Point-in-time archive: the scores Coil actually published on an archived date — sha256-committed BEFORE outcomes were known, append-only, never revised. Backtest-grade provenance; no free counterpart. Omit date for the list of archived dates. $0.02 USDC per call (x402).
| Name | Required | Description | Default |
|---|---|---|---|
| date | No | Archived date, YYYY-MM-DD; omit for the archive index |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It discloses critical behavioral traits: data is sha256-committed before outcomes were known, append-only, never revised, and costs $0.02 USDC per call. This provides transparency about data integrity, mutation policy, and pricing beyond the basic tool name.
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 compact, using three sentences to convey purpose, integrity guarantees, usage variation, and pricing. It front-loads the core value proposition ('Point-in-time archive') and avoids redundancy; every sentence earns its place.
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?
Given the tool's simplicity (one optional parameter, no output schema), the description is complete. It covers the purpose, the historical/provenance use case, the append-only behavior, the cost, and the variation when omitting the date. It does not explain the return format, but that's not necessary given the tool's straightforward nature.
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 of the single 'date' parameter, describing it as an archived date in YYYY-MM-DD format and noting it can be omitted for the archive index. The tool description repeats this usage ('omit date for the list of archived dates') without adding new semantics, so a baseline score of 3 is appropriate.
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 identifies the tool as a point-in-time archive of Coil's published scores for a specific archived date, distinguishing it from live/free counterparts by emphasizing sha256-committed, append-only data and backtest-grade provenance. This specific verb+resource+scope definition differentiates it from sibling tools like get_full_board.
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 provides clear context: it is for backtest-grade historical data, explicitly noting 'no free counterpart' and instructing to 'omit date for the list of archived dates.' It implies use for archival/provenance needs rather than live data, though it doesn't name explicit alternative tools.
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.
Most market-data and storefront tools are cleanly scoped, but there is meaningful overlap: get_full_board and get_engine_feed are near-duplicates, get_buy_list overlaps with get_morning_brief's candidate list, and get_orders / get_order_status are identical NotOffered stubs. The free/live twin pairs are clearly labeled so agents can distinguish latency/payment intent, but the sheer number of related reads creates selection friction.
The large get_* data family is very predictable (get_stock_read, get_stock_read_live, get_rotation, get_rotation_live), and storefront ools use clear verb_noun actions (add_to_cart, remove_from_cart, save_memory). Minor deviations like checkout_handoff and the get_live_setup / get_live_rules naming (which could be misread as *_live twins) keep it from a perfect score.
46 tools is well abovethe 25+ threshold for a coherent toset; at leat13 are direc live twins of free readable plus a 14-tool storefont/memory subdomain. Many could be consoliated into ingle tools with a delay/live param or a single storefront resource, making the surface feel bloated for agents.
The market-data surface is very comphehensive: regime, buy-list, hold-state, rotation, sector, stock, crypto day/night/trend/setups, full board, archived board, changes, and engine-feed coverage leave few dead ends for the stated trading-intelligence purpose. Storefont is adequate but has minor acknowledged gaps—orders/order-status are NotOffered stubs and session memories have no update/delete—so it is not a perfect 5.
