Skip to main content
Glama
Sign In

aipricepatterns

by com.aipricepatterns
OverviewDiscussionsAdmin

Run Candle Research Watch Cycle

run_candle_research_watch
Read-onlyIdempotent

Autonomous researcher and standing-watch loop in one call. Scans BTC, ETH and SOL, inspects the tightest neighbour bands, and checks walk-forward range lift. It replays a price order only when the caller supplies levelPrice, invalidationPrice and targetPrice. It does not invent those prices from P90 or P10. Without a complete order it reports the band finding only. UTC session context and watchId cooldown stay. Always returns NO_TRADE.

Input Schema

NameRequiredDescriptionDefault
fNoForward horizon in bars. Only the chart presets 5, 10, 30, and 50 are supported
qNoCandles in the matched pattern. Only the chart presets 3, 5, 8, and 12 are supported
topNNoHow many top radar rows (tightest neighbour bands) to inspect. Default 3. A named order is replayed on each row. Prices are not invented.
symbolsNoOptional subset of BTCUSDT, ETHUSDT, SOLUSDT. Default: all three.
watchIdNoOptional persistent identifier for this agent's watch session (e.g. 'claude-btc-watch'). When supplied, the server remembers what was already reported and suppresses duplicate alerts for quietWindowMinutes unless a new finding type appears or the replay gap shifts by >= 0.5 pp.
intervalsNoOptional timeframes from 5m, 15m, 1h, 4h, 1d. Default: 15m, 1h, 4h.
levelPriceNoEntry the caller named. Replay runs only when levelPrice, invalidationPrice and targetPrice are all set. Do not invent this from a band. Alias: level.
targetPriceNoTarget the caller named, on the opposite side of levelPrice from the invalidation. Do not invent this from a band. Aliases: target, targetLevel.
feePctPerSideNoFee percent per side for costed barrier replay of a named order. Default 0.1.
invalidationPriceNoInvalidation the caller named, on the opposite side of levelPrice. Do not invent this from P10 or P90. Aliases: stop, stopPrice, invalidation.
quietWindowMinutesNoCooldown window in minutes per symbol/interval when watchId is supplied. Default 360 (6 hours).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed5 schema fields changed
    • changedInput schema / properties / feePctPerSide / description
      Previous value: -"Fee percent per side for costed barrier replay. Default 0.1."New value: +"Fee percent per side for costed barrier replay of a named order. Default 0.1."
    • addedInput schema / properties / invalidationPrice
      Added value: +{
+  "description": "Invalidation the caller named, on the opposite side of levelPrice. Do not invent this from P10 or P90. Aliases: stop, stopPrice, invalidation.",
+  "exclusiveMinimum": 0,
+  "type": "number"
+}
    • addedInput schema / properties / levelPrice
      Added value: +{
+  "description": "Entry the caller named. Replay runs only when levelPrice, invalidationPrice and targetPrice are all set. Do not invent this from a band. Alias: level.",
+  "exclusiveMinimum": 0,
+  "type": "number"
+}
    • addedInput schema / properties / targetPrice
      Added value: +{
+  "description": "Target the caller named, on the opposite side of levelPrice from the invalidation. Do not invent this from a band. Aliases: target, targetLevel.",
+  "exclusiveMinimum": 0,
+  "type": "number"
+}
    • changedInput schema / properties / topN / description
      Previous value: -"How many top radar rows (tightest neighbour bands) to inspect with competing hypotheses and costed replay. Default 3."New value: +"How many top radar rows (tightest neighbour bands) to inspect. Default 3. A named order is replayed on each row. Prices are not invented."
  2. Added

TDQS

A3.6/5.0
Behavior4/5

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

Annotations already establish the safe, read-only, idempotent profile, and the description adds real behavioral context on top: it never fabricates prices from P90/P10, it degrades gracefully to a band-only report, watchId cooldown suppression applies, and UTC session context is retained. The closing 'Always returns NO_TRADE' is a notable disclosure but is stated without explanation, leaving some uncertainty about what a caller actually receives.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Six short sentences, front-loaded with the core capability before the conditional rules. Most sentences earn their place, though 'UTC session context and watchId cooldown stay' is elliptical to the point of being unclear and could be tightened or dropped.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 11-parameter, zero-required orchestration tool with no output schema, the description carries the full burden of explaining return values and it does so only partially. 'Always returns NO_TRADE' plus 'reports the band finding only' hints at the output shape but does not describe what a band finding contains, and the relationship between the research scan and the watch loop's persisted state is left implicit.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents every parameter including the 'do not invent this from a band' caveats, meaning the description largely restates schema content. It adds marginal meaning by framing the three price parameters as a joint precondition for replay and by tying quietWindowMinutes behavior to watchId, but supplies no format, units, or ordering details beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence names the composite behavior precisely (research pass plus standing-watch loop in a single call), and the following sentences enumerate the concrete steps: scanning BTC/ETH/SOL, inspecting tightest neighbour bands, and checking walk-forward range lift. It is clearly distinguishable from narrower siblings like scan_candle_memory_radar or check_candle_order, though it never names those siblings explicitly. Domain jargon ('neighbour bands', 'range lift') adds ambiguity for a first-time caller.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives conditional behavior ('replays a price order only when the caller supplies levelPrice, invalidationPrice and targetPrice') and the fallback ('without a complete order it reports the band finding only'), which tells an agent what happens but not when to choose this tool over check_candle_order or replay_candle_hypothesis. No explicit when-to-use or when-not-to-use routing against the 19 siblings is present.

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

All aipricepatterns toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources