Finance & Investment Banking portfolio-managementcrypto-tradingnav-trackingposition-monitoringopenclawtrade-bookhedge-fund

AIMM Portfolio Management Skill

Get an instant snapshot of your AI hedge fund portfolio: NAV, exposure, and all open positions in one query.

FollowSkills review · FSRS-2.0
Not recommended
40/ 100 5-point scale 2.0 / 5
1 2 3 4 5 6
1Trust12 / 25 · 2.4/5

The SKILL.md declares a read-only state provider ('No decision authority — provides data only'), with no fund movement, external writes, or sensitive-data exfiltration; risk surface is small. However, it does not disclose which accounts/exchanges data flows through, required API key privileges, failure rollback/recovery paths, and the publisher is unverified; data sources are only loosely named ('exchange/OMS').

2Reliability7 / 20 · 1.8/5

The capability claims are broadly consistent with the LangGraph architecture and trade_book dependency documented in README/pyproject, and persona.md gives explicit NAV/margin formulas and a JSON output example. But the interfaces of 'trade_book' and 'execution engine adapter' are undefined in the skill files, the query endpoints' error handling and failure feedback are absent, and key-path reproduction cannot be confirmed statically.

3Adaptability5 / 15 · 1.7/5

The scenario (querying NAV, positions, available margin) is clear with two natural-language trigger examples; but non-fit boundaries (real-time quotes, multi-account, venue support), trigger precision, and false-trigger safeguards are undeclared. Core function depends on overseas LLM/exchange APIs with no Chinese-language support or mainland-China reachability notes.

4Convention6 / 15 · 2.0/5

The SKILL.md is cleanly structured (capabilities/data sources/interface/dependencies) and persona.md layers reasonably with an output example; however the skill file is very thin — no install notes, versioning, changelog, known-limitation disclosure, or FAQ. Maintenance responsibility is only implicit at repo level via SECURITY.md; AGPL-3.0 licensing is clear at the repo level.

5Effectiveness6 / 15 · 2.0/5

As a read-only state query skill, the declared NAV/position lookups would be directly useful with structured JSON output if the platform stack runs, and README documents run/test commands; but static review cannot verify output correctness, and the skill reads more as an internal system component than a standalone user-facing skill, so its value claim is narrow.

6Verifiability4 / 10 · 2.0/5

The formulas and output schema in persona.md are auditable, and the repo has CI (ci.yml) and pytest infrastructure plus reproducible commands in README; but there is no dedicated test evidence covering this skill's key paths (portfolio state queries), and the performance figures are author claims explicitly not live-edge claims, with low static confidence.

Evidence confidence:Low Reviewed Sep 09, 2026 Reviewed revision dd7115076904
Before you use it
  • This skill is an internal read-only state component of the repo, not a standalone installable tool; it presumes the full platform stack (Docker, Postgres, LLM API key) is already running.
  • Core function depends on overseas LLM and exchange APIs with no mainland-China reachability or Chinese-language documentation.
  • Query endpoints document no error handling or failure feedback; the actual contracts of trade_book and the execution-engine adapter must be verified in source code by the user.
  • Backtest/return figures are author-claimed and LLM-dependent; do not treat them as live-return expectations, and note this is not investment advice.
  • The publisher is unverified and maintenance relies solely on a community repository; audit the code yourself before use.
See the full review method →

What does this skill do, and when should you use it?

This skill is one of 14 bundled in olaxbt/ai-market-maker (AIMM, Agentic AI Hedge Fund OS), located at src/agents/governance/portfolio_management/. It does not trade; it reads the trade book, the execution engine's position feed, and cash balances from the exchange/OMS, then returns a portfolio snapshot with NAV, exposure, and positions. It is invoked via a simple query interface (/portfolio_management?action=state or ?action=positions) or plain-language questions like "What's my current NAV?". It depends on the trade_book skill and an execution engine adapter from the same repo, so it only works as part of the AIMM platform, not standalone.

Reads the trade book (in-memory plus persistent), the execution engine's position feed, and cash balances from the exchange/OMS; exposes get_state() to return a portfolio snapshot containing NAV, exposure, and position details; serves two query actions — action=state (portfolio snapshot) and action=positions (all open positions); answers direct queries such as "What's my current NAV?" and "Show all open positions".

  1. A quant trader running paper trades on Binance Testnet via AIMM who wants current NAV and exposure on demand instead of querying the database
  2. An operator of a multi-agent trading stack who needs to verify all open positions before execution or during a risk review
  3. A developer debugging AIMM's execution engine who uses it to check that position records match cash balances
  4. A portfolio manager rotating across multiple assets (BTC/ETH/SOL) who wants one unified view of positions and exposure

What are this skill's strengths and limitations?

Pros
  • Minimal interface: two query actions plus natural-language questions return NAV, exposure, and positions
  • Clear data provenance: trade book, execution engine position feed, and exchange/OMS cash balance cross-checked
  • Inherits the platform's quant discipline: fully traceable execution records and risk governance
  • Open source under AGPL-3.0 with auditable code
Limitations
  • Tightly coupled to AIMM: the trade_book skill and execution engine adapter must be present; it cannot run outside the repo
  • Read-only — no position optimization, rebalancing, or trade advice (the broader Portfolio Management desk duties in the README are not declared in this SKILL.md, which only covers snapshot queries)
  • SKILL.md does not document refresh rates, performance characteristics, or error handling
  • Execution venues are Binance Testnet, Hyperliquid dry-run, and Futu paper trading — no live-trading evidence

How do you install this skill?

The skill ships inside the full ai-market-maker repo with no standalone install path: git clone https://github.com/olaxbt/ai-market-maker.git && cd ai-market-maker, then follow the README (pip install uv; install TA-Lib, Conda recommended; uv sync --extra dev). OpenClaw installation for the repo: claw install https://github.com/olaxbt/ai-market-maker. Note: this specific skill has no dedicated OpenClaw packaging; the documented openclaw/ package covers the whole repository.

How do you use this skill?

Prerequisite: the AIMM platform must be running (full stack via docker compose -f docker-compose.prod.yml up --build -d, or CLI trading mode via uv run python src/main.py). Then query: /portfolio_management?action=state for a snapshot with NAV, exposure, and positions; /portfolio_management?action=positions for all open positions; or ask in plain language, e.g. "Show all open positions". Run database migrations (alembic upgrade head) before first use.

How does this skill compare with similar options?

Sibling skills in the same repo (e.g., the trade book, execution engine) are its data upstream; this is the single entry point answering "what is the current portfolio state?". The README names no external competitors to compare against.

FAQ

Can it decide what to buy?
No. It only returns a portfolio snapshot (NAV, exposure, positions). Signal generation, risk-weighted allocation, and trade proposals are handled by other agents in the AIMM repo.
Can I use it without running the AIMM platform?
No. It reads from the trade_book skill and an execution engine adapter; both must be running. Outside the repo it has no data source.
Does it connect to real exchanges?
The README describes Binance Testnet paper trading, a Hyperliquid adapter in dry-run, and Futu simulated orders; live trading is not documented.
What about cost and licensing?
AGPL-3.0 open source. If you run modified software as a network service, AGPL obligations (including source offer to users) may apply. Additionally, metered usage through the OlaXBT Nexus stack is settled on BNB Chain, requiring BNB or supported stablecoin credits.

More skills from this repository

All from olaxbt/ai-market-maker

Finance & Investment Banking

Risk Guard

A hard-veto risk gate that validates signals against drawdown and policy limits before execution — the final safety layer of the AIMM agentic hedge fund OS.

Finance & Investment Banking

AIMM: OpenClaw Multi-Agent Trading Skill

A hedge-fund-style multi-agent crypto trading system packaged for OpenClaw, with a hard risk-control veto and quant-grade backtesting.

Finance & Investment Banking

Profile Agent — Trader Profiler

Turns a short questionnaire on risk tolerance and horizon into a trader profile with weight deltas and a readable narrative, feeding personalized configuration into a multi-agent trading system.

Finance & Investment Banking

News Narrative Miner

Mines breaking news and narrative phases for a given crypto asset, scoring impact and decay so trading teams know what story is moving the market.

Finance & Investment Banking

Whale Behavior Analyst (4.1)

Answers the critical crypto question — are whales dumping? — using on-chain large-transaction and exchange-flow data.

Finance & Investment Banking

Statistical Alpha Engine (Skill 2.2)

Ranks every asset in a trading universe cross-sectionally to answer questions like 'where does BTC rank right now?'

Finance & Investment Banking

Liquidity Order Flow Analyst (4.2)

Assesses slippage risk and order-book imbalance from depth data, giving crypto trading decisions liquidity-side evidence.

Finance & Investment Banking

Pro Bias Analyst

Determines whether smart money is buying or selling an asset like BTC by fusing ETF flows, futures basis, and options skew into a single pro bias score.

Finance & Investment Banking

Pattern Recognition Bot (2.1)

A technical-pattern recognition skill for crypto traders: given market data, it returns the forming pattern label, setup score, support level and reasoning so you know what structure the market is building.

Finance & Investment Banking

Technical TA Engine

A standardized technical-analysis layer for crypto trading that computes an indicator bundle (RSI, MACD, etc.) from OHLCV data and returns a 0–100 composite score with confidence.

Finance & Investment Banking

Retail Hype Tracker (Skill 3.1)

Detects whether retail FOMO is building on a trading pair, using social sentiment, funding rates and divergence warnings, so you don't buy the top of a hype cycle.

Finance & Investment Banking

Policy Orchestrator

Centralized policy check and routing for the AIMM agentic trading system: decide whether a given ticker is allowed for a user, with reason and policy ID.

Finance & Investment Banking

Monetary Sentinel

Reads macro and on-chain liquidity data to classify the market as Risk-On or Risk-Off, returning a liquidity score with cited reasoning.

Related skills