Skip to main content
Glama

WhaleScope MCP

by io.github.osindo-dev

Taker Imbalance Aggregator (Event-Domain Sampling)

taker_imbalance_aggregator
Read-only

Hitung rasio taker buy vs taker sell dari N trade PALING BARU (lookback_trades, bukan window waktu tetap) dari agg-trades yang di-supply caller (BUKAN fetch sendiri -- pass hasil binance_get_agg_trades / binance_get_spot_agg_trades). Event-domain sampling (jumlah trade tetap) dipilih ketimbang time-domain (durasi tetap) supaya ukuran sample -- dan reliabilitas statistik -- konsisten di kondisi market rame maupun sepi; lihat komentar kode buat rasional lengkap dan link ke probe CVD yang buktiin masalah ini.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
symbolNoLabel header saja -- kalkulasinya sendiri symbol-agnostic.
tradesYesAgg-trades URUT WAKTU NAIK (T non-decreasing), format native Binance (binance_get_agg_trades / binance_get_spot_agg_trades).
lookbackTradesNoJumlah trade PALING BARU yang dipakai (event-domain, bukan durasi waktu). Default 500 -- BELUM ADA VALIDASI EMPIRIK buat angka spesifik ini (beda dari window 60-menit analyze_cvd_divergence yang sudah diprobe 5 ronde). 500 dipilih sebagai angka bulat order-of-magnitude yang lebih kecil dari N 30-menit TERKECIL yang teramati di probe CVD buat pair kurang likuid (DOGEUSDT ~1770-5821 trade/window, lihat docs/mm_detection_framework.md) -- supaya kemungkinan besar tetap fillable bahkan di pair less-liquid, BUKAN hasil analisis statistik soal N optimal buat kualitas sinyal imbalance. INVESTIGASI N=500-vs-1000 SUDAH DIJALANIN (3 ronde probe, 2026-08-26) DAN DITUTUP INKONKLUSIF -- hipotesis regime-mixing yang sempat diajukan buat jelasin BTCUSDT-nya kebalik arah gugur diuji (lihat komentar UPDATE di atas), tapi akar masalah instability-nya sendiri BELUM ketemu. Default ini TETAP placeholder, bukan nilai tervalidasi -- jangan ulang metodologi span-vs-deviation yang sama tanpa hipotesis mekanisme baru.
maxLookbackSecondsNoAmbang staleness (detik) -- kalau rentang waktu buat ngumpulin lookbackTrades (trade terbaru dikurangi trade terlama di window) MELEBIHI ini, hasil di-flag stale=true (bukan reject keras -- tetap dihitung, tapi ditandai low-confidence). Default 3600 (60 menit) REUSE angka window terluas yang SUDAH divalidasi empirik di probe CVD analyze_cvd_divergence (fenomena sama: taker order flow) -- BUKAN angka yang diprobe khusus buat tool ini, cuma ceiling yang masuk akal by analogi.

TDQS

A4.5/5.0
Behavior5/5

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

Beyond the readOnlyHint annotation, the description discloses substantial behavior: staleness is handled by flagging 'stale=true' rather than hard-rejecting, defaults are admitted to be unvalidated placeholders ('BELUM ADA VALIDASI EMPIRIK'), and the maxLookbackSeconds default is an analogical reuse. It also warns against repeating an inconclusive methodology, which is genuinely useful behavioral context.

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?

The description is dense but front-loaded: the core computation and caller-supply requirement come first, followed by rationale and caveats. Some redundancy exists in the repeated empirical-validation caveats, but every sentence adds operational or decision-relevant detail.

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

Completeness4/5

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

Given four parameters, a 100%-covered schema, and readOnly annotations, the description is nearly complete: it specifies input requirements, sampling semantics, default validity, and staleness behavior. The only minor gap is that the exact output shape is not stated, though 'rasio' plus the stale flag description make it inferable.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3, but the description adds meaning beyond the schema: lookbackTrades=500 is a placeholder chosen for fillability, maxLookbackSeconds=3600 is borrowed by analogy, and maxLookbackSeconds behavior (stale flag vs reject) is described. This elevates it above the baseline.

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

Purpose5/5

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

Description opens with a specific verb and resource: 'Hitung rasio taker buy vs taker sell' over the N most recent trades, and explicitly scopes the calculation to caller-supplied agg-trades ('BUKAN fetch sendiri -- pass hasil binance_get_agg_trades / binance_get_spot_agg_trades'). This clearly distinguishes it from fetching tools like binance_get_taker_volume_ratio.

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

Usage Guidelines4/5

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

It gives clear usage context: the caller must supply ordered agg-trades, and the tool deliberately uses event-domain instead of time-domain sampling for consistent statistical reliability. It does not explicitly name alternative tools or state when-not-to-use scenarios, but the 'BUKAN fetch sendiri' directive plus the sampling rationale effectively routes usage.

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

All WhaleScope MCP toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.7/5.0
Disambiguation4/5

Most tools map to a distinct Binance metric or analytic concept, and descriptions explicitly contrast near-neighbors (spot vs futures, snapshot vs delta, 'BEDA dari...' notes). A few pairs could still be confused—`binance_get_basis` vs `binance_get_basis_history` and `binance_get_agg_trades` vs `binance_get_recent_trades`—but their purpose differences are explained well enough for careful agents.

Naming Consistency4/5

The dominant pattern is `binance_<verb>_<object>` in snake_case, with consistent complementary pairs like `get_*` and `get_*_history`. There are minor style breaks: `orderbook` vs `order_book`, the `whalescope_*` prefix, and `whalescope_full_pipeline` which lacks a verb, but the overall structure is readable and predictable.

Tool Count1/5

56 tools cross the explicit '50+ tools' extreme threshold. Although the Binance Futures domain is broad, many tools are single-endpoint or single-metric wrappers—multiple klines variants, order book variants, and ticker variants—that could be consolidated into parameterized composite tools. The surface is far too large for most agents to navigate efficiently.

Completeness4/5

The public market-data and analytics surface is remarkably complete: klines, funding, open interest, long/short ratios, top-trader data, liquidations, basis, order book behavior, regime detection, and full pipeline scoring are all covered. The main gaps are documented limitations such as unavailable liquidation-by-price data and non-public account/execution tooling, but agents can work around them without dead ends.

Resources