usewrit-app Data API

openapi: 3.2.0
info:
  title: Writ Agent Data API
  version: 0.1.0
  license:
    name: MIT
    identifier: MIT
  description: 'Loopback HTTP API of the Writ local agent (`writ-agentd`), the daemon embedded in the

    Writ desktop app and the self-host build. This document is generated against the Rust

    daemon source (`writ-agent/src/local/`), which is the wire truth.


    ## Transport


    - Plain HTTP on `127.0.0.1:8131` (default; `WRIT_PORT` env overrides).

    - Optional HTTPS twin on `127.0.0.1:8132`, served with a certificate minted by a

    per-install, name-constrained local CA (`~/.writ/tls/ca.pem`). Clients must trust that

    CA file (or pin it) to use the TLS port; the payloads are identical on both ports.

    - The daemon enforces a DNS-rebind guard: `Host`/`Origin` must be loopback. Talk to

    `127.0.0.1`, not `localhost`.


    ## Errors


    Domain errors are JSON `{"error": "", "code": ""}` (see the

    `Error` schema). Codes include `bad_request`, `unauthorized`, `forbidden`, `not_found`,

    `vault_locked` (HTTP 423), `too_many_requests`, `device_capacity` (HTTP 409),

    `captcha_required` (HTTP 402), `internal`.


    **Caveat:** some 4xx bodies are `text/plain`, not JSON — axum framework rejections

    (e.g. a JSON body that fails to deserialize returns

    `Failed to deserialize the JSON body into the target type: missing field ''url''` as

    plain text with a 4xx status). Clients must tolerate non-JSON error bodies.


    ## List envelopes


    The wire is deliberately documented as-is (SDKs normalize; this spec does not):


    - `{"data": [...], "count": n}` — workflows, personas, persona runs, secrets, files, keys

    - `{"data": [...], "count": n, "total": n}` — runs

    - **bare JSON array** — monitors, monitor selectors, extractors, automations,

    `/v1/changes/recent`

    - bespoke objects — `/v1/monitors/{id}/changes`, `/v1/data`, the workflow-data table'
servers:
- url: http://127.0.0.1:8131
  description: Plain-HTTP loopback listener (default port; WRIT_PORT overrides).
- url: https://127.0.0.1:8132
  description: HTTPS twin of the same API, terminated with a certificate from the per-install local CA at ~/.writ/tls/ca.pem. Trust that CA (or pin it) in your HTTP client.
security:
- bearerAuth: []
tags:
- name: Data
  description: Extracted-data explorer — aggregate tables, facets, exports, lineage runs.
paths:
  /v1/data:
    get:
      operationId: listDataWorkflows
      tags:
      - Data
      summary: Data-explorer workflow picker
      description: Every workflow that has produced extracted data in a SUCCESSFUL run, with a contributing-run count, last-data timestamp and a small `last_delta` summary (null for large or single-snapshot datasets). Sorted by most recent data first.
      responses:
        '200':
          description: Picker entries under a `workflows` key.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/DataWorkflowsList'
        '401':
          $ref: '#/components/responses/Unauthorized'
  /v1/workflows/{id}/data:
    parameters:
    - $ref: '#/components/parameters/WorkflowId'
    get:
      operationId: getWorkflowData
      tags:
      - Data
      summary: Aggregated extracted-data table
      description: 'Aggregates the workflow''s successful runs'' `extracted_data` into one sortable / searchable table. `view=all` (default) is the flat grid; `view=latest` is the deduplicated current dataset (adds `identity`, `counts`, `sources`, per-row `lineage`); `view=run&run_id=N` is one snapshot annotated vs the previous chain member (adds `delta`, `removed_records`, `prev_run_id`). Lens-parameter violations return a FastAPI-shaped 400 body `{"detail": "..."}` (not the standard Error shape).'
      parameters:
      - name: q
        in: query
        description: Global substring filter across extracted fields.
        schema:
          type: string
      - name: filter
        in: query
        description: Per-column legacy filter `column:substring` (repeatable).
        schema:
          type: array
          items:
            type: string
        explode: true
      - name: filters
        in: query
        description: Structured smart filters — a JSON array of clauses, e.g. `[{"col":"price","op":"between","min":10,"max":50}]`.
        schema:
          type: string
      - name: sort_by
        in: query
        description: A data column or `run_at` / `status` / `duration_ms` / `run_id` (lens views also accept `first_seen_at` / `last_seen_at` / `versions`).
        schema:
          type: string
      - name: sort_dir
        in: query
        schema:
          type: string
          enum:
          - asc
          - desc
          default: desc
      - name: limit
        in: query
        description: Page size (clamped to 1..=500; default 50).
        schema:
          type: integer
          default: 50
      - name: offset
        in: query
        schema:
          type: integer
          default: 0
      - name: include_inputs
        in: query
        description: Also surface run input values as `input.<name>` columns (view=all only).
        schema:
          type: boolean
          default: false
      - name: view
        in: query
        description: 'Lens: `latest` | `run` | `all` (default `all`).'
        schema:
          type: string
          enum:
          - all
          - latest
          - run
          default: all
      - name: run_id
        in: query
        description: The snapshot to show — required when `view=run`.
        schema:
          type: integer
          format: int64
      - name: key
        in: query
        description: Explicit identity key — comma-separated field names (identity pinning echo).
        schema:
          type: string
      - name: include_missing
        in: query
        description: view=latest only. Include records missing from the newest snapshot. Parsed STRICTLY — only the literal string `true` enables it.
        schema:
          type: string
      - name: source
        in: query
        description: 'view=latest/run: only records from this originating list key (empty string selects untagged records).'
        schema:
          type: string
      responses:
        '200':
          description: The aggregated data table (view-dependent extras included).
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/WorkflowDataTable'
        '400':
          description: 'Lens-parameter violation. NOTE the body is the FastAPI-parity shape `{"detail": "..."}`, not the standard Error object.'
          content:
            application/json:
              schema:
                type: object
                required:
                - detail
                properties:
                  detail:
                    type: string
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
    delete:
      operationId: deleteWorkflowData
      tags:
      - Data
      summary: Delete extracted-data rows
      description: Removes extracted rows from the workflow's runs. The delete set is the union of the explicit `records` refs and every stored version of each uid in `record_uids`. The run rows themselves are preserved — only their `extracted_data` is mutated.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/DeleteWorkflowDataRequest'
      responses:
        '200':
          description: Deletion summary.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/DeleteWorkflowDataResult'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
  /v1/workflows/{id}/data/runs:
    get:
      operationId: listWorkflowDataRuns
      tags:
      - Data
      summary: Data snapshot index
      description: Every successful data-bearing run (newest first) with record count, explicit-empty flag and delta vs the previous chain member (null for the oldest), plus the chosen identity.
      parameters:
      - $ref: '#/components/parameters/WorkflowId'
      - name: key
        in: query
        description: Explicit identity key — comma-separated field names.
        schema:
          type: string
      responses:
        '200':
          description: The snapshot index.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/DataRunsIndex'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
  /v1/workflows/{id}/data/facets:
    get:
      operationId: getWorkflowDataFacets
      tags:
      - Data
      summary: Per-column facets
      description: Per-column facets over the workflow's extracted data (inferred type, non-empty count, numeric min/max, distinct pick-list when low-cardinality). `view=latest|run` computes over exactly that lens's rowset and adds a `sources` envelope.
      parameters:
      - $ref: '#/components/parameters/WorkflowId'
      - name: include_inputs
        in: query
        schema:
          type: boolean
          default: false
      - name: view
        in: query
        schema:
          type: string
          enum:
          - all
          - latest
          - run
          default: all
      - name: run_id
        in: query
        schema:
          type: integer
          format: int64
      - name: key
        in: query
        schema:
          type: string
      - name: include_missing
        in: query
        description: Strict `true` includes missing records in the view=latest rowset.
        schema:
          type: string
      - name: source
        in: query
        schema:
          type: string
      responses:
        '200':
          description: Facets over the (lens) rowset.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/DataFacets'
        '400':
          description: Lens-parameter violation (FastAPI-shaped `detail` body).
          content:
            application/json:
              schema:
                type: object
                required:
                - detail
                properties:
                  detail:
                    type: string
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
  /v1/workflows/{id}/data/export:
    get:
      operationId: exportWorkflowData
      tags:
      - Data
      summary: Export the extracted-data table
      description: Downloads the full (search/sort/filter applied, UN-paginated) table as CSV (`?format=csv`, default) or JSON (`?format=json`), bounded by the 1000-run scan cap. Lens views (`view=latest|run`) attach lineage — extra CSV columns, or a `_lineage` object on each JSON row. Accepts the same query parameters as the table endpoint (`q`, `filter`, `filters`, `sort_by`, `sort_dir`, `view`, `run_id`, `key`, `include_missing`, `source`) plus `format`.
      parameters:
      - $ref: '#/components/parameters/WorkflowId'
      - name: format
        in: query
        description: Output shape. `csv` (default) or `json` return the tabular data; `markdown` and `html` render it and are CONTENT-AWARE — a crawl's pages (records carrying a long-form `markdown` column) export as readable documents, structured data as a table.
        schema:
          type: string
          enum:
          - csv
          - json
          - markdown
          - html
          default: csv
      - name: q
        in: query
        schema:
          type: string
      - name: filter
        in: query
        schema:
          type: array
          items:
            type: string
        explode: true
      - name: filters
        in: query
        schema:
          type: string
      - name: sort_by
        in: query
        schema:
          type: string
      - name: sort_dir
        in: query
        schema:
          type: string
          enum:
          - asc
          - desc
          default: desc
      - name: view
        in: query
        schema:
          type: string
          enum:
          - all
          - latest
          - run
          default: all
      - name: run_id
        in: query
        schema:
          type: integer
          format: int64
      - name: key
        in: query
        schema:
          type: string
      - name: include_missing
        in: query
        schema:
          type: string
      - name: source
        in: query
        schema:
          type: string
      responses:
        '200':
          description: The exported table as a CSV or JSON attachment.
          content:
            text/csv:
              schema:
                type: string
                description: RFC-4180 CSV; run-meta columns, data columns, then lineage columns (lens views).
            application/json:
              schema:
                type: array
                description: One object per row; lens views add a `_lineage` object per row.
                items:
                  type: object
        '400':
          description: Lens-parameter violation (FastAPI-shaped `detail` body).
          content:
            application/json:
              schema:
                type: object
                required:
                - detail
                properties:
                  detail:
                    type: string
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
components:
  parameters:
    WorkflowId:
      name: id
      in: path
      required: true
      description: Workflow row id.
      schema:
        type: integer
        format: int64
  schemas:
    Error:
      type: object
      description: 'The daemon''s domain error body. NOTE: some 4xx responses (axum framework rejections) are `text/plain` instead of this JSON shape. Certain codes attach extra fields (e.g. `device_capacity` adds `would_use_units_per_min` / `budget_units_per_min`; `captcha_required` adds `captcha_type` / `url`).'
      required:
      - error
      - code
      properties:
        error:
          type: string
          description: 'Human-readable message (e.g. `not found: workflow 999999`).'
        code:
          type: string
          description: 'Stable machine code: `bad_request`, `unauthorized`, `forbidden`, `not_found`, `vault_locked`, `too_many_requests`, `captcha_required`, `device_capacity`, `cloud_unreachable`, `db_error`, `io_error`, `json_error`, `internal`, …'
      additionalProperties: true
    DataRunsIndex:
      type: object
      description: The snapshot index (`GET /v1/workflows/{id}/data/runs`).
      required:
      - runs
      - identity
      - scanned_runs
      - truncated
      properties:
        runs:
          type: array
          description: Newest-first successful data-bearing runs; each entry carries `run_id`, record count, explicit-empty flag, per-snapshot delta (null for the oldest) and the run's `status`.
          items:
            type: object
            additionalProperties: true
        identity:
          type: object
          description: The chosen record identity.
          additionalProperties: true
        scanned_runs:
          type: integer
        truncated:
          type: boolean
    DeleteWorkflowDataResult:
      type: object
      required:
      - deleted
      - resolved
      - unmatched
      properties:
        deleted:
          type: integer
          description: Number of extracted rows removed.
        resolved:
          type: object
          description: '`{uid: n_versions}` for each matched uid.'
          additionalProperties:
            type: integer
        unmatched:
          type: array
          description: Requested uids that matched nothing.
          items:
            type: string
    WorkflowDataTable:
      type: object
      description: The aggregated extracted-data table. The base fields are always present; lens views add `identity` (+ `counts`/`sources` on latest, + `delta`/`removed_records`/ `prev_run_id`/`sources` on run views).
      required:
      - workflow_id
      - workflow_name
      - columns
      - declared
      - rows
      - total
      - scanned_runs
      - truncated
      - limit
      - offset
      properties:
        workflow_id:
          type: integer
          format: int64
        workflow_name:
          type: string
        columns:
          type: array
          items:
            type: string
        declared:
          type: boolean
          description: Whether the columns come from the workflow's declared output fields.
        rows:
          type: array
          items:
            $ref: '#/components/schemas/DataTableRow'
        total:
          type: integer
          description: Post-filter row total before pagination.
        scanned_runs:
          type: integer
          description: Successful data-bearing runs scanned (capped at 1000).
        truncated:
          type: boolean
          description: True when the scan hit the 1000-run cap.
        limit:
          type:
          - integer
          - 'null'
        offset:
          type: integer
        identity:
          type: object
          description: Lens views only — the chosen record identity.
          additionalProperties: true
        counts:
          type: object
          description: view=latest only — post-dedup change counts.
          properties:
            new:
              type: integer
            changed:
              type: integer
            same:
              type: integer
            missing:
              type: integer
        sources:
          type: object
          description: 'Lens views only — `{<list key | "">: count}` buckets of the rowset by originating list key (pre-source-filter).'
          additionalProperties:
            type: integer
        delta:
          type: object
          description: view=run only — the snapshot delta vs the previous chain member.
          additionalProperties: true
        removed_records:
          type: array
          description: view=run only — records that vanished vs the previous snapshot.
          items:
            type: object
            required:
            - uid
            - fields
            properties:
              uid:
                type: string
              fields:
                type: object
                additionalProperties: true
        prev_run_id:
          type:
          - integer
          - 'null'
          format: int64
          description: view=run only — the previous chain member (null for the oldest).
    DataFacets:
      type: object
      description: Per-column facets (`GET /v1/workflows/{id}/data/facets`).
      required:
      - workflow_id
      - columns
      - facets
      - row_count
      - scanned_runs
      - truncated
      properties:
        workflow_id:
          type: integer
          format: int64
        columns:
          type: array
          items:
            type: string
        facets:
          type: object
          description: Per-column facet descriptors (inferred type, non-empty count, numeric min/max, distinct pick-list when low-cardinality).
          additionalProperties: true
        row_count:
          type: integer
        scanned_runs:
          type: integer
        truncated:
          type: boolean
        sources:
          type: object
          description: Lens views only — pre-filter source buckets.
          additionalProperties:
            type: integer
    DataWorkflowSummary:
      type: object
      required:
      - workflow_id
      - workflow_name
      - run_count
      - last_data_at
      - last_delta
      properties:
        workflow_id:
          type: integer
          format: int64
        workflow_name:
          type: string
        run_count:
          type: integer
          description: Distinct successful runs that contributed at least one row.
        last_data_at:
          type:
          - string
          - 'null'
          description: Newest contributing run's timestamp.
        last_delta:
          description: Newest two snapshots compared under the sampled identity; null when the dataset has fewer than 2 snapshots or is too large.
          oneOf:
          - type: object
            required:
            - new
            - changed
            - removed
            properties:
              new:
                type: integer
              changed:
                type: integer
              removed:
                type: integer
          - type: 'null'
    DataTableRow:
      type: object
      description: 'One table row: run-meta at the top level, data columns nested under `fields`. Lens views (`view=latest|run`) add a `lineage` sibling (`uid`, `change`, `changed_fields` — present only when change is `changed` —, `changed_leaf_count`, `first_seen_at`, `last_seen_at`, `versions`, `source`, and `prev_run_id` on run views).'
      required:
      - run_id
      - run_at
      - status
      - record_index
      - fields
      properties:
        run_id:
          type: integer
          format: int64
        run_at:
          type:
          - string
          - 'null'
        status:
          type:
          - string
          - 'null'
        record_index:
          type: integer
        fields:
          type: object
          description: The extracted data columns (redacted; a column absent for a row is null).
          additionalProperties: true
        lineage:
          type: object
          description: Lens views only.
          additionalProperties: true
    DataWorkflowsList:
      type: object
      description: The Data-explorer workflow picker payload.
      required:
      - workflows
      properties:
        workflows:
          type: array
          items:
            $ref: '#/components/schemas/DataWorkflowSummary'
    DeleteWorkflowDataRequest:
      type: object
      description: Body for `DELETE /v1/workflows/{id}/data`. The delete set is `records` ∪ every stored version of each uid in `record_uids`. An empty body deletes nothing (200 with zeros).
      properties:
        records:
          type: array
          items:
            type: object
            required:
            - run_id
            properties:
              run_id:
                type: integer
                format: int64
              record_index:
                type: integer
                default: 0
        record_uids:
          type: array
          description: Lineage uids — each expands to EVERY stored version.
          items:
            type: string
        key:
          type:
          - string
          - 'null'
          description: Identity pinning echo — the identity fields the uids came from.
  responses:
    NotFound:
      description: The addressed resource does not exist.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
    Unauthorized:
      description: Missing/invalid bearer token, or the Origin/Host guard rejected the request.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: "Every route (including `/v1/agent` and `/v1/health`) requires\n`Authorization: Bearer <token>`. Token families:\n\n- `wlt_…` — the runtime token from `runtime.json` (full access, used by the desktop\n  UI and SDK discovery).\n- `wlk_…` — a scoped API key minted via `POST /v1/keys` (CSV scopes `read|run|admin`;\n  `/v1/keys` itself requires the full-access token).\n- `wlo_…` — an OAuth 2.1 access token from the daemon's local authorization server\n  (`run` scope).\n\nTreat the token as an opaque string. The daemon additionally enforces a loopback\nOrigin/Host DNS-rebind guard.\n"