- Home
- Skills
- Business & Operations
- two witness trade check
two witness trade check
A two-witness trading validation system that enforces price-pattern history, news sentiment, and risk guard rules.
Free
Works with the AI Agents you already use
two witness trade check
Example session with this skill installed
I want to buy 50 shares of AAPL at 190.00 with a stop at 185.00. My account has 10k, no losses today, and no events. Here is the price and news data. Should I trade this?
- Read your context and instructions
- Compiled the two witness trade
I have processed the trade plan for AAPL using the two-witness verification protocol. The analysis included 400 days of price history and 8 recent news facts.
VERDICT: TRADE_CANDIDATE
The trade idea passed all witness checks and risk guard requirements.
Trade Analysis Report
| Witness | Stance | Details |
|---|---|---|
| Price Witness | BULLISH | Found 24 similar patterns in history. 95% bootstrap CI for returns is positive (+1.2% to +3.8%). |
| Walk-Forward | PASSED | Out-of-sample testing showed a 62% hit rate across 12 historical signals. |
| News Witness | BULLISH | Sentiment shift of +0.35 detected based on 4 unique stories regarding new product launches. |
| Risk Guard | PASSED | Risk at stop is $250 (2.5% of equity). Proposed position is 9.5% of total account. |
| Independence | PASSED | Price move and news events occurred on different dates; no single-event bias detected. |
Reasoning: Both independent witnesses agree on a bullish outlook. The price pattern has a statistically significant edge in historical walk-forward testing, and recent news sentiment has shifted positively. The risk guard confirms the $5.00 stop loss keeps the potential loss within acceptable account limits.
Next steps
- Verify the 190.00 entry price is still available in the live market.
- Confirm the stop loss is set at 185.00 immediately upon execution.
- Monitor for any new scheduled events that were not included in this check.
This is an educational check, not financial advice.
Connects securely to your tools. The creator never sees your data.
About this skill
Trading decisions often suffer from emotional bias, lack of evidence, or poor risk management. This skill enforces a "two-witness" protocol, requiring both price-pattern history and news sentiment to align before a trade is considered valid. It replaces gut feelings with a systematic verification process that runs locally on your machine.
What it does
- Multi-source validation cross-references historical price patterns against recent news sentiment to confirm trade signals.
- Walk-forward testing replays signals over historical data to verify if the strategy actually beats chance before suggesting a trade.
- Automated risk guarding calculates position sizing and enforces stop-loss limits to ensure no single trade exceeds account risk parameters.
- Exit monitoring evaluates existing positions to provide specific HOLD, REVIEW, or EXIT recommendations based on changing market evidence.
- Event awareness blocks new trades near major scheduled events like earnings or rate decisions to avoid high-volatility traps.
How it works
- Input collection requires providing daily closing prices (minimum 350 days), recent news articles, and your specific trade plan including account equity.
- Fact extraction involves identifying specific topics and polarities from news text to generate a structured data file for the scripts.
- Script execution runs the bundled Python analysis tools, starting with
check.pyfor new trades orexit_check.pyfor existing holdings. - Verdict reporting delivers a final status (e.g., TRADE_CANDIDATE or NO_TRADE) based on the combined output of price, news, and risk witnesses.
Frameworks & tools
This skill is built with Python 3.9 using only the standard library. It requires no external dependencies, API keys, or network access, ensuring all financial data remains local.
Why this beats prompting it yourself
Standard LLM prompts often hallucinate price trends or give vague financial advice without calculating actual risk. This skill uses deterministic Python logic to enforce strict statistical thresholds, such as 95% bootstrap confidence intervals and out-of-sample walk-forward testing, which a text-only prompt cannot perform.
Use cases
- Validating a "buy" signal for a specific stock by checking if similar historical patterns were profitable.
- Deciding whether to hold or sell a current position when news sentiment turns negative.
- Stress-testing a trading bot's logic by adding a risk-guard layer that prevents oversized positions.
- Auditing a trade idea to ensure it isn't being placed too close to a high-risk earnings event.
Known limitations
Requires at least 350 days of price data for meaningful analysis. It does not connect to brokers or execute real trades. Results are educational and do not account for slippage or trading fees.
Requires 350+ days of price history. No broker integration or live execution. Does not account for slippage or fees. Python 3.9+ standard library only.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 14 days ago
- Free with an Agensi account
Requires