Borrowing it
Nothing to install: this file belongs to CastleZJC/Platform-MCP. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/CastleZJC/Platform-MCP/main/CLAUDE.mdgit clone --depth 1 https://github.com/CastleZJC/Platform-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/castlezjc/platform-mcp/claude-md)<a href="https://agentmods.dev/instructions/castlezjc/platform-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/castlezjc/platform-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/castlezjc/platform-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/castlezjc/platform-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.06943 | $0.06943 |
| Opus 5.5 | $0.02777 | $0.02777 |
| Sonnet 5.5 | $0.01389 | $0.01389 |
| Haiku 4.5 | $0.00694 | $0.00694 |
Grade C, and why
Platform-MCP CLAUDE.md scanned grade C 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 2mo 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **Shell 风控 4 级**:CRITICAL(rm -rf /、mkfs、dd、fork bomb、shutdown...)/ HIGH(sudo、systemctl stop...)/ MEDIUM(curl、nohup、| sh...)/ LOW(ls、cat、grep...);PROD 自动升 CRITICAL Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Shell 风控 4 级**:CRITICAL(rm -rf /、mkfs、dd、fork bomb、shutdown...)/ HIGH(sudo、systemctl stop...)/ MEDIUM(curl、nohup、| sh...)/ LOW(ls、cat、grep...);PROD 自动升 CRITICAL How it starts
The opening of the file, as written. The whole thing — 335 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
Platform-MCP is an internal MCP (Model Context Protocol) capability platform. Phase 1 focuses on database Skill — executing SQL via MCP tools called from Claude Code, with a Web management portal for datasource config, user/auth, encryption, and audit logging.
Project state: Phase 1-5 complete + Server Skill 二期专项(Linux SSH/SFTP)已落地。Application code, tests (525 backend pytest + 110 frontend vitest), and management portal all implemented. API Key authentication and dual-transport (stdio + streamable-http) MCP Server are live. 11 MCP tools across 2 skill packages (database 5 + server 6). POC verification tests remain in poc/.
Architecture
Dual-entry monolithic Python app with shared business logic:
Claude Code ──stdio(env PLATFORM_MCP_API_KEY)──▶ MCP Server (mcp_server/__init__.py) ──┐
╲ ├──▶ skills/ datasource/ audit/ common/ (shared logic)
╲ │
Claude Code ──HTTP(header PLATFORM_MCP_API_KEY)──▶ MCP Server (streamable-http) ──┤
│ │
Browser ──HTTP(session cookie)──▶ FastAPI Web (main.py) ──┐ ├──▶ PostgreSQL 16.4 (system DB, ORM)
│ ├──▶ Oracle 11g (target DB, thick mode, raw driver)
│ └──▶ MySQL 5.6 (target DB, aiomysql)
↕ API Key 校验 (validate_api_key) │
pmcp_api_key (key_hash + key_encrypted) ◀──────────┘
- MCP Server (
mcp_server/__init__.py): dual transport. stdio mode (env var) OR streamable-http mode (ASGI middleware readsPLATFORM_MCP_API_KEYheader).settings.mcp.transportswitches modes. - FastAPI Web (
main.py): REST API for management portal (auth, datasource CRUD, API Key management, encryption, audit queries). Runs via Gunicorn + Uvicorn workers. - Shared logic: Both entries use the same skill registry (
mcp_server/skill/registry.py), SQL executor (skills/database/executor.py), risk engine (skills/database/risk.py), datasource manager (datasource/manager.py), and crypto module (common/crypto.py). - Identity propagation: API Key 校验结果存入
contextvars.ContextVar(mcp_server/__init__.py:_mcp_identity_var)。McpContext.operator从中读取,未校验时回退mcp://{settings.mcp.operator_role}。HTTP 模式每请求独立,stdio 模式进程级绑定. - Key constraint: Database Skill logic must not leak into
mcp_server— it only handles protocol, param normalization, context wrapping, and response formatting.
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.
- 2mo ago First seen · 335 lines · 6,943 tokens per session scan C 1d256b12f89e
Platform-MCP CLAUDE.md is an instructions file published in the GitHub repository CastleZJC/Platform-MCP (0 stars, last pushed 1mo ago), licensed MIT. It adds 6,943 tokens to every session, about $0.0278 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Platform-MCP CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| Platform-MCP CLAUDE.mdCastleZJC/Platform-MCP | Instructions file | 0 | 1mo ago | MIT | C | 6,943 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 196k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 144k | today | MIT | A | 8,130 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 196k | today | MIT | A | 5,685 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 150k | today | MIT | A | 4,680 | Codex, OpenCode |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 283k | today | MIT | A | 4,342 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 196k | today | MIT | A | 3,153 | GitHub Copilot |