> ## Documentation Index > Fetch the complete documentation index at: https://docs.usetuner.ai/llms.txt > Use this file to discover all available pages before exploring further. # Get Call > Retrieve a single call with its full transcript and analysis. ## OpenAPI ````yaml https://api.usetuner.ai/public/openapi.json get /api/v1/workspaces/{workspace_id}/agents/{agent_id}/calls/{call_id} openapi: 3.1.0 info: title: My Public API version: 1.0.0 servers: - url: https://api.usetuner.ai security: [] tags: - name: calls x-group: Calls - name: workspaces x-group: Workspaces - name: simulation x-group: Simulation - name: agent-settings x-group: Agent Settings - name: agents x-group: Agents - name: scenarios x-group: Scenarios - name: traces x-group: Traces - name: providers x-group: Providers paths: /api/v1/workspaces/{workspace_id}/agents/{agent_id}/calls/{call_id}: get: tags: - calls - calls summary: Get Call description: Retrieve a single call with its full transcript and analysis. operationId: >- get_call_details_api_v1_workspaces__workspace_id__agents__agent_id__calls__call_id__get parameters: - name: call_id in: path required: true schema: type: integer description: Tuner's internal numeric ID for the call. title: Call Id description: Tuner's internal numeric ID for the call. - name: agent_id in: path required: true schema: type: integer description: Tuner's internal numeric ID for the agent. title: Agent Id description: Tuner's internal numeric ID for the agent. - name: workspace_id in: path required: true schema: type: integer description: Workspace ID. Find this in Workspace > General Settings. title: Workspace Id description: Workspace ID. Find this in Workspace > General Settings. responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/CallDetailsResponse' '401': description: Missing or invalid API key. content: application/json: example: detail: Not authenticated schema: $ref: '#/components/schemas/ErrorResponse' '403': description: You do not have access to this workspace. content: application/json: example: detail: User does not have access to workspace 42 schema: $ref: '#/components/schemas/ErrorResponse' '404': description: Agent or call not found. content: application/json: example: detail: Call with ID 88 not found schema: $ref: '#/components/schemas/ErrorResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - Bearer: [] components: schemas: CallDetailsResponse: properties: id: type: integer title: Id description: Call database ID call_intent: anyOf: - type: string - type: 'null' title: Call Intent description: Call intent call_outcome: anyOf: - $ref: '#/components/schemas/CallOutcome' - type: 'null' description: Call outcome with value and type date: type: string title: Date description: Call start date in ISO 8601 format duration_ms: anyOf: - type: integer - type: 'null' title: Duration Ms description: Call duration in milliseconds call_status: anyOf: - type: string - type: 'null' title: Call Status description: Call status disconnection_reason: anyOf: - type: string - type: 'null' title: Disconnection Reason description: Disconnection reason cost: type: number title: Cost description: >- Call cost in cents (canonical unit). Frontend converts to dollars for display. default: 0 latency: anyOf: - additionalProperties: true type: object - type: 'null' title: Latency description: >- Raw persisted latency payload from DB with top-level keys (for example: e2e/llm/stt/tts/eou). Any metadata key is excluded. voice_metrics: anyOf: - $ref: '#/components/schemas/VoiceMetricsSummary' - type: 'null' description: Aggregated voice metrics derived from call latency + transcript customer_number: type: string title: Customer Number description: Customer phone number default: '' caller_phone_number: anyOf: - type: string - type: 'null' title: Caller Phone Number description: Caller phone number (for phone_call type) call_id: type: string title: Call Id description: Provider call ID recording_url: anyOf: - type: string - type: 'null' title: Recording Url description: URL to call recording extracted_fields: additionalProperties: type: string type: object title: Extracted Fields description: Extracted data fields from LLM evaluations: additionalProperties: true type: object title: Evaluations description: Call evaluations (placeholder) transcript_with_tool_calls: anyOf: - items: additionalProperties: true type: object type: array - type: 'null' title: Transcript With Tool Calls description: >- Raw transcript segments as stored, including agent/user messages, node transitions, and tool calls. end_goal: anyOf: - type: string - type: 'null' title: End Goal description: 'Deprecated: use `scenario.goal`. Simulation scenario end goal.' deprecated: true persona: anyOf: - type: string - type: 'null' title: Persona description: 'Deprecated: use `scenario.label`. Simulation scenario persona.' deprecated: true prompt: anyOf: - type: string - type: 'null' title: Prompt description: 'Deprecated: use `scenario.situation`. Simulation scenario prompt.' deprecated: true scenario_type: anyOf: - type: string - type: 'null' title: Scenario Type description: >- Deprecated: use `scenario.type` / `simulation_kind`. 'Routine call' for functional, 'Pressure call' for adversarial, 'recorded_clips' for a replay call. deprecated: true simulation_kind: anyOf: - type: string enum: - scenario - replay - type: 'null' title: Simulation Kind description: >- 'scenario' for a generated/library simulation call (see `scenario`), 'replay' for a deterministic recorded-clips replay call (see `simulation_replay`). Null for a non-simulation call. scenario: anyOf: - $ref: '#/components/schemas/SimulationScenarioDetails' - type: 'null' description: >- Scenario content and context. Null for non-simulation calls and for replay calls (see simulation_replay instead). simulation_replay: anyOf: - $ref: '#/components/schemas/SimulationReplayInfo' - type: 'null' description: >- Deterministic-replay context -- non-null only for calls produced by a recorded-clips replay run. Null for generated simulation calls and non-simulation calls. pii_redaction_status: anyOf: - type: string enum: - redacted - partial_failure - failed - type: 'null' title: Pii Redaction Status description: >- PII redaction outcome. Null when redaction was not attempted (disabled workspace, simulation call, or pre-feature call); otherwise one of 'redacted', 'partial_failure', or 'failed'. retention_erased_scope: anyOf: - type: string enum: - transcript_recording - caller_data - type: 'null' title: Retention Erased Scope description: >- Retention erasure scope applied to this call. Null when the call was never erased. The 'full' scope never appears here -- it deletes the call row entirely, so those calls 404 instead. retention_erased_at: anyOf: - type: string - type: 'null' title: Retention Erased At description: >- When retention erasure ran for this call, ISO 8601. Null when never erased. metadata: anyOf: - $ref: '#/components/schemas/CallMetadataResponse' - type: 'null' description: >- Call-level metadata: AI models, AI usage, and remaining customer-supplied keys. Null when the call has none. type: object required: - id - date - call_id title: CallDetailsResponse description: Detailed call information response. ErrorResponse: properties: detail: type: string title: Detail description: Human-readable description of what went wrong. type: object required: - detail title: ErrorResponse description: Error body returned when a request fails. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError CallOutcome: properties: value: type: string title: Value description: Call outcome value/label type: anyOf: - type: string - type: 'null' title: Type description: Call outcome type (e.g., 'success', 'failure') type: object required: - value title: CallOutcome description: Call outcome with value and type. VoiceMetricsSummary: properties: p50_latency: anyOf: - type: number - type: 'null' title: P50 Latency description: P50 e2e latency in milliseconds p75_latency: anyOf: - type: number - type: 'null' title: P75 Latency description: P75 e2e latency in milliseconds p90_latency: anyOf: - type: number - type: 'null' title: P90 Latency description: P90 e2e latency in milliseconds p95_latency: anyOf: - type: number - type: 'null' title: P95 Latency description: P95 e2e latency in milliseconds p99_latency: anyOf: - type: number - type: 'null' title: P99 Latency description: P99 e2e latency in milliseconds longest_agent_monologue_ms: anyOf: - type: number - type: 'null' title: Longest Agent Monologue Ms description: Longest continuous agent monologue duration in milliseconds type: object title: VoiceMetricsSummary description: Aggregated voice metrics surfaced on call details. SimulationScenarioDetails: properties: type: type: string enum: - routine - pressure title: Type description: functional->routine, adversarial->pressure label: anyOf: - type: string - type: 'null' title: Label description: 'scenario_label_from_snapshot: persona, else name' goal: anyOf: - type: string - type: 'null' title: Goal description: Snapshot goal; blank treated as null situation: anyOf: - type: string - type: 'null' title: Situation description: situation_from_generated_scenario; blank -> null first_line: anyOf: - type: string - type: 'null' title: First Line description: Snapshot first_line (library runs only); blank -> null stop_condition: anyOf: - type: string - type: 'null' title: Stop Condition description: Snapshot stop_condition (when the caller hangs up); blank -> null intent: anyOf: - type: string - type: 'null' title: Intent description: Snapshot intent label; blank -> null tags: items: type: string type: array title: Tags description: Snapshot tags (library runs only); [] otherwise target_eval: anyOf: - $ref: '#/components/schemas/SimulationScenarioTargetEval' - type: 'null' description: From the execution's target_ai_evaluation column test_profile: anyOf: - $ref: '#/components/schemas/SimulationTestProfile' - type: 'null' description: The run's simulation profile, if one was used language: anyOf: - type: string - type: 'null' title: Language description: Display string for the run's caller persona language/accent noise: anyOf: - $ref: '#/components/schemas/SimulationNoise' - type: 'null' description: The run's noise environment origin: type: string enum: - manual - ai_generated title: Origin description: >- 'manual' iff the linked library scenario (not soft-deleted) has origin 'manual'; 'ai_generated' otherwise, including when the link is deleted or absent. library_scenario_id: anyOf: - type: integer - type: 'null' title: Library Scenario Id description: >- The linked library scenario's id, when it exists and is not soft-deleted type: object required: - type - origin title: SimulationScenarioDetails description: >- Scenario content and context for a simulation call (ENG-1543). All content fields (everything but `origin`/`library_scenario_id`) come from the execution's frozen snapshot -- never from the live library `scenario` row, so editing or deleting a library scenario never changes what a past call's details show. SimulationReplayInfo: properties: voice_clip_set_id: type: integer title: Voice Clip Set Id description: Clip set replayed into this call voice_clip_set_name: anyOf: - type: string - type: 'null' title: Voice Clip Set Name description: Clip set display name at the time of the run source_call_id: anyOf: - type: integer - type: 'null' title: Source Call Id description: Production call the clips were sliced from source_channel_mode: anyOf: - type: string enum: - multi_channel - mono - type: 'null' title: Source Channel Mode description: Channel layout of the source recording repetition: anyOf: - type: integer - type: 'null' title: Repetition description: 1-based repetition index of this call within its clip set clip_count: type: integer title: Clip Count description: Number of caller-turn clips played back replay_settings: additionalProperties: true type: object title: Replay Settings description: Replay tuning knobs frozen onto the run turn_timings: items: $ref: '#/components/schemas/SimulationReplayTurnTiming' type: array title: Turn Timings description: >- Per-turn timings, ordered by turn_index. Empty when the replay runtime's best-effort write of timings never landed -- never null and never an error. type: object required: - voice_clip_set_id - clip_count title: SimulationReplayInfo description: >- Deterministic-replay context for a call produced by a recorded-clips replay run. Everything except ``turn_timings`` comes from the scenario's ``generated_scenario`` snapshot, frozen at run time; ``turn_timings`` comes from the call's own metadata. CallMetadataResponse: properties: ai_models: anyOf: - $ref: '#/components/schemas/CallAiModelsResponse' - type: 'null' description: AI model identifiers used during the call; None when all are missing ai_usage: anyOf: - $ref: '#/components/schemas/CallAiUsageResponse' - type: 'null' description: AI usage/cost metrics for the call; None when all are missing extra: additionalProperties: true type: object title: Extra description: >- Remaining customer-supplied metadata keys from general_meta_data_raw, excluding the keys already surfaced elsewhere (ai_models, usage_token, simulation_correlation_id). type: object title: CallMetadataResponse description: >- Call-level metadata: AI models, AI usage, and remaining customer-supplied keys. ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError SimulationScenarioTargetEval: properties: type: type: string enum: - eval - outcome title: Type description: Target evaluation kind value: type: string title: Value description: Raw stored value (eval name, or the outcome text) label: type: string title: Label description: >- Display label. For 'eval', the agent's current eval display name resolved from `value`, falling back to `value` itself when no active eval matches (renamed or deleted). For 'outcome', always equal to `value`. type: object required: - type - value - label title: SimulationScenarioTargetEval description: Target AI evaluation for a simulation scenario, resolved for display. SimulationTestProfile: properties: name: anyOf: - type: string - type: 'null' title: Name description: Simulation profile name at run time fields: additionalProperties: true type: object title: Fields description: The profile's frozen key/value snapshot type: object title: SimulationTestProfile description: The customer-facts test profile used by a simulation run, if any. SimulationNoise: properties: name: type: string title: Name description: Noise environment display name intensity: anyOf: - type: number - type: 'null' title: Intensity description: Noise gain applied for the run type: object required: - name title: SimulationNoise description: Background noise environment used by a simulation run. SimulationReplayTurnTiming: properties: turn_index: type: integer title: Turn Index description: Zero-based position of the clip within the set clip_started_ms: type: integer title: Clip Started Ms description: Offset into the call at which the clip started playing, in ms clip_ended_ms: type: integer title: Clip Ended Ms description: Offset into the call at which the clip finished playing, in ms agent_first_audio_ms: anyOf: - type: integer - type: 'null' title: Agent First Audio Ms description: >- Offset at which the agent's first audio for this turn arrived. Null when the agent never spoke after this clip -- must not be coerced to 0. agent_turn_ended_ms: anyOf: - type: integer - type: 'null' title: Agent Turn Ended Ms description: >- Offset at which the agent's turn was detected as finished. Null if it never finished. waited_ms: type: integer title: Waited Ms description: >- How long the replayer waited for the agent before playing the next clip wait_capped: type: boolean title: Wait Capped description: True when the wait hit `max_wait_ms` instead of the agent finishing overlapped: type: boolean title: Overlapped description: True when the clip started while the agent was still speaking type: object required: - turn_index - clip_started_ms - clip_ended_ms - waited_ms - wait_capped - overlapped title: SimulationReplayTurnTiming description: >- One caller turn's playback/response timing from a deterministic replay call. A straight read of ``simulation_calls.call_metadata["replay_turn_timings"]``, written best-effort by the replay runtime -- see ``app.services.simulation_clip_replay``. CallAiModelsResponse: properties: stt: anyOf: - type: string - type: 'null' title: Stt description: Speech-to-text (ASR) model identifier used during the call llm: anyOf: - type: string - type: 'null' title: Llm description: LLM model identifier used during the call tts: anyOf: - type: string - type: 'null' title: Tts description: Text-to-speech model identifier used during the call extra: additionalProperties: true type: object title: Extra description: >- Any other key in the customer-supplied ai_models object we don't recognize by name. type: object title: CallAiModelsResponse description: AI model identifiers surfaced on the call detail 'AI Models' section. CallAiUsageResponse: properties: input_tokens: anyOf: - type: integer - type: 'null' title: Input Tokens description: LLM input (prompt) token count output_tokens: anyOf: - type: integer - type: 'null' title: Output Tokens description: LLM output (completion) token count llm_cost: anyOf: - type: number - type: 'null' title: Llm Cost description: >- LLM cost in cents (canonical unit, matching CallDetailsResponse.cost). voice_cost: anyOf: - type: number - type: 'null' title: Voice Cost description: >- Combined STT + TTS cost in cents (canonical unit). Covers both directions of speech; None only when neither component cost is available. stt_duration_seconds: anyOf: - type: number - type: 'null' title: Stt Duration Seconds description: >- Speech-to-text audio duration processed, in seconds. Sent by non-Vapi providers (custom API, LiveKit, Dograh, Pipecat, some Retell-labeled integrations) that do not populate llm_cost/voice_cost. llm_tokens: anyOf: - type: integer - type: 'null' title: Llm Tokens description: >- Total LLM token count for the call. Sent by non-Vapi providers as a single total rather than the input_tokens/output_tokens split Vapi provides. tts_characters: anyOf: - type: integer - type: 'null' title: Tts Characters description: >- Number of characters synthesized by text-to-speech during the call. Sent by non-Vapi providers in place of a per-direction voice cost breakdown. extra: additionalProperties: true type: object title: Extra description: >- Any other key in the customer-supplied usage_token object we don't recognize by name. type: object title: CallAiUsageResponse description: AI usage/cost metrics surfaced on the call detail 'AI Usage' section. securitySchemes: Bearer: type: http description: >- Use Tuner API key (tr_api_...) or user session token. Find your API key in Workspace Settings > API Keys. scheme: bearer ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.