Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/aranjan/kite-mcp/claude-mdgit clone --depth 1 https://github.com/aranjan/kite-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/aranjan/kite-mcp/claude-md)<a href="https://agentmods.dev/instructions/aranjan/kite-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/aranjan/kite-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02140 | $0.02140 |
| Opus 5 | $0.01070 | $0.01070 |
| Sonnet 5 | $0.00428 | $0.00428 |
| Haiku 4.5 | $0.00214 | $0.00214 |
Grade D, and why
kite-mcp CLAUDE.md scanned grade D with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- PyPI recovery codes: ~/.config/pypi/recovery_codes.txt (chmod 600) Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
cd ~/kite-mcp && rm -rf dist/ && ./venv/bin/python -m build How it starts
The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md - Project Context for kite-mcp
Project Overview
kite-mcp is an MCP (Model Context Protocol) server for Zerodha Kite Connect that enables trading Indian stocks through natural conversation with any MCP-compatible AI assistant.
- Author: Amit Ranjan (aranjan)
- Repository: https://github.com/aranjan/kite-mcp
- PyPI: https://pypi.org/project/kite-mcp/
- Smithery: https://smithery.ai/server/@aranjan/kite-mcp
- License: MIT
Architecture
User (natural language) -> AI Assistant -> kite-mcp (MCP/stdio) -> Zerodha Kite API
- Built with FastMCP (from the
mcpPython package) - Runs as a stdio server (local, not hosted)
- Auto-authenticates using TOTP via
pyotp - Token cached at
~/.zerodha_kite_token.json(expires daily)
Broader System Context
kite-mcp is part of a multi-broker trading system:
Claude Desktop
├── kite-mcp → Zerodha portfolio + orders (auto TOTP)
├── zerodha-official → Free quotes for any NSE stock (browser OAuth)
├── icici-mcp → ICICI Direct portfolio + orders (Playwright TOTP)
├── finance MCP → Technical analysis (RSI, MACD, Bollinger, MA)
└── Slack MCP → Post to #kite-portfolio
Related projects:
~/icici-mcp/— ICICI Direct MCP server (separate repo)~/finance-mcp-wrapper.py— Async wrapper for finance-mcp-server~/trading-agent-prompt.md— Scheduled task prompts for daily/EOD/weekly reports~/test_claude_config.py— Validates Claude Desktop config~/improvements-26-march.md— Planned improvements backlog
Project Structure
~/kite-mcp/
src/kite_mcp/
__init__.py # Version string
auth.py # Shared auth: load_credentials, automated_login, get_authenticated_kite
server.py # FastMCP server with 14 @mcp.tool() decorated tools
cli.py # CLI entry points: login(), status()
tests/
test_auth.py # 7 tests for auth module
test_server.py # 3 tests for tool registration
test_cli.py # 2 tests for CLI
pyproject.toml # Hatchling build, entry points: kite-mcp, kite-mcp-login
smithery.yaml # Smithery directory config
CHANGELOG.md
CONTRIBUTING.md
SECURITY.md
LICENSE # MIT
README.md
logo.svg
promotion-activities.md # Marketing plan with draft posts
.github/
workflows/ci.yml # CI: ruff lint, syntax, imports, tests across Python 3.10-3.13
ISSUE_TEMPLATE/ # Bug report + feature request templates
pull_request_template.md
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 205 lines · 2,140 tokens per session scan D 9c89680df2c7
kite-mcp CLAUDE.md is an instructions file published in the GitHub repository aranjan/kite-mcp (5 stars, last pushed 5mo ago), licensed MIT. It adds 2,140 tokens to every session, about $0.0107 per session on Opus 5. A static security scan graded it D with 2 findings (asks for root, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
tasty-agent AGENTS.md
AGENTS.md instructions for ferdousbhai/tasty-agent, covering tasty-agent, code index, boundaries and commands.
tasty-agent CLAUDE.md
Claude Code instructions for ferdousbhai/tasty-agent, a project described as: Let Claude manage your tastytrade portfolio.
icici-mcp CLAUDE.md
Claude Code instructions for aranjan/icici-mcp, covering claude.md - project context for icici-mcp, project overview, architecture, broader system context and project structure.
jesse AGENTS.md
AGENTS.md instructions for jesse-ai/jesse, covering jesse repository guide for ai agents, skills, key characteristics, central framework and technology stack.
pumpfun-bonkfun-bot CLAUDE.md
Instructions for chainstacklabs/pumpfun-bonkfun-bot, covering agent guide, ground rules, layout, naming inside learning-examples/ and commands.
tradememory-protocol AGENTS.md
Instructions for mnemox-ai/tradememory-protocol, covering tradememory protocol — agent context, project rules, recent changes and current status.