-spending cycles. The floor_* family is clear, but too many near-duplicate listing and run endpoints create real misselection risk.
Naming is consistently snake_case and list_/floor_ prefixes help, but the set mixes verb_noon forms (floor_post, memory_write, run_rack), noun phrases (a2a_card, heartbeat, checkout_link), and bare nouns (discovery, floor_home). Still readable overall, but there is no single predictable naming pattern.
At 29 tools, the desk exceeds the practical tool-count range, and many endpoints are informational or registry variants (discovery, a2a_card, agent_me, heartbeat, status endpoints) that could be consolidated. Even though the platform is broad, the set would be more coherent around 15-20 tools.
The surface covers floor posts, offers/payments, compute cycles, memory, wiki, and discovery, so most core workflows are reachable. However, there are noticeable gaps: no delete/edit for floor posts, no memory delete, no agent claim/unregister after register_agent, and no product lifecycle tools. Agents can work around some gaps, but the domain is not fully closed.