Llm Cli Gateway
未认领MCP server providing unified access to Claude Code, Codex, Gemini, Grok and Mistral CLIs with session management, retry logic, and async job orchestration.
安装
npm install -g llm-cli-gateway工具（40 个）
addDir
(string[], optional): Additional workspace directories. A non-empty value is a high-risk managed input.
agent
(string, optional): Named sub-agent to run as. A non-empty value is a high-risk managed input because it can change tool and permission posture.
agents
(string, optional): Inline agent definitions JSON. A non-empty value is a high-risk managed input for the same reason.
allowedTools
(string[], optional): Restrict Claude tools to this allow-list. A non-empty allow-list is a high-risk managed input because it can change the tool posture.
approvalStrategy
(string, optional): "legacy" (default) or "mcpmanaged". Managed mode uses acceptEdits by default and forces strictMcpConfig:true, so Claude uses only the gateway-generated MCP configuration. A direct full-permission request requires all of an explicit caller request, an approval-manager approval, an
ask_model
ask one provider through the simplified surface.
bare
(boolean, optional): Start Claude in minimal mode, skipping local customization discovery. true is a high-risk managed input.
claude_request
Execute a Claude Code request with optional session management.
consensus_check
check whether providers agree with a claim.
continueSession
(boolean, optional): Continue the active session. It has the same managed-approval requirement as sessionId. Because Claude --continue selects by cwd, it requires workingDir or a registered workspace selected explicitly, through caller-owned session metadata, or by the configured default. That works
createNewSession
(boolean, optional): Always create a new session
dangerouslySkipPermissions
(boolean, optional): Deprecated, maps to permissionMode: "bypassPermissions"; permissionMode wins when both are set. It is a direct full-permission request under mcpmanaged.
debugFile
(string, optional): Write Claude debug output to a file. A non-empty path is a high-risk managed input.
disallowedTools
(string[], optional): Explicitly deny listed Claude tools
effort
(string, optional): Reasoning effort (lowmediumhighxhighmax)
excludeDynamicSystemPromptSections
(boolean, optional): Trim dynamic system prompt sections
fallbackModel
(string, optional): Auto-fallback model when the default is overloaded
forkSession
(boolean, optional): Fork the resumed session instead of appending to it. Under mcpmanaged, it is a high-risk native-fork input that requires approval and LLMGATEWAYAPPROVALALLOWBYPASS=1, but stays bounded.
job_status
and jobresult: poll and collect validation job outputs.
jsonSchema
(string, optional): JSON Schema literal constraining structured output
list_available_models
list the models each provider CLI exposes through the simplified surface.
maxBudgetUsd
(number, optional): Budget cap in USD for the request
maxTurns
(integer, optional): Agent-loop turn cap
model
(string, optional): Model name or alias (use listmodels for available values; supports latest)
noSessionPersistence
(boolean, optional): Ephemeral session (not persisted to disk)
outputFormat
(string, optional): Output format (textjsonstream-json), default: stream-json — the gateway parses NDJSON usage events for token/cost observability; override to text only when you want unparsed stdout
permissionMode
(string, optional): Claude permission mode (defaultacceptEditsplanautodontAskbypassPermissions); preferred over dangerouslySkipPermissions. bypassPermissions is a direct full-permission request under mcpmanaged.
pluginDir
/ pluginUrl (string[], optional): Load Claude plugins from local directories or URLs. Non-empty values are high-risk managed inputs.
prompt
(string, optional): The prompt to send (1-100,000 chars). Exactly one of prompt or promptParts is required (mutually exclusive)
red_team_review
challenge a plan, answer, or document for risks and failure modes.
review_changes
capture one complete Git review artifact, fence repository
safeMode
(boolean, optional): Start Claude with local customizations disabled, including CLAUDE.md, skills, plugins, hooks, MCP, commands, and agents. true is a high-risk managed input.
second_opinion
ask one provider to review an answer.
sessionId
(string, optional): Specific session ID to use. Under mcpmanaged, native continuation is a high-risk input because it can inherit an unverified provider posture; it requires approval and LLMGATEWAYAPPROVALALLOWBYPASS=1, but does not select a full-permission profile.
settingSources
/ settings / tools (optional): Setting sources to load, settings JSON path/literal, built-in tool restriction. Non-empty setting sources, settings, or tool selections are high-risk managed inputs.
synthesize_validation
run an explicit judge model after provider results have
systemPrompt
/ appendSystemPrompt (string, optional): Replace or extend the system prompt. A non-empty value is a high-risk managed input.
systemPromptFile
/ appendSystemPromptFile (string, optional): Replace or extend the system prompt from a file. A non-empty file path is a high-risk managed input.
validate_with_models
ask two or more providers to independently validate a question.
validation_receipt
retrieve the canonically hashed immutable receipt of a terminal cross-LLM validation run by validationId (returns minted pending verificationfailed expiredunminted notfound, own-or-not-found). verificationfailed means a stored receipt exists but disagrees with its durable run, which is a defect to i