scout scan
Scan transcripts and read results.
Pass a FILE which is either a Python script that contains @scanner or @scanjob decorated functions or a config file (YAML or JSON) that adheres to the ScanJobConfig schema.
Usage
scout scan [OPTIONS] COMMAND [ARGS]...Options
| Name | Type | Description | Default |
|---|---|---|---|
-S |
text | One or more scanjob or scanner arguments (e.g. -S arg=value) | |
-T, --transcripts |
text | One or more transcript sources (e.g. -T ./logs) | |
--results |
text | Location to write scan results to. | ./scans |
--worklist |
path | Transcript ids to process for each scanner (JSON or YAML file). | |
-V, --validation |
text | One or more validation sets to apply for scanners (e.g. -V myscanner:deception.csv) | |
--model |
text | Model used by default for llm scanners. | |
--model-base-url |
text | Base URL for for model API | |
-M |
text | One or more native model arguments (e.g. -M arg=value) | |
--model-config |
text | YAML or JSON config file with model arguments. | |
--model-role |
text | Named model role with model name or YAML/JSON config, e.g. –model-role critic=openai/gpt-4o or –model-role grader=“{model: mockllm/model, temperature: 0.5}” | |
--max-transcripts |
integer | Maximum number of transcripts to scan concurrently (defaults to 25) | |
--max-processes |
integer | Number of worker processes. Defaults to 4. | |
--limit |
integer | Limit number of transcripts to scan. | |
--shuffle |
text | Shuffle order of transcripts (pass a seed to make the order deterministic) | |
--tags |
text | Tags to associate with this scan job (comma separated) | |
--metadata |
text | Metadata to associate with this scan job (more than one –metadata argument can be specified). | |
--cache |
text | Policy for caching of model generations. Specify –cache to cache with 7 day expiration (7D). Specify an explicit duration (e.g. (e.g. 1h, 3d, 6M) to set the expiration explicitly (durations can be expressed as s, m, h, D, W, M, or Y). Alternatively, pass the file path to a YAML or JSON config file with a full CachePolicy configuration. |
|
--batch |
text | Batch requests together to reduce API calls when using a model that supports batching (by default, no batching). Specify –batch to batch with default configuration, specify a batch size e.g. --batch=1000 to configure batches of 1000 requests, or pass the file path to a YAML or JSON config file with batch configuration. |
|
--max-connections |
integer | Maximum number of concurrent connections to Model API (defaults to max_transcripts) | |
--max-retries |
integer | Maximum number of times to retry model API requests (defaults to unlimited) | |
--timeout |
integer | Model API request timeout in seconds (defaults to no timeout) | |
--max-tokens |
integer | The maximum number of tokens that can be generated in the completion (default is model specific) | |
--temperature |
float | What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. | |
--top-p |
float | An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. | |
--top-k |
integer | Randomly sample the next word from the top_k most likely next words. Anthropic, Google, HuggingFace, and vLLM only. | |
--reasoning-effort |
choice (minimal | low | medium | high) |
Constrains effort on reasoning for reasoning models (defaults to medium). Open AI o-series and gpt-5 models only. |
|
--reasoning-tokens |
integer | Maximum number of tokens to use for reasoning. Anthropic Claude models only. | |
--reasoning-summary |
choice (concise | detailed | auto) |
Provide summary of reasoning steps (defaults to no summary). Use ‘auto’ to access the most detailed summarizer available for the current model. OpenAI reasoning models only. | |
--reasoning-history |
choice (none | all | last | auto) |
Include reasoning in chat message history sent to generate (defaults to “auto”, which uses the recommended default for each provider) | |
--response-schema |
text | JSON schema for desired response format (output should still be validated). OpenAI, Google, and Mistral only. | |
--dry-run |
boolean | Print resolved scanners and transcript counts without scanning. | |
--display |
choice (rich | plain | log | none) |
Set the display type (defaults to ‘rich’) | rich |
--log-level |
choice (debug | trace | http | info | warning | error | critical | notset) |
Set the log level (defaults to ‘warning’) | warning |
--debug |
boolean | Wait to attach debugger | |
--debug-port |
integer | Port number for debugger | 5678 |
--fail-on-error |
boolean | Re-raise exceptions instead of capturing them in results | |
--help |
boolean | Show this message and exit. |
Subcommands
| complete | Complete a scan which is incomplete due to errors (errors are not retried). |
| list | List the scans within the scans dir. |
| resume | Resume a scan which is incomplete due to interruption or errors (errors are retried). |
scout scan complete
Complete a scan which is incomplete due to errors (errors are not retried).
Usage
scout scan complete [OPTIONS] SCAN_LOCATIONOptions
| Name | Type | Description | Default |
|---|---|---|---|
--display |
choice (rich | plain | log | none) |
Set the display type (defaults to ‘rich’) | rich |
--log-level |
choice (debug | trace | http | info | warning | error | critical | notset) |
Set the log level (defaults to ‘warning’) | warning |
--debug |
boolean | Wait to attach debugger | |
--debug-port |
integer | Port number for debugger | 5678 |
--fail-on-error |
boolean | Re-raise exceptions instead of capturing them in results | |
--help |
boolean | Show this message and exit. |
Subcommands
scout scan list
List the scans within the scans dir.
Usage
scout scan list [OPTIONS] [SCANS_DIR]Options
| Name | Type | Description | Default |
|---|---|---|---|
--display |
choice (rich | plain | log | none) |
Set the display type (defaults to ‘rich’) | rich |
--log-level |
choice (debug | trace | http | info | warning | error | critical | notset) |
Set the log level (defaults to ‘warning’) | warning |
--debug |
boolean | Wait to attach debugger | |
--debug-port |
integer | Port number for debugger | 5678 |
--fail-on-error |
boolean | Re-raise exceptions instead of capturing them in results | |
--help |
boolean | Show this message and exit. |
Subcommands
scout scan resume
Resume a scan which is incomplete due to interruption or errors (errors are retried).
Usage
scout scan resume [OPTIONS] SCAN_LOCATIONOptions
| Name | Type | Description | Default |
|---|---|---|---|
--display |
choice (rich | plain | log | none) |
Set the display type (defaults to ‘rich’) | rich |
--log-level |
choice (debug | trace | http | info | warning | error | critical | notset) |
Set the log level (defaults to ‘warning’) | warning |
--debug |
boolean | Wait to attach debugger | |
--debug-port |
integer | Port number for debugger | 5678 |
--fail-on-error |
boolean | Re-raise exceptions instead of capturing them in results | |
--help |
boolean | Show this message and exit. |