> ## Documentation Index > Fetch the complete documentation index at: https://docs.tracks.alpic.ai/llms.txt > Use this file to discover all available pages before exploring further. # List a run's events > Returns what the agent did, in order: each prompt, then its messages, reasoning, tool calls and subagents, then a `stop` once the agent stopped. Events read alike whatever the harness, and are read from the run's output: `frame_seqs` points at the lines each one comes from. To get the next page, pass the `seq` of the last event received as `after`. ## OpenAPI ````yaml https://tracks.alpic.ai/api/openapi.json get /runs/{id}/events openapi: 3.1.0 info: title: Tracks API version: 0.0.1 description: >- Run coding agents such as Claude Code in isolated sandboxes with an API key of your organization. Create a run with a prompt, then follow its output and what it costs, or launch an experiment to compare agents on your scenarios. servers: - url: https://tracks.alpic.ai/api security: [] tags: - name: Runs description: Start coding agents and follow their work. - name: Files description: Upload files to place in a run's sandbox. - name: Connections description: >- Let agents use MCP servers as one of their users, without their credentials entering the sandbox. - name: Experiments description: >- Define scenarios to give several agents in several variants, judged on what they did. - name: Executions description: Launch experiments and read their results. - name: Library description: What your experiments can give their agents. - name: System paths: /runs/{id}/events: get: tags: - Runs summary: List a run's events description: >- Returns what the agent did, in order: each prompt, then its messages, reasoning, tool calls and subagents, then a `stop` once the agent stopped. Events read alike whatever the harness, and are read from the run's output: `frame_seqs` points at the lines each one comes from. To get the next page, pass the `seq` of the last event received as `after`. operationId: runs.listEvents parameters: - name: id in: path schema: $ref: '#/components/schemas/RunId' required: true - name: after in: query schema: type: string pattern: ^[+-]?(?:\d+(?:\.\d+)?|\.\d+)(?:[Ee][+-]?\d+)?$ allOf: - description: Return only items after this position. required: false - name: limit in: query schema: type: string pattern: ^[+-]?(?:\d+(?:\.\d+)?|\.\d+)(?:[Ee][+-]?\d+)?$ allOf: - description: >- Maximum number of items to return, from 1 to 1000. Defaults to 100. required: false - name: prompt in: query schema: type: string pattern: ^[+-]?(?:\d+(?:\.\d+)?|\.\d+)(?:[Ee][+-]?\d+)?$ allOf: - description: Return only this prompt's events. required: false responses: '200': description: Success content: application/json: schema: type: object properties: data: type: array items: anyOf: - $ref: '#/components/schemas/PromptEvent' - $ref: '#/components/schemas/MessageEvent' - $ref: '#/components/schemas/ReasoningEvent' - $ref: '#/components/schemas/ToolStartedEvent' - $ref: '#/components/schemas/ToolEndedEvent' - $ref: '#/components/schemas/SubagentStartedEvent' - $ref: '#/components/schemas/SubagentEndedEvent' - $ref: '#/components/schemas/PageVisitedEvent' - $ref: '#/components/schemas/StopEvent' description: >- One thing the agent did, read from its output so that runs of every harness read alike. has_more: type: boolean description: Whether more items are available after this page. required: - data - has_more additionalProperties: false '400': description: >- The request is invalid: a value is missing, has the wrong type or is out of range. Fix it before sending the request again. content: application/json: schema: $ref: '#/components/schemas/InvalidRequestEncoded' '401': description: >- The API key is missing, malformed or revoked, or its organization is closed. content: application/json: schema: $ref: '#/components/schemas/ApiKeyRefusedEncoded' '404': description: No run with this id exists in your organization. content: application/json: schema: $ref: '#/components/schemas/RunNotFoundEncoded' security: - bearer: [] components: schemas: RunId: type: string description: Unique identifier of the run. PromptEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - prompt text: type: string description: What you asked the agent to do. required: - run_id - seq - prompt - captured_at - frame_seqs - type - text additionalProperties: false description: A prompt given to the agent; the first event of its prompt. MessageEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - message text: type: string description: What the agent wrote. required: - run_id - seq - prompt - captured_at - frame_seqs - type - text additionalProperties: false description: Text the agent wrote, apart from reasoning and tool calls. ReasoningEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - reasoning text: type: string description: The reasoning, or a summary of it. Absent when the harness hides it. required: - run_id - seq - prompt - captured_at - frame_seqs - type additionalProperties: false description: The agent reasoning before it acts. ToolStartedEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - tool_started tool_call_id: type: string description: Identifier of the tool call, unique in its run. tool_name: type: string description: Name of the tool, as the harness calls it. input: description: What the agent passed to the tool. required: - run_id - seq - prompt - captured_at - frame_seqs - type - tool_call_id - tool_name - input additionalProperties: false description: The agent called a tool. ToolEndedEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - tool_ended tool_call_id: type: string description: Identifier of the tool call, unique in its run. outcome: $ref: '#/components/schemas/Outcome' output: type: string description: What the tool returned, possibly truncated. structured: description: >- The structured content an MCP tool returned beside its text, unchanged. Absent when the tool returned none or it was too large. screenshot: $ref: '#/components/schemas/Screenshot' required: - run_id - seq - prompt - captured_at - frame_seqs - type - tool_call_id - outcome - output additionalProperties: false description: >- A tool call ended; its `tool_call_id` is the one of the `tool_started` event. SubagentStartedEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - subagent_started description: type: string description: What the agent asked the subagent to do, in short. required: - run_id - seq - prompt - captured_at - frame_seqs - type additionalProperties: false description: >- A subagent started; its `subagent_id` is the `tool_call_id` of the `tool_started` event that launched it. That tool call may end before the subagent does: the subagent is done at its `subagent_ended`. SubagentEndedEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - subagent_ended outcome: $ref: '#/components/schemas/Outcome' required: - run_id - seq - prompt - captured_at - frame_seqs - type - outcome additionalProperties: false description: A subagent ended. PageVisitedEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - page_visited tool_call_id: type: string description: >- The browser tool call that led to the page, when the agent had made one. page_id: type: string description: >- Identifier of the page in the run: a later visit to the same page keeps it. version: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- How many times the page was seen changed, this one included, starting at 1. url: type: string description: URL of the page. title: type: string description: Title of the page, empty when it has none. reason: type: string enum: - arrival - settled - after_submit - error - after_action - prompt_end description: >- Why it was captured: on arrival, once settled, after a form was submitted, on an error, after a browser action, or when the prompt ended. screenshot: $ref: '#/components/schemas/Screenshot_1' required: - run_id - seq - prompt - captured_at - frame_seqs - type - page_id - version - url - title - reason additionalProperties: false description: The agent's browser showed a page, new or changed. StopEvent: type: object properties: run_id: $ref: '#/components/schemas/RunId' seq: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- Position of the event in the run, starting at 1. Always increasing, but some numbers may be skipped. prompt: type: integer minimum: 1 maximum: 9007199254740991 description: Index of the prompt the event belongs to. captured_at: type: string subagent_id: type: string description: >- The subagent that did it: the `tool_call_id` of the `tool_started` event that launched it. Absent when the run's agent did it. frame_seqs: type: array items: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- The `seq` of each output line the event was read from. Empty for events we record ourselves. type: type: string enum: - stop stop_reason: type: string enum: - completed - failed - canceled description: >- Why the agent stopped. When `failed`, `error` may explain what went wrong. error: type: string description: What went wrong. required: - run_id - seq - prompt - captured_at - frame_seqs - type - stop_reason additionalProperties: false description: >- The agent stopped working on the prompt; the last event of its prompt. It ends everything still open in the prompt: any tool call or subagent without its own end event ended with it. InvalidRequestEncoded: type: object properties: _tag: type: string enum: - InvalidRequest message: type: string description: Every problem found, in one sentence. location: type: string enum: - path - query - header - body description: The part of the request that is wrong. issues: type: array items: $ref: '#/components/schemas/RequestIssue' description: Each problem found, with where it is. required: - _tag - message - location - issues additionalProperties: false ApiKeyRefusedEncoded: type: object properties: _tag: type: string enum: - ApiKeyRefused message: type: string required: - _tag - message additionalProperties: false RunNotFoundEncoded: type: object properties: _tag: type: string enum: - RunNotFound message: type: string required: - _tag - message additionalProperties: false Outcome: type: string enum: - succeeded - failed description: Whether the tool call or subagent succeeded. Screenshot: type: object properties: page_url: type: string description: Address of the page shown. required: - page_url additionalProperties: false description: >- The page a browser tool call left its browser on, as the harness captured it. Absent when it captured none. Screenshot_1: type: object properties: page_url: type: string description: Address of the page shown. required: - page_url additionalProperties: false description: >- The page as the browser captured it. Absent when it captured none; its image may be missing when the sandbox was gone before it was kept. RequestIssue: type: object properties: path: type: array items: anyOf: - type: string - type: number - type: string enum: - Infinity - '-Infinity' - NaN description: >- Keys leading to the wrong value inside that part of the request, such as `["agent", "model"]`. Empty when the problem concerns that part as a whole. message: type: string description: What is wrong with the value. required: - path - message additionalProperties: false description: One problem found in the request. securitySchemes: bearer: description: >- An API key of your organization, starting with `trk_`, sent as `Authorization: Bearer `. type: http scheme: bearer ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.