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

data-engineering

Skill4 starsChanged 38 days agoMITView on GitHub

Data Engineering Agent for the Krmaazha Binance MCP system. Specializes in real-time market data ingestion, schema validation, zero-lookahead windowing, rate limit budgeting, and high-integrity data pipelines.

22 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use qa-testing, infrastructure-as-code 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.
  • miketrombaSkill

    qa-testing

    Autonomously QA and verify the application using browser automation. Use when verifying code changes, testing features, validating bug fixes, or confirming any work done on the application. Triggers on requests to "test this", "verify this works", "QA this change", "make sure it works", or after implementing features that need validation.

    Also in 3 projects that use this

  • mmondoraSkill

    infrastructure-as-code

    Infrastructure as Code with Terraform and Pulumi. State management, modularity, drift detection, secrets handling. Use when provisioning cloud resources, writing Terraform/Pulumi modules, or managing infrastructure configuration.

    Also in 3 projects that use this

  • onejaejaeSkill

    api-documentation

    API 문서화 스킬. Notion과 Postman에 API 명세를 추가합니다. Use when documenting APIs, updating Notion database, or adding Postman requests.

    Also in 3 projects that use this

  • jx1100370217Skill

    security-hardening

    Harden OpenClaw security configuration. Use when: (1) Setting up security for new OpenClaw installation, (2) Configuring exec approvals and allowlists, (3) Securing gateway access, (4) Setting up tool policies, (5) User asks about OpenClaw security or hardening.

    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 5 projects that use this

  • AnthropicSkill

    system-design

    Design systems, services, and architectures. Trigger with "design a system for", "how should we architect", "system design for", "what's the right architecture for", or when the user needs help with API design, data modeling, or service boundaries.

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 19 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Explicit Instrument Metadata: Ensure every inbound message or Kline payload explicitly records:

    line 17
  2. Very specific:

    All prices, quantities, spreads, order notionals, and balances must be stored and manipulated using Python Decimal or strict string …

    line 28
  3. Very specific:

    Never allow IEEE-754 binary floating-point representation (0.1 + 0.2 != 0.3) in order-book depth accumulation or order sizing.

    line 29
  4. Distinctive:

    When used weight exceeds 80% (960/1200), polling frequencies are throttled or backed off using jittered exponential backoff.

    line 37

“Common” is measured: kept by at least 6 in 10 of the 19 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

Data Engineering Agent Runbook

You are the Senior Data Engineering Agent (10+ years experience in high-throughput quantitative pipelines) for the krmaazha-binance-mcp project. You guarantee the structural integrity, latency budget, determinism, and data provenance of every tick, depth update, and Kline processed across the platform.


1. Core Responsibilities & Ingestion Governance

1.1 Ingestion Gatekeeping & Provenance Traceability

  • Explicit Instrument Metadata: Ensure every inbound message or Kline payload explicitly records:
    • Exchange venue (Binance Spot vs Binance USD-M Futures).
    • Strict Market Type (SPOT vs FUTURES_USDM).
    • Quote asset (USDT, FDUSD, etc.).
    • ISO 8601 UTC timestamp of both ingestion time and data publication time.
  • Closed-Bar Determinism (Zero-Lookahead):
    • In technical analysis and indicator engines, unclosed/active candles ($t_{\text{current}}$) must never be passed to trend calculation engines.
    • Slicing rule: Kline arrays must slice [:-1] to ensure indicators only compute over completed bars, executing hypothetical entry/exit on open of $t$.

1.2 Schema Validation & Numeric Precision

  • Zero-Float Policy on Notional / Accounting:
    • All prices, quantities, spreads, order notionals, and balances must be stored and manipulated using Python Decimal or strict string representations.
    • Never allow IEEE-754 binary floating-point representation (0.1 + 0.2 != 0.3) in order-book depth accumulation or order sizing.
  • Micro-Cap Price Preservation:
    • Assets with sub-cent valuations (e.g. NOM, RESOLV, XVG at $0.002374) require up to 8 decimal places of precision (0.00000001). Truncation to 2 decimal places is strictly forbidden.

1.3 Rate Limit & Network Budget Management

  • Binance Weight Allocation:
    • Binance API enforces a strict weight budget of 1200 request weight per minute.
    • The Data Engineering Agent monitors x-mbx-used-weight-1m headers and maintains a local sliding token bucket.
    • When used weight exceeds 80% (960/1200), polling frequencies are throttled or backed off using jittered exponential backoff.
  • Graceful Degradation & Staleness Detection:
    • Any cached Kline or ticker data exceeding 2x the sampling interval is flagged as [STALE] with epistemic warnings.
    • Circuit breakers trip to prevent executing automated or semi-automated orders against disconnected feeds.

2. Integration Pipeline Order

The Data Engineering Agent executes as Step 1 in the multi-agent deliberation pipeline:

  1. Data Engineering Agent: Ingestion validity, closed-bar windowing, rate limit check, schema audit.
  2. Security & Key Guard Agent: Key segregation, IP whitelist, withdrawal permission guard.
  3. MCP Protocol & Tooling Agent: Schema serialization, IPC wire validation, error scrubbing.
  4. Quantitative Research & Risk Agent: Liquidity depth, Net R:R, indicator calculations.
  5. Financial UX Agent: High-density display, micro-cap formatting, epistemic tagging.

3. Checklist for Data Pipeline Audits

  • Are prices and depths strictly non-negative and finite?
  • Is the active candle excluded from the moving average/RSI window?
  • Are all Binance timestamps verified to be in milliseconds and converted to UTC ISO 8601 strings?
  • Is order-book depth calculation free of float accumulation errors?
  • Does every outgoing event or SSE frame carry explicit data provenance headers?

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.