Maintainer of this skill?
by cplj
Holdings exposure, what-if scenarios, cost and capacity, backtest checks, and published research.
# Add to your Claude Code skills
git clone https://github.com/cplj/axiocairn-mcpGuides for using ai agents skills like axiocairn-mcp.
See how axiocairn-mcp compares with popular alternatives.
axiocairn-mcp is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by cplj. Holdings exposure, what-if scenarios, cost and capacity, backtest checks, and published research. It has 58 GitHub stars.
axiocairn-mcp's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/cplj/axiocairn-mcp" and add it to your Claude Code skills directory (see the Installation section above).
axiocairn-mcp is primarily written in Python. It is open-source under cplj on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh axiocairn-mcp against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
Research & Learning publishes exploratory papers and research notes with their methods and limitations; tutorials are coming on November 6, 2026.
Axiocairn MCP lets an AI assistant check holdings exposure, run what-if scenarios, estimate trading cost and capacity, review backtest discipline, and read published research reports with their data dates and limits.
Traditional Chinese · Connection guide · Service status
"I hold 50% NVDA, 30% MSFT, 20% SPY. How exposed am I to the market?"
| Reading | Value |
|---|---|
| Largest position | NVDA, 50% |
| Technology, direct weights (NVDA + MSFT) | 80% |
| Concentration (HHI) | 0.38 |
| Daily volatility | 1.58% |
| Beta to SPY | 1.47 |
| Beta on SPY down days (121 of 252) | 1.58 |
| Correlation with SPY: NVDA / MSFT | 0.66 / 0.39 |
Data date 2026-10-30. Window 2025-10-30 to 2026-10-30, 252 daily returns, licensed market data. This is a simulated holding with fixed weights, not an account and not advice. The readings describe how the three symbols moved together in that window, not a forecast. The 80% counts only the two direct technology positions, not what SPY holds.
Works with Claude (claude.ai web).
Access follows your plan.
In your AI app, add a custom connector named Axiocairn with this address. For ChatGPT, use the ChatGPT address in the table below.
https://axiocairn.com/mcp
Sign in with Google or GitHub when asked.
Ask:
What Axiocairn reports are available today?
What is the data date of the latest Axiocairn report you can see, and is it today's edition?
In a few short bullets, what are the limits of Axiocairn reports?
Connect to the hosted server at https://axiocairn.com/mcp using Streamable HTTP and the account authorization flow. No local server installation is needed.
| Client | URL | Connection status |
|---|---|---|
| Claude (claude.ai web) | https://axiocairn.com/mcp | Verified on 2026-10-27 |
| ChatGPT (web) | https://axiocairn.com/mcp/chatgpt | Verified on 2026-10-27 |
| Grok | https://axiocairn.com/mcp/grok | Verification pending |
| Muse | https://axiocairn.com/mcp/muse | Verification pending |
| Other MCP clients (Claude Code, Cursor, Gemini CLI) | https://axiocairn.com/mcp | Verification pending |
"Verified" means we added the connector, authorized an account, and got answers to report questions in that client on the date shown. These are deployed, account-protected endpoints. A listed URL alone does not establish client compatibility. The examples below are configuration templates; they do not establish a successful connection in each client.
After authorization, request tools/list for the current tool names, descriptions, and input schemas. This minimal tools/call request asks for report timing:
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "get_report_schedule",
"arguments": {"language": "en"}
}
}
Keep the report link, data date, missing fields, and limitations in the answer. An expected publication time is not evidence that a report is available.
On claude.ai, open Customize → Connectors → Add custom connector, name it Axiocairn, and use https://axiocairn.com/mcp. This was verified on October 27, 2026. Claude Code has not been verified. For Claude Code, the official remote HTTP configuration supports this template using the general endpoint:
claude mcp add --transport http axiocairn https://axiocairn.com/mcp
Complete the client's authorization flow if offered. This command describes configuration, not verified Axiocairn compatibility. A verified Claude Desktop configuration is not currently published in the connection guide.
Use these values in a custom connector. This was verified in ChatGPT on the web on October 27, 2026.
Name: Axiocairn
Connector URL: https://axiocairn.com/mcp/chatgpt
Review the account authorization screen before approving access. Follow the connection guide for the current setup status.
Review the server configuration in Cursor before installing. When prompted, sign in to Axiocairn and review the OAuth authorization screen (the account access approval step). No API key belongs in this configuration.
The Cursor MCP documentation describes remote server entries in mcp.json. Merge this entry into the existing mcpServers object without replacing other servers:
{
"mcpServers": {
"axiocairn": {
"url": "https://axiocairn.com/mcp"
}
}
}
Cursor is not listed as verified in Axiocairn's published guide. This template uses the general endpoint and requires a compatible account authorization flow.
The Gemini CLI MCP documentation describes httpUrl in settings.json. Merge this entry into the existing settings:
{
"mcpServers": {
"axiocairn": {
"httpUrl": "https://axiocairn.com/mcp"
}
}
}
Gemini CLI is not listed as verified in Axiocairn's published guide. This template requires compatible remote HTTP and account authorization support.
Muse verification is pending. Add a custom connector with these values:
Name: Axiocairn
Connector URL: https://axiocairn.com/mcp/muse
Follow the Muse guide, authorize the account, and refresh the tool list.
Tool discovery requires account authorization. Use the authorized server's tools/list response as the authority for the tools available to your connection; unauthenticated discovery returns an authorization error. Some tools are limited by plan.
| Tool | What it does |
|---|---|
analyze_holdings |
Review pasted holdings: exposures, sector concentration and a linked SPY stress estimate. |
check_portfolio |
Concentration, daily volatility, correlation and SPY beta of a basket. |
what_if |
Apply a stress scenario, an SPY move or a weight change to a prior holdings result. |
estimate_cost_capacity |
Estimate the cost of a hypothetical order and the size a stock or ETF can absorb. |
calculate_options |
Descriptive Black-Scholes-Merton prices and Greeks from dated historical inputs. |
check_backtest_discipline |
List supplied trial counts, multiple-testing adjustment and train/test gaps. |
check_idea |
Turn a symbol, direction and horizon into a falsifiable statement with its evidence gaps. |
get_screen |
Compare one symbol across dated simulated-position lists: listed, not listed, missing. |
get_bench_result |
Read a queued calculation by job_id: result, chart data and interpretation guide. |
get_data_quality |
Dataset dates, missing observations, duplicate rows and source-version changes. |
get_alerts |
Read recorded comparisons against limits you defined. |
get_alert_conditions |
Read your saved daily-close conditions and their versions. |
set_alert_condition |
Save or disable one daily-close condition; sends no notification. |
get_todays_reports |
Find reports for today or a given date, including the weekly House View. |
get_latest_report_by_type |
Retrieve the newest available report in a selected report family. |
search_report_archive |
Find published reports by type, date, ticker or language; samples are labelled. |
search_reports |
Search existing reports by ticker, date range and category. |
get_symbol_reports |
List the existing reports for a ticker, then retrieve the one you pick. |
get_house_view |
This week’s House View, or the next expected publication date. |
get_report_schedule |
Expected report times, time zone and recorded generation times. |
get_usage_guide |
Which report categories exist, what each is for and what generation costs in allowance. |
list_guides |
List help-center topics. |
get_guide |
Return one full help-center entry. |
get_plans |
Plans, allowances and how to upgrade, from the website plan settings. |
Sign in once when the connector prompts you. Use the same Google or GitHub account as your Axiocairn account. Review the requested access before approving it. Reauthorization may be needed after access expires or is revoked.
Reader is the free tier. It includes limited ticker look-ups, delayed market briefs, and a limited history window; some in-depth content is a preview. Samples and report access follow the account's permissions. See the current plan details for allowances and access conditions. Connecting a client does not expand those permissions.
If a connection fails, use the website to read the reports your account can access. Reconnect to refresh an outdated tool list. Do not paste account credentials into prompts or shared configuration files.
Read each report's data cutoff and publication date separately. An older report d