IBKR Autonomous Trader
v1.0.0Autonomous paper trading research agent for Interactive Brokers (IBKR). Use this skill whenever the user wants to connect to IBKR, execute trades, manage pos...
⭐ 0· 60·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the code and files: strategy, execution, risk, news, decision, and performance modules all align with an Interactive Brokers paper-trading agent. No unrelated cloud credentials or unrelated binaries are requested.
Instruction Scope
SKILL.md and scripts stay within trading scope (connect to local TWS/IB Gateway, fetch market data, generate signals, place orders, fetch Google News RSS). Two concerns: (1) SKILL.md instructs the agent to 'Always use this skill for trading-related tasks' which is an overly broad trigger instruction and increases invocation frequency; (2) the skill includes an automatic 'self-improvement' loop that updates strategy parameters/weights (writes to memory/strategies.json), which can change runtime behavior over time and should be reviewed/controlled.
Install Mechanism
No install spec that downloads remote archives; skill is delivered as source files and expects Python packages to be installed (pip). No evidence of downloads from untrusted URLs or extract/install steps.
Credentials
The skill requests no environment variables or external API keys — it connects to a locally-running TWS/IB Gateway (127.0.0.1:7497 for paper), which is appropriate. Note: news fetching uses network access to Google News RSS (expected). Also: SKILL.md / code show minor naming mismatches (SKILL.md refers to 'ENABLE_LIVE_TRADING' while config uses 'enable_live_trading'), and feedparser is used by news_engine but not listed in the SKILL.md 'pip' compatibility line, so an extra dependency must be installed.
Persistence & Privilege
always:false (good). The skill writes logs and persistent JSON files under its directory (memory/*.json, logs/*) and updates strategies.json as part of learning — expected for a trading agent but worth noting because it persistently modifies local state and can autonomously tune weights over time. It does not appear to modify other skills or system-wide agent configs.
Assessment
This package appears coherent for paper trading with IBKR, but review and take these precautions before use: 1) Run only in paper mode initially — the code can perform live trades if you flip the live flag; confirm live trading requires an explicit 'CONFIRM' prompt but test with no live enabled. 2) Install required Python packages (note: feedparser is used by news_engine but not listed in SKILL.md's pip list). 3) Inspect and control the self-improvement flow: it writes to memory/strategies.json and can update decision weights/params automatically — consider disabling or reviewing automated updates until you validate behavior. 4) Ensure TWS/IB Gateway runs locally and you understand account scope (the agent reads account summary and can close positions). 5) The instruction 'always use this skill for trading-related tasks' is broad — consider limiting invocation or reviewing triggers. If you need higher assurance, run the code in an isolated environment (dedicated paper account, network-restricted VM) and perform a code review of the omitted/truncated files referenced by the repository.Like a lobster shell, security has layers — review code before you run it.
latestvk97bwcxn3bbpgwsbfqtrgbr9fn84fjyw
License
MIT-0
Free to use, modify, and redistribute. No attribution required.