committee

Original：🇨🇳 Chinese

Translated

Multi-Model Investment Committee. Use this skill when the user says "start a meeting", "investment committee", "multi-model analysis", or "let three models review together".

1installs

Sourceallenai2014/ai-investment-advisor

Added on2026-02-08

NPX Install

npx skill4agent add allenai2014/ai-investment-advisor committee

SKILL.md Content (Chinese)

View Translation Comparison →

/committee - Investment Committee (Data-Driven Version)

Let Claude, Codex, and Gemini conduct independent analyses based on the same objective data, perform cross-validation when necessary, and finally extract actionable consensus.

Usage

```
/committee
```
- Convene an investment committee
```
/committee How to operate dividend ETFs
```
- Convene for a specific question
```
/committee consensus
```
- Summarize existing viewpoints from the three models

Execution Steps (Recommended Enhanced Version)

0. Select Mode

Simplified Version: One round of independent analysis → Summarize consensus
Enhanced Version: R1 Independent Analysis → R2 Cross-Validation → Summarize consensus
Default: Ask the user; use the enhanced version if not specified

Step 1: Prepare Unified Input Data

Obtain Market Data

bash

cd "Stock Market Info" && python3 scripts/fetch_market_data.py

The script supports
technicals
(MA/RSI/MACD/range position). If it's too slow, you can disable it in
MODULES
.

Read Context Read
```
Stock Market Info/Config/Context.md
```
, and supplement risk preferences and constraints from
```
Profile.md
```
,
```
Principles.md
```
, and
```
Insight.md
```
Determine Today's Decision Question Ask the user what issue they want the committee to discuss today
Generate Input File Integrate the above content and save it to
```
Stock Market Info/Committee/Input/YYYY-MM-DD-Input.md
```

Use template:
```
Committee/Templates/prompt_template.md
```
Must include a "Data Quality Checklist", clearly indicating missing fields (valuation/capital/technical, etc.)
```
technicals
```
and transaction data of
```
holdings
```
/
```
watchlist
```
from the output of
```
fetch_market_data.py
```
must be included

Step 2: Round1 Independent Analysis

Based on the input data, output according to

Committee/Templates/opinion_template.md

:

Overall market judgment
Position signal matrix and operation suggestions (each target suggestion must reference >=2 objective data fields)
Risk warning
Answer the decision question
Self-assessment of confidence level

Save to:

```
Committee/Opinions/Claude.md
```
```
Committee/Opinions/Codex.md
```
```
Committee/Opinions/Gemini.md
```

Step 3: Cross-Validation (Enhanced Version)

Provide the R1 viewpoints to other models, requiring:

Only point out omissions/misinterpretations/over-inferences based on the same data
Allow adjustments: suggestions, confidence, risk warnings
Output R2 correction table

Save to:

```
Committee/Opinions/Claude-R2.md
```
```
Committee/Opinions/Codex-R2.md
```
```
Committee/Opinions/Gemini-R2.md
```

Step 4: Summarize Consensus

When the user indicates that all viewpoints have been collected:

Read Opinion Files Prioritize using R2; use R1 if R2 is not available
Compare Target by Target

3/3 agreement → Strong consensus
2/3 agreement → Weak consensus
All different → Disagreement

Weight and Risk Control Rules

Weight = Confidence level × Data completeness
Viewpoints that do not reference objective data will be downgraded in weight
High-risk warnings from any model must be included in the summary

Generate Consensus Report Output according to

Templates/consensus_template.md

and save to

Committee/Sessions/YYYY-MM-DD.md

Step 5: Record Recommendations to Track

Record strong consensus recommendations in the "Adoption Records" section of
```
Stock Market Info/Config/Insight.md
```
Set the status to "Pending Adoption", and verify later with
```
/trade
```

Output Format Requirements

Uniformly use
```
Committee/Templates/opinion_template.md
```
Missing fields must be marked as "Missing", self-completion is prohibited
Key suggestions must include "Key Data Basis"

Consensus Extraction Rules (Supplementary)

If there is inconsistency in the understanding of key data (e.g., different interpretations of RSI/trends), enter the "Disagreement Zone" and mark the data dispute
If there is a large data gap, the original strong consensus will be downgraded to weak consensus
If a "risk red line" appears (e.g., RSI>75 and price is much higher than MA60), a caution prompt must be given and confidence level reduced

Template File Locations

Input prompt template:
```
Committee/Templates/prompt_template.md
```
Opinion output template:
```
Committee/Templates/opinion_template.md
```

Consensus summary template:

Committee/Templates/consensus_template.md