> ## Documentation Index > Fetch the complete documentation index at: https://benchgen.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Get a benchmark bundle manifest > The `competition.yaml` this benchmark would be described by, plus every file it names: its pages, its terms and its logo. Together with the four task files from `task_files`, that is a complete bundle, which is what makes a saved copy something `create_from_repo` can build from. Without it a saved folder holds a benchmark’s files but not its settings, and cannot be created from at all. No zip is returned: the heavy parts already come down through their own download links, so sending them again here would double the traffic to say nothing new. Requires the `benchmark:read` scope, and the caller must be the creator or a collaborator. The manifest always names an image, terms and at least one page, with stand-ins where the benchmark has none, so a folder written from this answer passes the same checks an upload does. The `competition.yaml` this benchmark would be described by, plus the text files it names. Together with the four task files from [list benchmark files](/docs/api-reference/endpoint/list-benchmark-files.md), that is a complete bundle. Which matters more than it sounds: `competition.yaml` sits in the root of a bundle and names everything else, so a saved copy without it holds a benchmark's files but not its settings, and [create from repository](/docs/api-reference/endpoint/create-benchmark-repo.md) cannot read it at all. ## No zip, on purpose The scoring program, ingestion program, questions and answers already come down through their own download links. Sending them again here would double the traffic to say nothing new. What this returns is the text that has nowhere else to come from. ## OpenAPI ````yaml GET /api/competitions/{id}/bundle_manifest/ openapi: 3.0.3 info: title: BenchGen Platform API version: 1.0.0 description: >- One API for the whole BenchGen platform: model catalogue and serving, fine-tuning and benchmarks, datasets (knowledge), and billing. ## Authentication Every request uses the same credential: a platform API token sent as `Authorization: Bearer bgn_...`. Create tokens in the web app under Profile Settings > Platform API tokens (the secret is shown exactly once), or via `POST /api/tokens/` with an interactive session. Revoking a token disables it platform-wide within 60 seconds. ## Scopes A token carries scopes chosen at creation; a request outside the token's scopes gets `403` with an explanatory message. | scope | grants | |---|---| | `models:read` | read model catalogues, job status, logs, GPU info | | `models:write` | deploy, train, merge, stop models and jobs | | `benchmark:read` | read benchmark runs and results | | `benchmark:run` | launch benchmark runs | | `benchmark:create` | create benchmarks in your account (drafts, Excel, bundles, specs); counts toward the creation limit | | `benchmark:manage` | edit and delete benchmarks you own or collaborate on | | `benchmark:publish` | publish and unpublish benchmarks you own or collaborate on | | `knowledge:read` | read your datasets and fine-tuning data | | `knowledge:write` | create, edit and delete datasets and fine-tuning data | | `billing:read` | read your balance and usage | | `agents:chat` | chat with your own agents through the API | | `agents:manage` | manage your agents, knowledge bases and channels | | `admin` | everything the account can do (staff accounts only) | ## For agents This document plus `/api/llms.txt` are the machine-readable entry points. Responses are JSON. Errors use conventional status codes; the body carries `error` or `message`. Knowledge endpoints return `[{"data": [...], "meta": {...}}]`. The complete auto-generated schema of every endpoint (including internal ones) lives at `/api/public-docs.json` (Swagger 2.0); this document is the curated, stable, supported surface. contact: url: https://benchgen.com servers: - url: https://api.benchgen.com security: - platformToken: [] tags: - name: auth description: Token introspection for services and integrations - name: tokens description: Manage your platform API tokens - name: models description: Model catalogue and serving - name: finetune description: Fine-tuning jobs, inference deployments, GPUs - name: knowledge description: Datasets and fine-tuning data (knowledge API) - name: billing description: Balance and usage - name: agents description: Chat with your agents (OpenAI-compatible facade) - name: benchmark description: Public benchmark (competition) listings. paths: /api/competitions/{id}/bundle_manifest/: get: tags: - benchmark summary: The bundle manifest for a benchmark description: >- The `competition.yaml` this benchmark would be described by, plus every file it names: its pages, its terms and its logo. Together with the four task files from `task_files`, that is a complete bundle, which is what makes a saved copy something `create_from_repo` can build from. Without it a saved folder holds a benchmark’s files but not its settings, and cannot be created from at all. No zip is returned: the heavy parts already come down through their own download links, so sending them again here would double the traffic to say nothing new. Requires the `benchmark:read` scope, and the caller must be the creator or a collaborator. The manifest always names an image, terms and at least one page, with stand-ins where the benchmark has none, so a folder written from this answer passes the same checks an upload does. parameters: - name: id in: path required: true schema: type: integer responses: '200': description: The manifest and the files it names. content: application/json: schema: type: object properties: competition_yaml: type: string description: Write this to competition.yaml in the folder root. files: type: object description: >- Text files the manifest names, path to contents: the pages and the terms. Always includes terms and at least one page, because the upload refuses a bundle without them. logo: type: string description: >- The path of the image the manifest names; its bytes are in binary_files. binary_files: type: object description: >- Binary files the manifest names, path to base64 contents. Holds the logo: the benchmark’s own, or the BenchGen mark when it has none, because the upload refuses a bundle without an image. security: - PlatformToken: [] components: securitySchemes: platformToken: type: http scheme: bearer bearerFormat: bgn_ opaque token description: >- Platform API token created under Profile Settings > Platform API tokens. Scopes are fixed at creation. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.