JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.
This page documents the structure of the main artifacts used and produced by ArkSim: scenario files, simulation output, evaluation output, and the run configuration file. Use it when building integrations, writing custom tooling, or debugging pipeline outputs.
For agent connection details (Chat Completions, A2A, and config), see Agent configuration.
Scenarios (scenarios.json)
The scenario file is a single JSON document that lists scenario objects. Each scenario describes one simulated user and one conversation session. It is the input to Simulation and is referenced during Evaluation.
Schema
string
required
Schema version identifier. Use "v1".
list[Scenario]
required
List of scenario objects.
Show Scenario fields
string
required
Unique identifier for this scenario.
string
required
Unique identifier for the simulated user.
string
required
The user’s conversational goal, passed into the simulated user prompt. Write in second person (“You want to…”).
string
required
Context about the agent (e.g. company or product overview) supplied to the simulated user.
string
required
Natural-language persona description for the simulated user. Used directly in the simulation prompt. Write in second person (“You are…”).
list[KnowledgeItem]
Knowledge items injected into the simulated user’s context. Can be an empty list. Each item has:
content (string, required): Text used as the user’s background knowledge. (Additional keys such as metadata may be present in the file for traceability but are not part of the loaded model.)
dict[string, Any]
Provenance metadata (e.g. target_agent_capability, goal_raw, user_attributes). Not used by the simulator or evaluator.
Simulation output (simulation.json)
Written by arksim simulate or the simulation step of arksim simulate-evaluate. Path is set by output_file_path in your config.
Schema
string
required
Output schema version (e.g. "v1").
string
required
ArkSim package version that produced this file.
string
required
UUID that uniquely identifies this simulation run. Referenced by the evaluation output as simulation_id to link the two artifacts.
string
required
ISO-8601 UTC timestamp when the file was generated.
list[Conversation]
required
One record per simulated conversation.
Show Conversation fields
string
required
Unique ID for this conversation.
string
required
ID of the scenario used for this conversation.
list[Message]
required
Ordered list of messages in the conversation.Each message has: turn_id (int), message_id (string), role ("simulated_user" or "assistant"), content (string).
SimulatedUserPrompt
required
Rendered simulated user prompt and variables used to drive the conversation.
simulated_user_prompt_template: The Jinja template used for the simulated user system prompt.
variables: Dict used to render the template. Keys are scenario.agent_context, scenario.goal, scenario.knowledge, and scenario.user_profile.
Evaluation output (evaluation.json)
Written by arksim evaluate or the evaluation step of arksim simulate-evaluate. Path is {output_dir}/evaluation.json.
Schema
string
required
Output schema version (e.g. "v1").
string
required
ISO-8601 UTC timestamp when the file was generated.
string
required
ArkSim package version that produced this file.
string
required
UUID that uniquely identifies this evaluation run. Generated fresh for each arksim evaluate or arksim simulate-evaluate invocation.
string
required
UUID of the simulation run that produced the conversations being evaluated. Copied from simulation.json to link the two artifacts.
list[ConversationEvaluation]
required
One record per conversation evaluated.
Show ConversationEvaluation fields
string
required
ID of the conversation.
float
required
How fully the user goal was achieved (0–1).
string
required
Explanation for the goal completion score.
float
required
Proportion of turns with no detected behavior failure (0–1).
float
required
Combined score for the conversation (0–1). Computed from turn_success_ratio and goal_completion_score.
string
required
One of: completed, partial_failure, failed.
list[TurnEvaluation]
required
Per-turn scores and failure labels.Each turn has: turn_id, scores (list of metric results with name, value, optional reason, optional metadata), turn_score, turn_behavior_failure, turn_behavior_failure_reason, qual_scores, unique_error_ids.
list[UniqueError]
required
Deduplicated behavior failures across all conversations.
Show UniqueError fields
string
required
Unique ID (e.g. UUID) for this error pattern.
string
required
Category of the failure (e.g. lack of specific information, repetition, false information).
string
required
Human-readable description of the error.
string
Default "medium". Severity of the error.
list[Occurrence]
required
List of (conversation_id, turn_id) where this error occurred.
Run configuration (config.yaml)
The same YAML file can be used for arksim simulate, arksim evaluate, and arksim simulate-evaluate. For simulate-evaluate, simulation and evaluation settings are merged from a single config.
Agent configuration
Key
Type
Description
agent_config
object
Inline agent configuration. See Agent configuration for chat_completions and a2a.
Simulation keys
Used by simulate and the simulation phase of simulate-evaluate.
Key
Type
Description
scenario_file_path
string
Path to scenarios.json.
num_conversations_per_scenario
int
Number of conversations to run per scenario.
max_turns
int
Maximum turns per conversation.
output_file_path
string
Path where simulation.json is written.
simulated_user_prompt_template
string
Optional. Jinja template for the simulated user system prompt.
model
string
LLM model (e.g. gpt-5.1).
provider
string
LLM provider (e.g. openai).
num_workers
string | int
Parallel workers (default 50). Set "auto" to auto-scale.
Evaluation keys
Used by evaluate and the evaluation phase of simulate-evaluate.
Key
Type
Description
scenario_file_path
string
Path to scenarios.json (for goal and knowledge during evaluation).
simulation_file_path
string
Path to simulation output (simulation.json). Omitted when running simulate-evaluate (output passed in memory).
output_dir
string
Directory for evaluation results; evaluation.json and optional reports are written here.
custom_metrics_file_paths
list[string]
Paths to Python files defining custom metrics.
metrics_to_run
list[string]
Names of built-in metrics to run. If empty, all built-in metrics run.
generate_html_report
boolean
Whether to generate final_report.html.
score_threshold
float | null
If set, exit with non-zero code when any conversation score is below this value.
model
string
LLM model for scoring.
provider
string
LLM provider.
num_workers
string | int
Parallel workers (default 50). Set "auto" to auto-scale.
model, provider, and num_workers are shared across simulation and evaluation when both run from the same config.
{"docsConfig":{"theme":"mint","$schema":"https://mintlify.com/docs.json","name":"ArkSim Docs","colors":{"primary":"#035E7B","light":"#8EC1D6","dark":"#024E66"},"logo":{"light":"https://mintcdn.com/arklex-ca4e8217/fzSC6u7V-KKJVmxu/logo/light-mode.svg?fit=max\u0026auto=format\u0026n=fzSC6u7V-KKJVmxu\u0026q=85\u0026s=32308e1e89cdae8ed5a84f09533879e5","dark":"https://mintcdn.com/arklex-ca4e8217/fzSC6u7V-KKJVmxu/logo/dark-mode.svg?fit=max\u0026auto=format\u0026n=fzSC6u7V-KKJVmxu\u0026q=85\u0026s=64700bcd373389ed6845561803108f2e"},"favicon":"https://mintcdn.com/arklex-ca4e8217/fzSC6u7V-KKJVmxu/favicon.svg?fit=max\u0026auto=format\u0026n=fzSC6u7V-KKJVmxu\u0026q=85\u0026s=bb2dfed5809b91768fab08bebddb7b26","navbar":{"links":[{"href":"mailto:support@arklex.ai","label":"Support"},{"icon":"github","href":"https://github.com/arklexai/arksim","label":""}],"primary":{"type":"button","label":"Arklex Home","href":"https://www.arklex.ai/"}},"footer":{"socials":{"linkedin":"https://www.linkedin.com/company/arklex","discord":"https://discord.gg/YNkdqAQjzA","github":"https://github.com/arklexai/arksim"}},"seo":{"metatags":{"canonical":"https://docs.arklex.ai/","og:site_name":"ArkSim Docs","og:type":"website","og:description":"ArkSim is an open-source agent testing framework. Simulate realistic multi-turn conversations with your AI agent and evaluate performance before production.","twitter:card":"summary_large_image"}},"fonts":{"family":"DM Sans","weight":400},"icons":{"library":"fontawesome"},"contextual":{"options":["copy","view","chatgpt","claude","perplexity","mcp","cursor","vscode"],"display":"header"},"styling":{"latex":false},"navigation":{"pages":[]}},"deployment":{"deploymentMetadataValue":{"subdomain":"arklex-ca4e8217","hasAudioNarration":false,"actualSubdomain":"arklex-ca4e8217","customDomains":["docs.arklex.ai"],"gitSource":{"type":"github","owner":"arklexai","repo":"arksim","deployBranch":"main","contentDirectory":"docs","isPrivate":false,"mountPath":"","label":""},"entitlements":{},"buildId":"6af7741daa641b5a6d867093:failure","clientVersion":"0.0.4270","searchFilterCounts":[{"language":"default","tag":"Core Capabilities","count":13},{"language":"default","tag":"Examples","count":10},{"language":"default","tag":"Getting started","count":6},{"language":"default","tag":"Help","count":3},{"language":"default","tag":"CI/CD","count":2},{"language":"default","tag":"Integrations","count":1}]},"assistantConfig":{"enableAskAiButton":false,"enableStarterQuestions":false,"enableAskAiSearchBar":true},"basePath":"","hasChatPermissions":false,"hasWebMcpPermissions":false,"toggles":[{"name":"kill-isr","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"custom-rbac-roles","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"agent-mintlify-slack","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"agent-integrations-settings","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"rbac-expansion","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automations-pr-grouping","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"custom-prefix","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"analytics-v7","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"user-flows-analytics","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"no-publish-flow","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-onboarding","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"snapshot-serving-live","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"stripTemperatureMiddleware","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"support-platform-pylon-zendesk","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"snapshot-serving","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"instant-rollback","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"granular-notifications-server","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"granular-notifications","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automations-fable","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"fixed-ai-action-pricing","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"site-roles","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automation-resolve-merge-conflict","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"scoped-publishing","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"mintql","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"page-level-analytics","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"editor-version-history-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"billing-page-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"full-page-search","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workspace-projects","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"assistant-v5","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"fork-pr-previews","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"revalidation-jwt","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-sso-scim","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"hobby-abuse-scan","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"dashboard-inbox","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-signup","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"turbopuffer-dual-write","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-enabled","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"approve-review-flow","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"editor-metadata","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"sso-setup-deadline-banner","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"turbopuffer-serve","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"live-agents","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"starter-pricing","enabled":true,"variant":{"name":"treatment","enabled":true,"payload":{"type":"string","value":"treatment"},"feature_enabled":true},"impressionData":false},{"name":"llms-txt-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"wildcard-email-auth","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false}],"googleFontsInlined":true},"customCss":[{"filePath":"styles.css","content":"#sidebar-title{font-size:medium}"}]}
{"page":{"pagePath":"v0.2.0/schema-reference","slug":"/v0.2.0/schema-reference","pageMetadata":{"title":"Schema Reference","description":"JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.","href":"/v0.2.0/schema-reference"},"description":{"compiledSource":"\"use strict\";\nconst {jsx: _jsx} = arguments[0];\nconst {useMDXComponents: _provideComponents} = arguments[0];\nfunction _createMdxContent(props) {\n const _components = {\n p: \"p\",\n ..._provideComponents(),\n ...props.components\n };\n return _jsx(_components.p, {\n children: \"JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.\"\n });\n}\nfunction MDXContent(props = {}) {\n const {wrapper: MDXLayout} = {\n ..._provideComponents(),\n ...props.components\n };\n return MDXLayout ? _jsx(MDXLayout, {\n ...props,\n children: _jsx(_createMdxContent, {\n ...props\n })\n }) : _createMdxContent(props);\n}\nreturn {\n default: MDXContent\n};\n","frontmatter":{},"scope":{}},"pageType":"default","mdx":{"mdxSource":{"compiledSource":"\"use strict\";const{Fragment:_Fragment,jsx:_jsx,jsxs:_jsxs}=arguments[0];const{useMDXComponents:_provideComponents}=arguments[0];function _createMdxContent(props){const _components={a:`a`,code:`code`,hr:`hr`,li:`li`,p:`p`,tbody:`tbody`,td:`td`,th:`th`,thead:`thead`,tr:`tr`,ul:`ul`,..._provideComponents(),...props.components},{Expandable,Heading,Note,ResponseField,Table,Tip}=_components;if(!Expandable)_missingMdxReference(`Expandable`,true);if(!Heading)_missingMdxReference(`Heading`,true);if(!Note)_missingMdxReference(`Note`,true);if(!ResponseField)_missingMdxReference(`ResponseField`,true);if(!Table)_missingMdxReference(`Table`,true);if(!Tip)_missingMdxReference(`Tip`,true);return _jsxs(_Fragment,{children:[_jsx(_components.p,{children:`This page documents the structure of the main artifacts used and produced by ArkSim: scenario files, simulation output, evaluation output, and the run configuration file. Use it when building integrations, writing custom tooling, or debugging pipeline outputs.`}),`\n`,_jsx(Tip,{children:_jsxs(_components.p,{children:[`For agent connection details (Chat Completions, A2A, and config), see `,_jsx(_components.a,{href:`/v0.2.0/simulate-conversation#agent-configuration`,children:`Agent configuration`}),`.`]})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsxs(Heading,{level:`2`,id:`scenarios-scenarios-json`,children:[`Scenarios (`,_jsx(_components.code,{children:`scenarios.json`}),`)`]}),`\n`,_jsxs(_components.p,{children:[`The scenario file is a single JSON document that lists scenario objects. Each scenario describes one simulated user and one conversation session. It is the input to `,_jsx(_components.a,{href:`/v0.2.0/simulate-conversation`,children:`Simulation`}),` and is referenced during `,_jsx(_components.a,{href:`/v0.2.0/evaluate-conversation`,children:`Evaluation`}),`.`]}),`\n`,_jsx(Heading,{level:`3`,id:`schema`,children:`Schema`}),`\n`,_jsx(ResponseField,{name:`schema_version`,type:`string`,required:true,id:`param-schema-version`,children:_jsxs(_components.p,{children:[`Schema version identifier. Use `,_jsx(_components.code,{children:`\"v1\"`}),`.`]})}),`\n`,_jsxs(ResponseField,{name:`scenarios`,type:`list[Scenario]`,required:true,id:`param-scenarios`,children:[_jsx(_components.p,{children:`List of scenario objects.`}),_jsxs(Expandable,{title:`Scenario fields`,children:[_jsx(ResponseField,{name:`scenario_id`,type:`string`,required:true,id:`param-scenario-id`,children:_jsx(_components.p,{children:`Unique identifier for this scenario.`})}),_jsx(ResponseField,{name:`user_id`,type:`string`,required:true,id:`param-user-id`,children:_jsx(_components.p,{children:`Unique identifier for the simulated user.`})}),_jsx(ResponseField,{name:`goal`,type:`string`,required:true,id:`param-goal`,children:_jsx(_components.p,{children:`The user’s conversational goal, passed into the simulated user prompt. Write in second person (“You want to…”).`})}),_jsx(ResponseField,{name:`agent_context`,type:`string`,required:true,id:`param-agent-context`,children:_jsx(_components.p,{children:`Context about the agent (e.g. company or product overview) supplied to the simulated user.`})}),_jsx(ResponseField,{name:`user_profile`,type:`string`,required:true,id:`param-user-profile`,children:_jsx(_components.p,{children:`Natural-language persona description for the simulated user. Used directly in the simulation prompt. Write in second person (“You are…”).`})}),_jsxs(ResponseField,{name:`knowledge`,type:`list[KnowledgeItem]`,id:`param-knowledge`,children:[_jsx(_components.p,{children:`Knowledge items injected into the simulated user’s context. Can be an empty list. Each item has:`}),_jsxs(_components.ul,{children:[`\n`,_jsxs(_components.li,{children:[_jsx(_components.code,{children:`content`}),` (string, required): Text used as the user’s background knowledge. (Additional keys such as `,_jsx(_components.code,{children:`metadata`}),` may be present in the file for traceability but are not part of the loaded model.)`]}),`\n`]})]}),_jsx(ResponseField,{name:`origin`,type:`dict[string, Any]`,id:`param-origin`,children:_jsxs(_components.p,{children:[`Provenance metadata (e.g. `,_jsx(_components.code,{children:`target_agent_capability`}),`, `,_jsx(_components.code,{children:`goal_raw`}),`, `,_jsx(_components.code,{children:`user_attributes`}),`). Not used by the simulator or evaluator.`]})})]})]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsxs(Heading,{level:`2`,id:`simulation-output-simulation-json`,children:[`Simulation output (`,_jsx(_components.code,{children:`simulation.json`}),`)`]}),`\n`,_jsxs(_components.p,{children:[`Written by `,_jsx(_components.code,{children:`arksim simulate`}),` or the simulation step of `,_jsx(_components.code,{children:`arksim simulate-evaluate`}),`. Path is set by `,_jsx(_components.code,{children:`output_file_path`}),` in your config.`]}),`\n`,_jsx(Heading,{level:`3`,id:`schema-2`,children:`Schema`}),`\n`,_jsx(ResponseField,{name:`schema_version`,type:`string`,required:true,id:`param-schema-version-1`,children:_jsxs(_components.p,{children:[`Output schema version (e.g. `,_jsx(_components.code,{children:`\"v1\"`}),`).`]})}),`\n`,_jsx(ResponseField,{name:`simulator_version`,type:`string`,required:true,id:`param-simulator-version`,children:_jsx(_components.p,{children:`ArkSim package version that produced this file.`})}),`\n`,_jsx(ResponseField,{name:`simulation_id`,type:`string`,required:true,id:`param-simulation-id`,children:_jsxs(_components.p,{children:[`UUID that uniquely identifies this simulation run. Referenced by the evaluation output as `,_jsx(_components.code,{children:`simulation_id`}),` to link the two artifacts.`]})}),`\n`,_jsx(ResponseField,{name:`generated_at`,type:`string`,required:true,id:`param-generated-at`,children:_jsx(_components.p,{children:`ISO-8601 UTC timestamp when the file was generated.`})}),`\n`,_jsxs(ResponseField,{name:`conversations`,type:`list[Conversation]`,required:true,id:`param-conversations`,children:[_jsx(_components.p,{children:`One record per simulated conversation.`}),_jsxs(Expandable,{title:`Conversation fields`,children:[_jsx(ResponseField,{name:`conversation_id`,type:`string`,required:true,id:`param-conversation-id`,children:_jsx(_components.p,{children:`Unique ID for this conversation.`})}),_jsx(ResponseField,{name:`scenario_id`,type:`string`,required:true,id:`param-scenario-id-1`,children:_jsx(_components.p,{children:`ID of the scenario used for this conversation.`})}),_jsxs(ResponseField,{name:`conversation_history`,type:`list[Message]`,required:true,id:`param-conversation-history`,children:[_jsx(_components.p,{children:`Ordered list of messages in the conversation.`}),_jsxs(_components.p,{children:[`Each message has: `,_jsx(_components.code,{children:`turn_id`}),` (int), `,_jsx(_components.code,{children:`message_id`}),` (string), `,_jsx(_components.code,{children:`role`}),` (`,_jsx(_components.code,{children:`\"simulated_user\"`}),` or `,_jsx(_components.code,{children:`\"assistant\"`}),`), `,_jsx(_components.code,{children:`content`}),` (string).`]})]}),_jsxs(ResponseField,{name:`simulated_user_prompt`,type:`SimulatedUserPrompt`,required:true,id:`param-simulated-user-prompt`,children:[_jsx(_components.p,{children:`Rendered simulated user prompt and variables used to drive the conversation.`}),_jsxs(_components.ul,{children:[`\n`,_jsxs(_components.li,{children:[_jsx(_components.code,{children:`simulated_user_prompt_template`}),`: The Jinja template used for the simulated user system prompt.`]}),`\n`,_jsxs(_components.li,{children:[_jsx(_components.code,{children:`variables`}),`: Dict used to render the template. Keys are `,_jsx(_components.code,{children:`scenario.agent_context`}),`, `,_jsx(_components.code,{children:`scenario.goal`}),`, `,_jsx(_components.code,{children:`scenario.knowledge`}),`, and `,_jsx(_components.code,{children:`scenario.user_profile`}),`.`]}),`\n`]})]})]})]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsxs(Heading,{level:`2`,id:`evaluation-output-evaluation-json`,children:[`Evaluation output (`,_jsx(_components.code,{children:`evaluation.json`}),`)`]}),`\n`,_jsxs(_components.p,{children:[`Written by `,_jsx(_components.code,{children:`arksim evaluate`}),` or the evaluation step of `,_jsx(_components.code,{children:`arksim simulate-evaluate`}),`. Path is `,_jsx(_components.code,{children:`{output_dir}/evaluation.json`}),`.`]}),`\n`,_jsx(Heading,{level:`3`,id:`schema-3`,children:`Schema`}),`\n`,_jsx(ResponseField,{name:`schema_version`,type:`string`,required:true,id:`param-schema-version-2`,children:_jsxs(_components.p,{children:[`Output schema version (e.g. `,_jsx(_components.code,{children:`\"v1\"`}),`).`]})}),`\n`,_jsx(ResponseField,{name:`generated_at`,type:`string`,required:true,id:`param-generated-at-1`,children:_jsx(_components.p,{children:`ISO-8601 UTC timestamp when the file was generated.`})}),`\n`,_jsx(ResponseField,{name:`evaluator_version`,type:`string`,required:true,id:`param-evaluator-version`,children:_jsx(_components.p,{children:`ArkSim package version that produced this file.`})}),`\n`,_jsx(ResponseField,{name:`evaluation_id`,type:`string`,required:true,id:`param-evaluation-id`,children:_jsxs(_components.p,{children:[`UUID that uniquely identifies this evaluation run. Generated fresh for each `,_jsx(_components.code,{children:`arksim evaluate`}),` or `,_jsx(_components.code,{children:`arksim simulate-evaluate`}),` invocation.`]})}),`\n`,_jsx(ResponseField,{name:`simulation_id`,type:`string`,required:true,id:`param-simulation-id-1`,children:_jsxs(_components.p,{children:[`UUID of the simulation run that produced the conversations being evaluated. Copied from `,_jsx(_components.code,{children:`simulation.json`}),` to link the two artifacts.`]})}),`\n`,_jsxs(ResponseField,{name:`conversations`,type:`list[ConversationEvaluation]`,required:true,id:`param-conversations-1`,children:[_jsx(_components.p,{children:`One record per conversation evaluated.`}),_jsxs(Expandable,{title:`ConversationEvaluation fields`,children:[_jsx(ResponseField,{name:`conversation_id`,type:`string`,required:true,id:`param-conversation-id-1`,children:_jsx(_components.p,{children:`ID of the conversation.`})}),_jsx(ResponseField,{name:`goal_completion_score`,type:`float`,required:true,id:`param-goal-completion-score`,children:_jsx(_components.p,{children:`How fully the user goal was achieved (0–1).`})}),_jsx(ResponseField,{name:`goal_completion_reason`,type:`string`,required:true,id:`param-goal-completion-reason`,children:_jsx(_components.p,{children:`Explanation for the goal completion score.`})}),_jsx(ResponseField,{name:`turn_success_ratio`,type:`float`,required:true,id:`param-turn-success-ratio`,children:_jsx(_components.p,{children:`Proportion of turns with no detected behavior failure (0–1).`})}),_jsx(ResponseField,{name:`overall_agent_score`,type:`float`,required:true,id:`param-overall-agent-score`,children:_jsx(_components.p,{children:`Combined score for the conversation (0–1). Computed from turn_success_ratio and goal_completion_score.`})}),_jsx(ResponseField,{name:`evaluation_status`,type:`string`,required:true,id:`param-evaluation-status`,children:_jsxs(_components.p,{children:[`One of: `,_jsx(_components.code,{children:`completed`}),`, `,_jsx(_components.code,{children:`partial_failure`}),`, `,_jsx(_components.code,{children:`failed`}),`.`]})}),_jsxs(ResponseField,{name:`turn_scores`,type:`list[TurnEvaluation]`,required:true,id:`param-turn-scores`,children:[_jsx(_components.p,{children:`Per-turn scores and failure labels.`}),_jsxs(_components.p,{children:[`Each turn has: `,_jsx(_components.code,{children:`turn_id`}),`, `,_jsx(_components.code,{children:`scores`}),` (list of metric results with `,_jsx(_components.code,{children:`name`}),`, `,_jsx(_components.code,{children:`value`}),`, optional `,_jsx(_components.code,{children:`reason`}),`, optional `,_jsx(_components.code,{children:`metadata`}),`), `,_jsx(_components.code,{children:`turn_score`}),`, `,_jsx(_components.code,{children:`turn_behavior_failure`}),`, `,_jsx(_components.code,{children:`turn_behavior_failure_reason`}),`, `,_jsx(_components.code,{children:`qual_scores`}),`, `,_jsx(_components.code,{children:`unique_error_ids`}),`.`]})]})]})]}),`\n`,_jsxs(ResponseField,{name:`unique_errors`,type:`list[UniqueError]`,required:true,id:`param-unique-errors`,children:[_jsx(_components.p,{children:`Deduplicated behavior failures across all conversations.`}),_jsxs(Expandable,{title:`UniqueError fields`,children:[_jsx(ResponseField,{name:`unique_error_id`,type:`string`,required:true,id:`param-unique-error-id`,children:_jsx(_components.p,{children:`Unique ID (e.g. UUID) for this error pattern.`})}),_jsx(ResponseField,{name:`behavior_failure_category`,type:`string`,required:true,id:`param-behavior-failure-category`,children:_jsx(_components.p,{children:`Category of the failure (e.g. lack of specific information, repetition, false information).`})}),_jsx(ResponseField,{name:`unique_error_description`,type:`string`,required:true,id:`param-unique-error-description`,children:_jsx(_components.p,{children:`Human-readable description of the error.`})}),_jsx(ResponseField,{name:`severity`,type:`string`,id:`param-severity`,children:_jsxs(_components.p,{children:[`Default `,_jsx(_components.code,{children:`\"medium\"`}),`. Severity of the error.`]})}),_jsx(ResponseField,{name:`occurrences`,type:`list[Occurrence]`,required:true,id:`param-occurrences`,children:_jsx(_components.p,{children:`List of (conversation_id, turn_id) where this error occurred.`})})]})]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsxs(Heading,{level:`2`,id:`run-configuration-config-yaml`,children:[`Run configuration (`,_jsx(_components.code,{children:`config.yaml`}),`)`]}),`\n`,_jsxs(_components.p,{children:[`The same YAML file can be used for `,_jsx(_components.code,{children:`arksim simulate`}),`, `,_jsx(_components.code,{children:`arksim evaluate`}),`, and `,_jsx(_components.code,{children:`arksim simulate-evaluate`}),`. For `,_jsx(_components.code,{children:`simulate-evaluate`}),`, simulation and evaluation settings are merged from a single config.`]}),`\n`,_jsx(Heading,{level:`3`,id:`agent-configuration`,children:`Agent configuration`}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Key`}),_jsx(_components.th,{children:`Type`}),_jsx(_components.th,{children:`Description`})]})}),_jsx(_components.tbody,{children:_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`agent_config`})}),_jsx(_components.td,{children:`object`}),_jsxs(_components.td,{children:[`Inline agent configuration. See `,_jsx(_components.a,{href:`/v0.2.0/simulate-conversation#agent-configuration`,children:`Agent configuration`}),` for `,_jsx(_components.code,{children:`chat_completions`}),` and `,_jsx(_components.code,{children:`a2a`}),`.`]})]})})]}),`\n`,_jsx(Heading,{level:`3`,id:`simulation-keys`,children:`Simulation keys`}),`\n`,_jsxs(_components.p,{children:[`Used by `,_jsx(_components.code,{children:`simulate`}),` and the simulation phase of `,_jsx(_components.code,{children:`simulate-evaluate`}),`.`]}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Key`}),_jsx(_components.th,{children:`Type`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`scenario_file_path`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`Path to `,_jsx(_components.code,{children:`scenarios.json`}),`.`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`num_conversations_per_scenario`})}),_jsx(_components.td,{children:`int`}),_jsx(_components.td,{children:`Number of conversations to run per scenario.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`max_turns`})}),_jsx(_components.td,{children:`int`}),_jsx(_components.td,{children:`Maximum turns per conversation.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`output_file_path`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`Path where `,_jsx(_components.code,{children:`simulation.json`}),` is written.`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`simulated_user_prompt_template`})}),_jsx(_components.td,{children:`string`}),_jsx(_components.td,{children:`Optional. Jinja template for the simulated user system prompt.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`model`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`LLM model (e.g. `,_jsx(_components.code,{children:`gpt-5.1`}),`).`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`provider`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`LLM provider (e.g. `,_jsx(_components.code,{children:`openai`}),`).`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`num_workers`})}),_jsx(_components.td,{children:`string | int`}),_jsxs(_components.td,{children:[`Parallel workers (default `,_jsx(_components.code,{children:`50`}),`). Set `,_jsx(_components.code,{children:`\"auto\"`}),` to auto-scale.`]})]})]})]}),`\n`,_jsx(Heading,{level:`3`,id:`evaluation-keys`,children:`Evaluation keys`}),`\n`,_jsxs(_components.p,{children:[`Used by `,_jsx(_components.code,{children:`evaluate`}),` and the evaluation phase of `,_jsx(_components.code,{children:`simulate-evaluate`}),`.`]}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Key`}),_jsx(_components.th,{children:`Type`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`scenario_file_path`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`Path to `,_jsx(_components.code,{children:`scenarios.json`}),` (for goal and knowledge during evaluation).`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`simulation_file_path`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`Path to simulation output (`,_jsx(_components.code,{children:`simulation.json`}),`). Omitted when running `,_jsx(_components.code,{children:`simulate-evaluate`}),` (output passed in memory).`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`output_dir`})}),_jsx(_components.td,{children:`string`}),_jsxs(_components.td,{children:[`Directory for evaluation results; `,_jsx(_components.code,{children:`evaluation.json`}),` and optional reports are written here.`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`custom_metrics_file_paths`})}),_jsx(_components.td,{children:`list[string]`}),_jsx(_components.td,{children:`Paths to Python files defining custom metrics.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`metrics_to_run`})}),_jsx(_components.td,{children:`list[string]`}),_jsx(_components.td,{children:`Names of built-in metrics to run. If empty, all built-in metrics run.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`generate_html_report`})}),_jsx(_components.td,{children:`boolean`}),_jsxs(_components.td,{children:[`Whether to generate `,_jsx(_components.code,{children:`final_report.html`}),`.`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`score_threshold`})}),_jsx(_components.td,{children:`float | null`}),_jsx(_components.td,{children:`If set, exit with non-zero code when any conversation score is below this value.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`model`})}),_jsx(_components.td,{children:`string`}),_jsx(_components.td,{children:`LLM model for scoring.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`provider`})}),_jsx(_components.td,{children:`string`}),_jsx(_components.td,{children:`LLM provider.`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.code,{children:`num_workers`})}),_jsx(_components.td,{children:`string | int`}),_jsxs(_components.td,{children:[`Parallel workers (default `,_jsx(_components.code,{children:`50`}),`). Set `,_jsx(_components.code,{children:`\"auto\"`}),` to auto-scale.`]})]})]})]}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[_jsx(_components.code,{children:`model`}),`, `,_jsx(_components.code,{children:`provider`}),`, and `,_jsx(_components.code,{children:`num_workers`}),` are shared across simulation and evaluation when both run from the same config.`]})})]})}function MDXContent(props={}){const{wrapper:MDXLayout}={..._provideComponents(),...props.components};return MDXLayout?_jsx(MDXLayout,{...props,children:_jsx(_createMdxContent,{...props})}):_createMdxContent(props)}return{default:MDXContent};function _missingMdxReference(id,component){throw new Error(`Expected `+(component?`component`:`object`)+\" `\"+id+\"` to be defined: you likely forgot to import, pass, or provide it.\")}","frontmatter":{},"scope":{"config":{},"pageMetadata":{"title":"Schema Reference","description":"JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.","href":"/v0.2.0/schema-reference"}}},"mdxSourceWithNoJs":null},"mdxExtracts":{"tableOfContents":[{"title":"Scenarios (scenarios.json)","slug":"scenarios-scenarios-json","depth":2,"children":[{"title":"Schema","slug":"schema","depth":3,"children":[]}]},{"title":"Simulation output (simulation.json)","slug":"simulation-output-simulation-json","depth":2,"children":[{"title":"Schema","slug":"schema-2","depth":3,"children":[]}]},{"title":"Evaluation output (evaluation.json)","slug":"evaluation-output-evaluation-json","depth":2,"children":[{"title":"Schema","slug":"schema-3","depth":3,"children":[]}]},{"title":"Run configuration (config.yaml)","slug":"run-configuration-config-yaml","depth":2,"children":[{"title":"Agent configuration","slug":"agent-configuration","depth":3,"children":[]},{"title":"Simulation keys","slug":"simulation-keys","depth":3,"children":[]},{"title":"Evaluation keys","slug":"evaluation-keys","depth":3,"children":[]}]}],"codeExamples":{}},"jsLinks":[],"subdomain":"arklex-ca4e8217","siteName":"ArkSim Docs","title":"Schema Reference - ArkSim Docs","navMaps":{"firstHrefInVersion":{"0.3.x":{"href":"/v0.3.x/schema-reference","matchLevel":1},"0.2.0":{"href":"/v0.2.0/schema-reference","matchLevel":0},"main":{"href":"/main/schema-reference","matchLevel":1}},"firstHrefInLanguage":{}},"lastModified":"2026-06-02T16:27:28.491Z","originalFileLocation":"v0.2.0/schema-reference.mdx"},"headHtml":"\u003cscript data-mint-head>(function(){var d=document.documentElement,p=null;if(!p){try{p=localStorage.getItem(\"isDarkMode\")}catch(e){}if(p==='true')p='dark';else if(p==='false')p='light';if(p!=='light'\u0026\u0026p!=='dark'\u0026\u0026p!=='system')p=\"system\"}var r=p==='system'?(matchMedia(\"(prefers-color-scheme: dark)\").matches?'dark':'light'):p;d.classList.remove(r==='dark'?'light':'dark');d.classList.add(r);d.setAttribute(\"data-theme-preference\",p)})();\u003c/script>\u003cmeta data-mint-head name=\"description\" content=\"JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.\">\u003cmeta data-mint-head name=\"application-name\" content=\"ArkSim Docs\">\u003cmeta data-mint-head name=\"generator\" content=\"Mintlify\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon/android-chrome-192x192.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"apple-touch-icon\" type=\"image/png\" sizes=\"180x180\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon/apple-touch-icon.png\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon/favicon-16x16.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon/favicon-32x32.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon/favicon.ico\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon-dark/android-chrome-192x192.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon-dark/favicon-16x16.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon-dark/favicon-32x32.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon-dark/favicon.ico\" media=\"(prefers-color-scheme: dark)\">\u003cmeta data-mint-head property=\"og:title\" content=\"Schema Reference - ArkSim Docs\">\u003cmeta data-mint-head property=\"og:description\" content=\"ArkSim is an open-source agent testing framework. Simulate realistic multi-turn conversations with your AI agent and evaluate performance before production.\">\u003cmeta data-mint-head property=\"og:site_name\" content=\"ArkSim Docs\">\u003cmeta data-mint-head property=\"og:url\" content=\"https://docs.arklex.ai//v0.2.0/schema-reference\">\u003cmeta data-mint-head property=\"og:type\" content=\"website\">\u003cmeta data-mint-head property=\"og:image\" content=\"https://arklex-ca4e8217.mintlify.site/_mintlify/api/og/v2?division=Core+Capabilities\u0026amp;title=Schema+Reference\u0026amp;description=JSON+and+YAML+schemas+used+by+ArkSim+for+scenarios%2C+simulation+output%2C+evaluation+output%2C+and+run+configuration.\u0026amp;theme=dd893450ac6a843c53399a87\u0026amp;signature=8af0c2cf8e9e04d57e5010a2c608dbfa24b3573bc410e75c3001c93c9a2f8062\">\u003cmeta data-mint-head property=\"og:image:width\" content=\"1200\">\u003cmeta data-mint-head property=\"og:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"twitter:card\" content=\"summary_large_image\">\u003cmeta data-mint-head name=\"twitter:title\" content=\"Schema Reference - ArkSim Docs\">\u003cmeta data-mint-head name=\"twitter:description\" content=\"ArkSim is an open-source agent testing framework. Simulate realistic multi-turn conversations with your AI agent and evaluate performance before production.\">\u003cmeta data-mint-head name=\"twitter:image\" content=\"https://arklex-ca4e8217.mintlify.site/_mintlify/api/og/v2?division=Core+Capabilities\u0026amp;title=Schema+Reference\u0026amp;description=JSON+and+YAML+schemas+used+by+ArkSim+for+scenarios%2C+simulation+output%2C+evaluation+output%2C+and+run+configuration.\u0026amp;theme=dd893450ac6a843c53399a87\u0026amp;signature=8af0c2cf8e9e04d57e5010a2c608dbfa24b3573bc410e75c3001c93c9a2f8062\">\u003cmeta data-mint-head name=\"twitter:image:width\" content=\"1200\">\u003cmeta data-mint-head name=\"twitter:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"msapplication-config\" content=\"/mintlify-assets/_mintlify/favicons/arklex-ca4e8217/i9XTpXMmpLhjemFV/_generated/favicon/browserconfig.xml\">\u003cmeta data-mint-head name=\"apple-mobile-web-app-title\" content=\"ArkSim Docs\">\u003cmeta data-mint-head name=\"msapplication-TileColor\" content=\"#035E7B\">\u003clink data-mint-head rel=\"canonical\" href=\"https://docs.arklex.ai//v0.2.0/schema-reference\">\u003clink data-mint-head rel=\"alternate\" type=\"application/xml\" href=\"/sitemap.xml\">\u003clink data-mint-head rel=\"alternate\" type=\"text/markdown\" href=\"/v0.2.0/schema-reference.md\">\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https://docs.arklex.ai/#organization\",\"name\":\"ArkSim Docs\",\"url\":\"https://docs.arklex.ai\",\"logo\":{\"@type\":\"ImageObject\",\"url\":\"https://mintcdn.com/arklex-ca4e8217/fzSC6u7V-KKJVmxu/logo/light-mode.svg?fit=max\u0026auto=format\u0026n=fzSC6u7V-KKJVmxu\u0026q=85\u0026s=32308e1e89cdae8ed5a84f09533879e5\"}},{\"@type\":\"WebSite\",\"@id\":\"https://docs.arklex.ai#website\",\"name\":\"ArkSim Docs\",\"url\":\"https://docs.arklex.ai\",\"publisher\":{\"@id\":\"https://docs.arklex.ai/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https://docs.arklex.ai//v0.2.0/schema-reference#webpage\",\"url\":\"https://docs.arklex.ai//v0.2.0/schema-reference\",\"name\":\"Schema Reference\",\"description\":\"JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.\",\"dateModified\":\"2026-06-02T16:27:28.491Z\",\"isPartOf\":{\"@id\":\"https://docs.arklex.ai#website\"},\"breadcrumb\":{\"@id\":\"https://docs.arklex.ai//v0.2.0/schema-reference#breadcrumb\"}},{\"@type\":\"BreadcrumbList\",\"@id\":\"https://docs.arklex.ai//v0.2.0/schema-reference#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Core Capabilities\",\"item\":\"https://docs.arklex.ai/v0.2.0/build-scenario\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Schema Reference\",\"item\":\"https://docs.arklex.ai//v0.2.0/schema-reference\"}]},{\"@type\":[\"Article\",\"TechArticle\"],\"@id\":\"https://docs.arklex.ai//v0.2.0/schema-reference#article\",\"headline\":\"Schema Reference\",\"name\":\"Schema Reference\",\"description\":\"JSON and YAML schemas used by ArkSim for scenarios, simulation output, evaluation output, and run configuration.\",\"url\":\"https://docs.arklex.ai//v0.2.0/schema-reference\",\"mainEntityOfPage\":{\"@id\":\"https://docs.arklex.ai//v0.2.0/schema-reference#webpage\"},\"dateModified\":\"2026-06-02T16:27:28.491Z\",\"publisher\":{\"@id\":\"https://docs.arklex.ai/#organization\"},\"isPartOf\":{\"@id\":\"https://docs.arklex.ai#website\"}}]}\u003c/script>\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@type\":\"WebSite\",\"name\":\"ArkSim Docs\",\"publisher\":{\"@type\":\"Organization\",\"name\":\"ArkSim Docs\"},\"provider\":{\"@type\":\"Organization\",\"name\":\"Mintlify\",\"url\":\"https://mintlify.com\"}}\u003c/script>","agentPointerHref":"/v0.2.0/schema-reference.md","siteName":"ArkSim Docs","nav":{"global":null,"versions":[{"version":"0.3.x","default":true,"tag":"Latest","pages":[{"title":"0.3.x","href":"/v0.3.x/schema-reference"}]},{"version":"0.2.0","pages":[{"title":"Overview","href":"/v0.2.0/overview"},{"group":"Getting started","expanded":true,"pages":[{"title":"Installation","href":"/v0.2.0/installation"},{"title":"Quickstart","href":"/v0.2.0/quickstart"}]},{"group":"Core Capabilities","pages":[{"title":"Scenarios","href":"/v0.2.0/build-scenario"},{"title":"Simulation","href":"/v0.2.0/simulate-conversation"},{"title":"Evaluation","description":"Score your agent's responses across every conversation turn and surface failure patterns with actionable fixes.","href":"/v0.2.0/evaluate-conversation"},{"title":"Schema Reference","href":"/v0.2.0/schema-reference"}]},{"group":"Examples","pages":[{"title":"Insurance Customer Service Agent","description":"A walkthrough of running ArkSim against a customer service agent for an insurance company use case.","href":"/v0.2.0/insurance-customer-service-agent-evaluation"},{"title":"E-Commerce Customer Service Agent","href":"/v0.2.0/e-commerce-customer-service-agent-evaluation"},{"title":"Personal AI Assistant (OpenClaw)","href":"/v0.2.0/personal-ai-assistant-openclaw-evaluation"}]},{"group":"Help","pages":[{"title":"Contact","href":"/v0.2.0/contact-email"}]}]},{"version":"main","pages":[{"title":"main","href":"/main/schema-reference"}]}]},"basePath":"","title":"Schema Reference - ArkSim Docs","deploymentId":"dpl_DyXGnKeCjZyA3sUdfMJ7FjkdBefg","buildId":"dpl_DyXGnKeCjZyA3sUdfMJ7FjkdBefg","dynamics":[],"chunks":[]}