|
| 1 | +--- |
| 2 | +title: "gptscript" |
| 3 | +--- |
| 4 | +## gptscript |
| 5 | + |
| 6 | + |
| 7 | + |
| 8 | +``` |
| 9 | +gptscript [flags] PROGRAM_FILE [INPUT...] |
| 10 | +``` |
| 11 | + |
| 12 | +### Options |
| 13 | + |
| 14 | +``` |
| 15 | + --cache-dir string Directory to store cache (default: $XDG_CACHE_HOME/gptscript) ($GPTSCRIPT_CACHE_DIR) |
| 16 | + --chat-state string The chat state to continue, or null to start a new chat and return the state ($GPTSCRIPT_CHAT_STATE) |
| 17 | + -C, --chdir string Change current working directory ($GPTSCRIPT_CHDIR) |
| 18 | + --color Use color in output (default true) ($GPTSCRIPT_COLOR) |
| 19 | + --config string Path to GPTScript config file ($GPTSCRIPT_CONFIG) |
| 20 | + --confirm Prompt before running potentially dangerous commands ($GPTSCRIPT_CONFIRM) |
| 21 | + --credential-context string Context name in which to store credentials ($GPTSCRIPT_CREDENTIAL_CONTEXT) (default "default") |
| 22 | + --credential-override string Credentials to override (ex: --credential-override github.com/example/cred-tool:API_TOKEN=1234) ($GPTSCRIPT_CREDENTIAL_OVERRIDE) |
| 23 | + --debug Enable debug logging ($GPTSCRIPT_DEBUG) |
| 24 | + --debug-messages Enable logging of chat completion calls ($GPTSCRIPT_DEBUG_MESSAGES) |
| 25 | + --default-model string Default LLM model to use ($GPTSCRIPT_DEFAULT_MODEL) (default "gpt-4o") |
| 26 | + --disable-cache Disable caching of LLM API responses ($GPTSCRIPT_DISABLE_CACHE) |
| 27 | + --disable-tui Don't use chat TUI but instead verbose output ($GPTSCRIPT_DISABLE_TUI) |
| 28 | + --dump-state string Dump the internal execution state to a file ($GPTSCRIPT_DUMP_STATE) |
| 29 | + --events-stream-to string Stream events to this location, could be a file descriptor/handle (e.g. fd://2), filename, or named pipe (e.g. \\.\pipe\my-pipe) ($GPTSCRIPT_EVENTS_STREAM_TO) |
| 30 | + --force-chat Force an interactive chat session if even the top level tool is not a chat tool ($GPTSCRIPT_FORCE_CHAT) |
| 31 | + --force-sequential Force parallel calls to run sequentially ($GPTSCRIPT_FORCE_SEQUENTIAL) |
| 32 | + -h, --help help for gptscript |
| 33 | + -f, --input string Read input from a file ("-" for stdin) ($GPTSCRIPT_INPUT) |
| 34 | + --list-models List the models available and exit ($GPTSCRIPT_LIST_MODELS) |
| 35 | + --list-tools List built-in tools and exit ($GPTSCRIPT_LIST_TOOLS) |
| 36 | + --listen-address string Server listen address ($GPTSCRIPT_LISTEN_ADDRESS) (default "127.0.0.1:9090") |
| 37 | + --no-trunc Do not truncate long log messages ($GPTSCRIPT_NO_TRUNC) |
| 38 | + --openai-api-key string OpenAI API KEY ($OPENAI_API_KEY) |
| 39 | + --openai-base-url string OpenAI base URL ($OPENAI_BASE_URL) |
| 40 | + --openai-org-id string OpenAI organization ID ($OPENAI_ORG_ID) |
| 41 | + -o, --output string Save output to a file, or - for stdout ($GPTSCRIPT_OUTPUT) |
| 42 | + -q, --quiet No output logging (set --quiet=false to force on even when there is no TTY) ($GPTSCRIPT_QUIET) |
| 43 | + --server Start server ($GPTSCRIPT_SERVER) |
| 44 | + --sub-tool string Use tool of this name, not the first tool in file ($GPTSCRIPT_SUB_TOOL) |
| 45 | + --ui Launch the UI ($GPTSCRIPT_UI) |
| 46 | + --workspace string Directory to use for the workspace, if specified it will not be deleted on exit ($GPTSCRIPT_WORKSPACE) |
| 47 | +``` |
| 48 | + |
| 49 | +### SEE ALSO |
| 50 | + |
| 51 | +* [gptscript credential](gptscript_credential.md) - List stored credentials |
| 52 | +* [gptscript eval](gptscript_eval.md) - |
| 53 | +* [gptscript fmt](gptscript_fmt.md) - |
| 54 | +* [gptscript parse](gptscript_parse.md) - |
| 55 | + |
0 commit comments