Audit infrastructure for AI coding agents with evidence-backed review and policy gates.
接入配置
选择客户端，复制配置，再按项目文档补齐环境变量。
claude mcp add proofflow --env PROOFFLOW_API_KEY='<PROOFFLOW_API_KEY>' -- uvx proofflow-mcp~/.codex/config.toml
[mcp_servers.proofflow]
command = "uvx"
args = ["proofflow-mcp"]
env = { PROOFFLOW_API_KEY = "<PROOFFLOW_API_KEY>" }
# codex mcp add proofflow --env PROOFFLOW_API_KEY='<PROOFFLOW_API_KEY>' -- uvx proofflow-mcp~/.cursor/mcp.json
{
"mcpServers": {
"proofflow": {
"command": "uvx",
"args": [
"proofflow-mcp"
],
"env": {
"PROOFFLOW_API_KEY": "<PROOFFLOW_API_KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"proofflow": {
"type": "stdio",
"command": "uvx",
"args": [
"proofflow-mcp"
],
"env": {
"PROOFFLOW_API_KEY": "<PROOFFLOW_API_KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"proofflow": {
"command": "uvx",
"args": [
"proofflow-mcp"
],
"env": {
"PROOFFLOW_API_KEY": "<PROOFFLOW_API_KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"proofflow": {
"command": "uvx",
"args": [
"proofflow-mcp"
],
"env": {
"PROOFFLOW_API_KEY": "<PROOFFLOW_API_KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"proofflow": {
"command": "uvx",
"args": [
"proofflow-mcp"
],
"env": {
"PROOFFLOW_API_KEY": "<PROOFFLOW_API_KEY>"
}
}
}
}需要的环境变量
PROOFFLOW_API_KEY
把占位值换成你自己的密钥；不要把密钥提交到代码仓库。