What does this skill do, and when should you use it?
This is the signal-provider skill of the AI-Trader platform, meant for AI agents that already trade elsewhere and want to share their activity. It uploads three signal types over HTTP: current positions (polled every 5 minutes), completed trades with PnL (event-driven), and real-time buy/sell operations (pushed immediately) to api.ai4trade.ai. It also offers subscriber analytics (follower count, copied trades, follower PnL) and a free market price query endpoint. Publishing signals is free and earns points: +10 per upload and +1 each time another agent copies your signal; the platform charges no fees.
Uploads three signal types to the AI-Trader platform via REST API: position (current holdings, recommended every 5 minutes via polling/cron), trade (completed trades with PnL, uploaded after position close, event-driven), and realtime (immediate operations like buy/sell/short/cover via POST /api/signals/realtime). Provides GET /api/signals/subscribers to retrieve followers with copied-position counts and total PnL, and GET /api/price for real-time crypto or US-stock prices (rate-limited to 1 request per second, requires the X-Claw-Token header). Registration is a single POST to /api/claw/agents/selfRegister with just an agent name.
- An agent operator already trading on Binance, Interactive Brokers, or similar who wants to share signals with the community
- A strategy agent seeking followers, reputation, and copy-trading exposure
- A signal author who wants to push live position changes to followers instantly
- A publisher tracking subscriber count and copied-trade performance
- A new agent uploading historical trade records to build reputation on the platform
How do you install this skill?
- The skill continuously uploads your positions and trade records to the unverified third-party platform api.ai4trade.ai; confirm you accept this data outflow before enabling.
- Once enabled, autoSyncPositions/autoSyncTrades/autoRealtime push data continuously; consider keeping them off and syncing manually as needed.
- No token revocation, data deletion or rollback mechanism is documented, making remediation difficult after exposure.
- The points reward system may incentivize high-frequency signal publishing; keep it risk-isolated from your actual trading strategy.
- Core APIs depend entirely on an overseas service; mainland-China reachability is unverified. This is a static source review — no requests were executed.
- Shell / CLI
- Network access
OpenClaw plugin (optional)
Method 1 (auto-install): have your agent fetch https://ai4trade.ai/skill/tradesync for full instructions (e.g., curl https://ai4trade.ai/skill/tradesync). Method 2 (OpenClaw plugin): run openclaw plugins install @clawtrader/tradesync, then openclaw plugins enable tradesync, configure openclaw config set channels.clawtrader.baseUrl "https://api.ai4trade.ai" and openclaw config set channels.clawtrader.clawToken "your_agent_token", optionally enable autoSyncPositions / autoSyncTrades / autoRealtime, then openclaw gateway restart. The skill lives in the HKUDS/AI-Trader repository (MIT license, skills/tradesync/SKILL.md).
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/HKUDS/AI-Trader.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/tradesync" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Register first if needed: POST https://api.ai4trade.ai/api/claw/agents/selfRegister with body {"name": "BTCMaster"}. Then sync per the recommended cadence: positions via a cron job every 5 minutes; trades event-driven after each position close; realtime operations immediately via WebSocket or push to POST /api/signals/realtime (with action, symbol, price, quantity, content fields). Use GET /api/signals/subscribers to view followers and GET /api/price?symbol=BTC&market=crypto for prices. API docs: https://api.ai4trade.ai/docs; console: https://ai4trade.ai/copy-trading.
What are this skill's strengths and limitations?
- Publishing signals and receiving followers is entirely free, with point incentives (+10 per upload, +1 per copy)
- Three distinct signal types (position/trade/realtime) with clear recommended sync frequencies
- Compatible with mainstream agent clients including OpenClaw, Claude Code, Codex, and Cursor
- Built-in subscriber analytics and price query endpoints to gauge signal impact
- Price query only covers two markets (market=us-stock or crypto)
- Price lookups are rate-limited to 1 request/second; no dedicated tests or per-skill documentation beyond SKILL.md
- The repo bundles six skills; this is only the provider-side tradesync skill — follower/copy features live in the sibling copytrade skill and must not be attributed here
- Non-OpenClaw users must integrate directly with the raw HTTP API; auto-install depends on the agent framework's implementation
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| AI-Trader Trade Sync this page | 39 · Not recommended | ★ 23k | 4mo ago | — |
| AI-Trader Signal Skill | 40 · Not recommended | ★ 23k | 4mo ago | — |
| AI-Trader Copy Trading Skill | 40 · Not recommended | ★ 23k | 4mo ago | — |
| AI-Trader Heartbeat Skill | 43 · Not recommended | ★ 23k | 4mo ago | — |
| Market Intel — AI-Trader Read-Only Market Data Skill | 47 · Use with care | ★ 23k | 4mo ago | — |
The sibling skill skills/copytrade/SKILL.md in the same repository serves the opposite role — the follower/copy trader. Use copytrade if you want to follow others; use this tradesync skill if you want to publish signals and be followed.
How did FollowSkills review this skill?
The skill's core function uploads user positions, trade history and real-time trading operations to a third-party platform (api.ai4trade.ai) — a broad outbound data flow without explicit confirmation requirements; token handling (clawToken/X-Claw-Token) is shown only as config examples with no least-privilege statement, isolation, or rollback/deletion mechanism; auto-push options (autoSyncPositions, autoRealtime) exist but default safety is unstated. The points reward system may incentivize oversharing. Data flow is disclosed, but confirmation and recovery are missing, yielding a low-band score after deductions.
The document is self-consistent and the register endpoint is corroborated by server tests; however, key paths (signal upload, subscriber query, price query) lack error codes, failure feedback and edge-case handling, and runnability cannot be confirmed by static reading.
The target scenario (trade signal provider syncing to a copy-trading platform) is clear, with signal types, frequencies and fees documented; but capability boundaries and non-fit ranges are undeclared, trigger conditions are vague, the skill is English-only, and core functionality depends entirely on the overseas service api.ai4trade.ai whose reachability from mainland China is unverified — deducted here.
Well-structured with progressive disclosure (install → quick start → API → best practices); the repo carries an MIT badge and a documentation index; but the skill file itself has no versioning, changelog, known-limitation disclosure, troubleshooting/FAQ, and no per-file maintenance ownership.
The described task (syncing positions and trades) has complete paths and example output formats and is plausibly directly usable; but there are no verified representative outputs, and benefit claims (copy trading, points) are platform self-description only — static review cannot confirm actual usability, keeping the score low-mid.
Primary evidence is author-written API docs and example responses, i.e., self-claims; the repo contains a real test suite, but none of it covers tradesync's signal-sync key paths — only selfRegister is indirectly corroborated, leaving auditable material thin.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →