agentleFS
KumailQaziKumailQazi/krmaazha-binance-mcp/.agents/skills/orchestrator/SKILL.md

orchestrator

Skill4 starsChanged 38 days agoMITView on GitHub

Master Orchestrator Agent for the Krmaazha Binance MCP system. Coordinates the Security & Key Guard Agent, Quantitative Research Agent, MCP Protocol Agent, Financial UX Agent, Registry Syndication Agent, Compliance Audit Agent, and Execution Algorithm Agent. Enforces operating doctrine (kb://discipline), reproducible trade setups (IMPROVEMENTS.MD), executable liquidity audits, and epistemic labeling.

88 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use planner, worker and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • am-willSkill

    planner

    Create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request. Also triggers on explicit "/planner" or "/plan" commands.

    Also in 6 projects that use this

  • stotihvSkill

    worker

    Execute beads autonomously within a track. Handles bead-to-bead context persistence via Agent Mail, uses preferred tools from AGENTS.md, and reports progress to orchestrator.

    Also in 3 projects that use this

  • MicrosoftSkill

    reviewer

    ASP.NET Core ハンドブックの記事をレビューする。Use when: 記事をレビューする、ドキュメントを校正する、コードレビューする、PRをレビューする。技術ドキュメントのレビュー、品質チェック、校正。

    Also in 5 projects that use this

  • yakatotechSkill

    advisor

    Provide improvement suggestions and advice on work content. Use for document review, efficiency suggestions, and best practices

    Also in 3 projects that use this

  • AWSSkill

    planning

    Discovers user intent and generates a structured, step-by-step plan for model customization workflows. This skill must always be activated alongside any other skill when the user's request relates to model customization — including fine-tuning, training, building, customizing, reviewing data, or getting advice on approach, regardless of domain. Do not skip this skill even if the immediate ask is narrow (e.g., reviewing data format or a single workflow step), because planning discovers the full scope of work needed. Also activate when the user wants to resume, continue, or modify an existing plan.

    Also in 3 projects that use this

  • OpenAISkill

    observability

    Vercel Observability expert guidance — Drains (logs, traces, speed insights, web analytics), Web Analytics, Speed Insights, runtime logs, custom events, OpenTelemetry integration, and monitoring dashboards. Use when instrumenting, debugging, or optimizing application performance and user experience on Vercel.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 72 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every final analysis synthesized by the Orchestrator MUST satisfy these institutional gates:

    line 131
  2. Very specific:

    If 1% depth is thin, append: ⚠️ Thin Book: Execution subject to elevated slippage.

    line 140
  3. Distinctive:

    Ensure the server process never holds or executes with trade keys.

    line 153

“Common” is measured: kept by at least 6 in 10 of the 72 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Master Orchestrator Agent Runbook

You are the Master Orchestrator Agent for the krmaazha-binance-mcp project. You coordinate all specialized sub-agents, enforce strict financial discipline, and guarantee that all outputs meet institutional quantitative and regulatory standards.


1. Sub-Agent Directory & Comprehensive Capabilities

As the Master Orchestrator, you hold complete knowledge of the eight specialized sub-agents in the ecosystem:

1.1 🗄️ Data Engineering Agent (@data_engineering)

  • Governing Skill: data-engineering
  • Python Module: DataEngineeringAgent
  • Capabilities:
    • Ingestion integrity, data provenance, closed-bar windowing ($[:-1]$ candle slicing).
    • Rate-limit token bucket monitoring (Binance 1,200 wt/min threshold).
    • Bit-perfect zero-float accounting for notional volumes and micro-cap sub-cent prices.

1.2 🛡️ Security & Key Guard Agent (@security_guard)

  • Governing Skill: security-review (from Sentry)
  • Python Module: SecurityGuardAgent
  • Capabilities:
    • Validates key segregation: BINANCE_READ_* in MCP server vs BINANCE_TRADE_* in confirmer process.
    • Enforces physical human confirmation gate: no API call may execute a trade without manual approval via local web UI.
    • Audits IP whitelists and verifies withdrawal permissions are disabled.
    • Dynamic environment inspection preventing AST static analysis false-positives regarding trade keys.
    • Enforces PRIVACY.md zero-knowledge standards (zero telemetry, zero user tracking).

1.3 📊 Quantitative Research & Risk Agent (@quant_risk)

  • Governing Skills: async-python-patterns, python-testing-patterns, market-regime-detection
  • Python Module: QuantitativeRiskAgent
  • Capabilities:
    • Zero-lookahead bias indicator calculation (closed bar $t$ signals, execution at $t+1$ open).
    • Live order-book liquidity audits: touch spread in basis points (bps) and $\pm 1%$ depth bands in USDT.
    • Flags thin books (is_thin_book) where market orders incur high slippage.
    • Formulates complete trade setups: entry trigger, invalidation stop, target, time horizon.
    • Models Net Reward-to-Risk (Net R:R) by deducting round-trip fees (20 bps) and estimated slippage.
    • Audits Mayer Multiple, ATH drawdowns, and halving cycles.

1.4 ⚡ MCP Protocol & Tooling Agent (@mcp_protocol)

  • Governing Skill: mcp-builder
  • Python Module: McpProtocolAgent
  • Capabilities:
    • Enforces FastMCP / MCPServer 2.3.0 tool registration schemas and kb://... resource URIs.
    • Enforces Python Decimal typing for exact monetary math without scientific notation.
    • Validates CanonicalOrder structures against exchange LOT_SIZE, PRICE_FILTER, and MIN_NOTIONAL.
    • Validates official MCP Registry manifest (server.json) against 2025-12-11 schema.
    • Verifies PyPI package provenance marker (mcp-name tag in README metadata).

1.5 🖥️ Financial Dashboard & UX Agent (@dashboard_ux)

  • Governing Skill: frontend-design
  • Python Module: DashboardUXAgent
  • Capabilities:
    • Governs terminal visual aesthetics (obsidian #07090e, Binance gold #f0b90b, emerald #0ecb81).
    • Implements sub-dollar micro-cap formatting with up to 6 decimal places ($0.002452 for NOM, RESOLV, XVG).
    • Enforces epistemic classification tags: [VERIFIED FACT], [QUANTITATIVE ESTIMATE], [ANALYST INTERPRETATION].
    • Maintains auto-polling sync countdowns and live "Last Fetched" UTC timestamps.

1.6 🌐 Registry Syndication & Distribution Agent (@registry_syndication)

  • Governing Skill: mcp-registry-syndication
  • Python Module: RegistrySyndicationAgent
  • Capabilities:
    • Coordinates multi-registry lifecycle: Official MCP Registry (io.github.KumailQazi/krmaazha-binance-mcp), Smithery (smithery.yaml), Glama (glama.json), PulseMCP.
    • Guarantees version lockstep across pyproject.toml, server.json, CHANGELOG.md, and Git release tags.
    • Manages GitHub Device Flow publishing credentials and indexation health.

1.7 ⚖️ Regulatory Compliance & Trust Audit Agent (@compliance_audit)

  • Governing Skill: regulatory-compliance-audit
  • Python Module: ComplianceAuditAgent
  • Capabilities:
    • Enforces non-custodial software classification under European MiCA and FATF Travel Rule frameworks.
    • Audits M8ven Trust Verification grade across all 23 security, governance, and architectural labels.
    • Safeguards codebase against AST static analysis false-positives regarding trade keys.
    • Monitors dependency hygiene and automated pip-audit CVE screenings.

1.8 🎯 Execution Algorithm & Smart Routing Agent (@execution_algorithms)

  • Governing Skill: execution-algorithms
  • Python Module: ExecutionAlgorithmAgent
  • Capabilities:
    • Formulates institutional trade execution schedules (TWAP, VWAP, Iceberg order slicing).
    • Analyzes order-book depth bands ($\pm 1%$, $\pm 2%$) to cap market participation under $10%$.
    • Computes comprehensive Net Execution Friction (exchange fee + half spread + market impact slippage).
    • Prepares structured execution batches for user confirmation at the local physical gate.

2. Multi-Agent Orchestration Protocols

sequenceDiagram
    participant User
    participant Orch as Master Orchestrator
    participant Quant as Quantitative Agent
    participant Exec as Execution Agent
    participant Sec as Security Agent
    participant Comp as Compliance Agent
    participant Proto as Protocol Agent
    participant Reg as Registry Agent
    participant UX as UX Agent

    User->>Orch: Query market / Propose order / Deep audit
    Orch->>Quant: Step 1: Liquidity depth & closed-bar technicals
    Quant-->>Orch: Spread (bps), ±1% depth, indicators, initial R:R
    Orch->>Exec: Step 2: Algorithmic slicing (TWAP/VWAP) & Net Friction
    Exec-->>Orch: Execution trajectory & participation rate
    Orch->>Sec: Step 3: Key isolation & physical confirmation gate
    Sec-->>Orch: Read-only confirmed, gate engaged
    Orch->>Comp: Step 4: MiCA non-custodial check & M8ven trust grade
    Comp-->>Orch: Non-custodial verified, CVE clean
    Orch->>Proto: Step 5: FastMCP schema & CanonicalOrder Decimal check
    Proto-->>Orch: CanonicalOrder compliant, error scrubber active
    Orch->>Reg: Step 6: Registry & version lockstep check
    Reg-->>Orch: PyPI / MCP Registry in sync
    Orch->>UX: Step 7: Microcap formatting & epistemic labeling
    UX-->>User: Synthesized Institutional Dossier & Bottom Line

3. Institutional Quality Gates (IMPROVEMENTS.MD)

Every final analysis synthesized by the Orchestrator MUST satisfy these institutional gates:

  1. Exact Instrument Identification:
    • State Base/Quote, Venue (e.g. Binance Spot), Market Type (SPOT), and ISO-8601 UTC Timestamp.
    • Include venue disclaimer: "Quotes, depth, and volume represent Binance Spot specifically."
  2. Signal Reproducibility:
    • Explicitly list: Timeframe interval, whether signal bar is CLOSED or FORMING, indicator periods (EMA 12/26, RSI 14, MACD 12/26/9, ATR 14), and formula version.
  3. Executable Liquidity Audit:
    • State touch spread in basis points (bps), $\pm 1%$ order-book depth in USDT, and recent 24h volume.
    • If 1% depth is thin, append: ⚠️ Thin Book: Execution subject to elevated slippage.
  4. Complete Trade Setup & Algorithmic Slicing:
    • Direction: LONG, SHORT (with spot cash-hedge context), AVOID, or NEUTRAL.
    • Entry Trigger: Exact observable condition (e.g. "Confirmed 1h close above $X").
    • Invalidation Level: Stop-loss price, with explicit warning of execution slippage.
    • Net Reward-to-Risk: Compute gross R:R and Net R:R after deducting round-trip fees (20 bps) and estimated slippage.
    • Execution Slicing: If notional exceeds $5%$ of book depth, recommend TWAP/Iceberg slicing.
  5. Epistemic Classification Tags:
    • Prefix every claim with:
      • [VERIFIED FACT]
      • [QUANTITATIVE ESTIMATE]
      • [ANALYST INTERPRETATION]
  6. Regulatory & Non-Custodial Integrity:
    • Ensure the server process never holds or executes with trade keys.
    • Physical confirmation gate is mandatory for all capital allocation.
  7. Standardized Bottom-Line Synthesizer:
    • Conclude every analysis with the mandated single-sentence bottom line:

      Bottom Line: "At [timestamp], [instrument] is [state]. The setup is valid only if [observable condition]; it is invalidated by [condition]. Costs/risks are [items]. Otherwise, no trade."

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.