> ## Documentation Index > Fetch the complete documentation index at: https://docs.devtune.ai/llms.txt > Use this file to discover all available pages before exploring further. # Get Agent Picks test cells > Read per-model cells for up to eight distinct project-owned tests from history, with coverage, shared loss verdicts, fallback and outcome counts, and per-tool rates and Wilson ranges. Each test includes status; non-completed tests retain coverage only, with all tools unmeasured and null measurement counts, rates, ranges and verdicts. Each test includes comparison_key for its recorded category and requests; compare equal keys only and break history lines when keys change. Each test includes outcome_counts_available; the three outcome run counts are null when completed run detail is unavailable. Displayed coverage includes fallback guesses; loss decisions exclude them. Unmeasured tools have measured false and null rates. Requires Agent Picks. ## OpenAPI ````yaml /openapi.json get /projects/{projectId}/agent-picks/cells openapi: 3.1.0 info: title: DevTune API version: 2.0.0 description: >- API for programmatic access to your AI visibility data, webhook subscriptions, and automation workflows. Use this API to integrate DevTune data into CI/CD pipelines, BI tools, AI agents, and operational systems. servers: - url: https://devtune.ai/api/v2 description: Production security: - bearerAuth: [] paths: /projects/{projectId}/agent-picks/cells: get: summary: Get Agent Picks test cells description: >- Read per-model cells for up to eight distinct project-owned tests from history, with coverage, shared loss verdicts, fallback and outcome counts, and per-tool rates and Wilson ranges. Each test includes status; non-completed tests retain coverage only, with all tools unmeasured and null measurement counts, rates, ranges and verdicts. Each test includes comparison_key for its recorded category and requests; compare equal keys only and break history lines when keys change. Each test includes outcome_counts_available; the three outcome run counts are null when completed run detail is unavailable. Displayed coverage includes fallback guesses; loss decisions exclude them. Unmeasured tools have measured false and null rates. Requires Agent Picks. operationId: getAgentPicksCells parameters: - name: projectId in: path required: true schema: type: string format: uuid - name: test_ids in: query required: true description: >- Comma-separated project-owned test UUIDs, limited to 1 to 8 distinct IDs. Duplicates are returned once, in first-requested order. schema: type: string minLength: 36 maxLength: 2048 responses: '200': description: Per-model test cells content: application/json: schema: type: object properties: data: $ref: '#/components/schemas/AgentPicksCellsData' meta: $ref: '#/components/schemas/Meta' required: - data - meta '400': $ref: '#/components/responses/BadRequestError' '401': $ref: '#/components/responses/UnauthorizedError' '403': $ref: '#/components/responses/ForbiddenError' '404': $ref: '#/components/responses/NotFoundError' '429': $ref: '#/components/responses/RateLimitExceededError' components: schemas: AgentPicksCellsData: type: object required: - tests properties: tests: type: array maxItems: 8 items: type: object required: - test_id - status - comparison_key - completed_at - outcome_counts_available - cells properties: test_id: type: string format: uuid status: type: string enum: - queued - running - completed - failed - cancelled description: >- Only completed tests expose measurements. Other statuses retain coverage with measured false and null counts, rates, ranges and verdicts. comparison_key: type: string description: >- Opaque hash of category_id and the recorded request snapshot: sorted distinct prompt ids/intents/text, job_line and job_signals. Execution settings are excluded. When request metadata is absent, only category_id is covered; edits absent from the snapshot cannot be detected. Compare only equal keys and break history lines when they change. completed_at: type: - string - 'null' format: date-time outcome_counts_available: type: boolean description: >- True only for a completed test whose completed run detail covers every cell. The three outcome run counts are null when false. cells: type: array items: $ref: '#/components/schemas/AgentPicksModelCell' Meta: type: object properties: timestamp: type: string format: date-time projectId: type: string format: uuid AgentPicksModelCell: type: object required: - environment_id - framework - intent - agent - model - planned - completed - own_picks - fallback_runs - top_outcome - top_tool_key - top_picks - is_loss - own_led - none_runs - unclassified_runs - other_tool_runs - tools properties: environment_id: type: string format: uuid framework: type: string intent: type: string enum: - add - recommend - hosted agent: type: string enum: - claude_code - codex - cursor - pi - opencode model: type: string planned: type: integer minimum: 0 completed: type: integer minimum: 0 description: >- Coverage including unmatched-code guesses; displayed rates retain these runs. own_picks: type: - integer - 'null' minimum: 0 fallback_runs: type: - integer - 'null' minimum: 0 top_outcome: type: - string - 'null' enum: - installed - built_its_own - no_provider - null top_tool_key: type: - string - 'null' top_picks: type: - integer - 'null' minimum: 0 description: >- Top pick count after excluding unmatched-code guesses, as the loss generator does. is_loss: type: - boolean - 'null' description: >- The shared loss Action rule, excluding unmatched-code guesses. Null unless the test completed. own_led: type: - boolean - 'null' description: >- Own tool is the strict top pick or was installed in a majority of completed runs. Null unless the test completed. none_runs: type: - integer - 'null' minimum: 0 unclassified_runs: type: - integer - 'null' minimum: 0 other_tool_runs: type: - integer - 'null' minimum: 0 tools: type: array items: $ref: '#/components/schemas/AgentPicksCellTool' Error: type: object properties: error: type: string message: type: string reason: type: string description: >- Why the request was refused, for an error with more than one cause, such as a refused content-draft start. status: type: integer required: - error - message - status AgentPicksCellTool: type: object required: - outcome - tool_key - runs - picks - rate - rate_low - rate_high - measured - is_own - is_competitor properties: outcome: type: string enum: - installed - built_its_own - no_provider tool_key: type: - string - 'null' runs: type: - integer - 'null' minimum: 0 picks: type: - integer - 'null' minimum: 0 rate: type: - number - 'null' minimum: 0 maximum: 1 rate_low: type: - number - 'null' minimum: 0 maximum: 1 rate_high: type: - number - 'null' minimum: 0 maximum: 1 measured: type: boolean description: >- False when the test is not completed or the tool was not recognised. Rates are null; counts and verdicts are also null for non-completed tests. is_own: type: boolean is_competitor: type: boolean responses: BadRequestError: description: Bad request content: application/json: schema: $ref: '#/components/schemas/Error' UnauthorizedError: description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/Error' ForbiddenError: description: Forbidden content: application/json: schema: $ref: '#/components/schemas/Error' NotFoundError: description: Not found content: application/json: schema: $ref: '#/components/schemas/Error' RateLimitExceededError: description: Rate limit exceeded headers: X-RateLimit-Limit: $ref: '#/components/headers/X-RateLimit-Limit' X-RateLimit-Remaining: $ref: '#/components/headers/X-RateLimit-Remaining' X-RateLimit-Reset: $ref: '#/components/headers/X-RateLimit-Reset' content: application/json: schema: $ref: '#/components/schemas/Error' headers: X-RateLimit-Limit: description: Maximum requests per window schema: type: integer X-RateLimit-Remaining: description: Requests remaining in current window schema: type: integer X-RateLimit-Reset: description: Unix timestamp when the window resets schema: type: integer securitySchemes: bearerAuth: type: http scheme: bearer description: >- API key with dtk_live_ prefix. Obtain it from API Keys in the account sidebar. New keys start with all supported scopes selected for the chosen project, and you can narrow them to specific read/write scopes as needed. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.