Synopsis

fcli ai-assist mcp start-stdio [--async-bg-threads=<bgThreads>] [--job-safe-return=<jobSafeReturnPeriod>] [-m=<module>] [--progress-interval=<progressIntervalPeriod>] [--progress-threads=<progressThreads>] [--work-threads=<workThreads>] [--import=<importFiles>[,<importFiles>…​]]…​ [[-h] [--env-prefix=<prefix>] [--log-file=<logFile>] [--log-level=<logLevel>] [--log-mask=<level>] [--debug]]

Description

Start the fcli MCP server over stdio. This command exposes fcli module commands and/or imported action functions as MCP tools to AI clients.

Note
 For better user experience and potentially lower token usage, consider Fortify skills from https://github.com/fortify/skills, which can be installed through your AI assistant marketplace (if available) or by utilizing the `fcli ai-assist extensions' commands.

THREAD MODEL: - Work threads (--work-threads): execute MCP tool calls. Each concurrent tool call occupies one thread for its full duration. Size to the maximum number of tool calls the AI may invoke in parallel. - Progress threads (--progress-threads): poll progress for long-running jobs at regular intervals. One thread is consumed per active long-running job during each poll. The default of 4 is sufficient for most use cases. - Async background threads (--async-bg-threads): run background async streaming jobs (e.g. run.fcli steps with streaming output). Increase if actions make heavy use of async streaming.

Options

--async-bg-threads=<bgThreads>

Number of background threads for running async streaming jobs. Default: 2.

--import=<importFiles>[,<importFiles>…​]

Action YAML files to import. Exported functions are registered as MCP tools or resources based on function metadata.

--job-safe-return=<jobSafeReturnPeriod>

Maximum time to wait synchronously for a job result before returning an in_progress placeholder. Specify duration like 25s, 2m, 1h.

-m, --module=<module>

Fcli module to expose through this MCP server instance.

--progress-interval=<progressIntervalPeriod>

Interval between internal progress counter updates for long-running jobs. Specify duration (e.g. 500ms, 1s, 2s).

--progress-threads=<progressThreads>

Number of threads used for updating and tracking job progress for long-running jobs.

--work-threads=<workThreads>

Number of worker threads used to execute MCP tool jobs concurrently. Increase for higher parallelism if AI invokes multiple tools simultaneously.

Generic fcli options (also see documentation link below)

--debug

Enable collection of debug logs.

--env-prefix=<prefix>

Prefix for resolving default option values. Default value: FCLI_DEFAULT.

-h, --help

Use 'fcli [command] -h' to display help for fcli (sub-)commands.

--log-file=<logFile>

Write log output to file. Default: ./fcli.log if logging is enabled.

--log-level=<logLevel>

Set logging level: TRACE, DEBUG, INFO, WARN, ERROR, NONE.

--log-mask=<level>

Log mask level: high, medium, low, none. Default: medium. Masking is done on a best-effort basis; no guarantee that all sensitive data will be masked.