> ## Documentation Index > Fetch the complete documentation index at: https://docs.staging.alpic.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create a submission package > Create a submission package for an environment. Alpic reads the environment's live MCP server and prefills every field it can infer, so the response already carries the populated package. The environment must have a successful deployment and must not require authentication. ## OpenAPI ````yaml https://api.alpic.ai/openapi.json post /v1/submissions openapi: 3.1.1 info: title: Alpic API version: 1.0.0 servers: - url: https://api.alpic.ai security: - bearerAuth: [] paths: /v1/submissions: post: tags: - submissions summary: Create a submission package description: >- Create a submission package for an environment. Alpic reads the environment's live MCP server and prefills every field it can infer, so the response already carries the populated package. The environment must have a successful deployment and must not require authentication. operationId: submissions.create.v1 requestBody: required: true content: application/json: schema: type: object properties: environmentId: type: string description: The ID of the environment to build the package from platform: enum: - chatgpt - claudeai type: string description: >- The directory the package targets: `chatgpt` for the OpenAI App Directory, `claudeai` for the Anthropic MCP Directory required: - environmentId - platform responses: '200': description: The created submission package, with its prefilled fields content: application/json: schema: $ref: '#/components/schemas/Submission' '400': description: '400' content: application/json: schema: oneOf: - type: object properties: defined: const: true code: const: BAD_REQUEST status: const: 400 message: type: string default: Bad Request data: {} required: - defined - code - status - message - type: object properties: defined: const: false code: type: string status: type: number message: type: string data: {} required: - defined - code - status - message '404': description: '404' content: application/json: schema: oneOf: - type: object properties: defined: const: true code: const: NOT_FOUND status: const: 404 message: type: string default: Not Found data: {} required: - defined - code - status - message - type: object properties: defined: const: false code: type: string status: type: number message: type: string data: {} required: - defined - code - status - message components: schemas: Submission: anyOf: - type: object properties: id: type: string environmentId: type: string auditId: anyOf: - type: string - type: 'null' urlVerification: anyOf: - type: object properties: verifiedAt: type: string format: date-time x-native-type: date results: type: array items: type: object properties: field: enum: - privacyPolicyUrl - termsOfServiceUrl - privacyPolicy type: string url: type: string outcome: enum: - pass - fail - unverified type: string reason: type: string concerns: type: array items: type: string scrapedWith: enum: - fetch - firecrawl type: string required: - field - url - outcome - reason required: - verifiedAt - results - type: 'null' default: null createdAt: type: string format: date-time x-native-type: date updatedAt: type: string format: date-time x-native-type: date platform: const: claudeai formData: $ref: '#/components/schemas/ClaudeSubmissionFormData' prefillSnapshot: anyOf: - $ref: '#/components/schemas/ClaudeSubmissionFormData' - type: 'null' default: null required: - id - environmentId - auditId - createdAt - updatedAt - platform - formData - type: object properties: id: type: string environmentId: type: string auditId: anyOf: - type: string - type: 'null' urlVerification: anyOf: - type: object properties: verifiedAt: type: string format: date-time x-native-type: date results: type: array items: type: object properties: field: enum: - privacyPolicyUrl - termsOfServiceUrl - privacyPolicy type: string url: type: string outcome: enum: - pass - fail - unverified type: string reason: type: string concerns: type: array items: type: string scrapedWith: enum: - fetch - firecrawl type: string required: - field - url - outcome - reason required: - verifiedAt - results - type: 'null' default: null createdAt: type: string format: date-time x-native-type: date updatedAt: type: string format: date-time x-native-type: date platform: const: chatgpt formData: $ref: '#/components/schemas/ChatgptSubmissionFormData' prefillSnapshot: anyOf: - $ref: '#/components/schemas/ChatgptSubmissionFormData' - type: 'null' default: null required: - id - environmentId - auditId - createdAt - updatedAt - platform - formData ClaudeSubmissionFormData: type: object properties: serverUrl: type: string name: type: string slug: type: string oneLiner: type: string description: type: string categories: type: array items: enum: - Productivity - Communication - Data & Analytics - Development tools - Financial Services - Consumer Health - Health & Life Sciences - Media & Entertainment - Commerce & Shopping - Legal - Sales and marketing - Creative - Nonprofit - Education - Other type: string iconUrl: type: string documentation: type: string support: type: string privacyPolicy: type: string screenshots: type: array items: type: object properties: prompt: type: string url: type: string required: - prompt - url primaryUseCases: type: string description: >- Describe the main tasks users will accomplish with this connector, and include a few example prompts a user might type. These may be shown publicly but are primarily used for internal review — reviewers test your server using them. connectionRequirements: type: string description: >- What accounts, permissions, or setup a user needs before connecting (e.g. 'Requires a paid Acme account with admin permissions'). If there are genuinely no special requirements, say so explicitly. readWriteCapabilities: enum: - read_only - write_only - read_write type: string description: |- What your server can do: • 'read_only': retrieves data but never modifies it. • 'write_only': creates or modifies data but does not read it. • 'read_write': both retrieves and modifies data (most common). companyName: type: string description: >- The company or organization behind this connector — its legal name or publicly known product name. companyWebsite: type: string primaryContactName: type: string description: >- Full name of the person Anthropic should contact about this submission. Receives review feedback and follow-up questions. primaryContactEmail: type: string primaryContactRole: type: string description: >- The primary contact's role at the company (e.g. 'Engineer', 'Product Manager'). Optional. anthropicPointOfContact: type: string description: >- If you've been working with someone at Anthropic, their name or email. Optional, can expedite review. authentication: enum: - oauth_dcr - oauth_cimd - oauth_anthropic_creds - custom_connection - none type: string description: >- How users authenticate with your server: • 'oauth_dcr': OAuth 2.0 + Dynamic Client Registration (RFC 7591) — clients register themselves. • 'oauth_cimd': OAuth 2.0 + Client ID Metadata Document — a published client_id metadata document. • 'oauth_anthropic_creds': OAuth 2.0 with Anthropic-held client credentials (a static client_id, and secret if needed). • 'custom_connection': users supply their own URL or client details at connection time (e.g. Snowflake). • 'none': no authentication (authless) — anyone can connect and call tools without credentials. partialAuth: type: boolean description: >- Whether some tools return 401 and prompt for auth on demand while the server otherwise starts authless (partial auth). apiOwnership: enum: - first_party - proxied - third_party type: string description: >- How your connector relates to the API it calls: • 'first_party': the MCP server calls your own production APIs. • 'proxied': calls go to a third-party API you're authorized to integrate (e.g. a Zapier-style integration). • 'third_party': calls a third-party API you don't control — typically blocks listing; explain in 'additionalNotes'. personalHealthData: anyOf: - type: boolean - type: object properties: other: type: string required: - other description: >- Whether this connector accesses, processes, or stores personal health information. No → false, Yes → true, or { other: explanation } when it depends on configuration. sponsoredContent: anyOf: - type: boolean - type: object properties: other: type: string required: - other description: >- Whether this connector surfaces sponsored, promoted, or advertising content (prohibited in directory connectors). No → false, Yes → true, or { other: explanation }. testInstructions: type: string description: >- Clear test-account setup and access instructions for reviewers: every link, credential, and step needed to access the MCP server autonomously. For accounts that need to be populated, provide test credentials granting access to a fully populated account. selfTested: type: boolean description: >- Confirmation that you have run every tool via MCP Inspector or as a custom connector in Claude. policyAcknowledgements: type: array items: enum: - >- I have read and understand the MCP Directory developer guidelines. - >- My server calls our own first-party APIs (or APIs we legitimately proxy with permission). - >- My connector does not transfer money, cryptocurrency, or other financial assets on a user's behalf. - >- My connector does not use AI models to generate images, video, or audio. - >- Tool descriptions contain no instructions about model behavior, other tools, or external instruction sources, and no hidden or encoded text. - >- My connector does not collect conversation data beyond what is needed for its function. - >- Public documentation for this connector will be live by the publish date. type: string description: >- Anthropic directory policy acknowledgements. All items must be acknowledged before the claude.ai wizard allows submission. Each is an attestation you are legally agreeing to by submitting. additionalNotes: type: string description: >- Anything else the review team should know about your connector. Optional. ChatgptSubmissionFormData: type: object properties: logoLight: type: string description: >- Logo icon for light mode. Square PNG, no borders or rounded corners (clients apply circular cropping). logoDark: type: string description: >- Optional dark-mode logo icon. Square PNG, same specs as the light icon. Leave empty if the light icon also works on dark backgrounds. appName: type: string description: The name users will see in ChatGPT and in the Apps Directory. tagline: type: string description: type: string description: >- Clear, engaging description highlighting what the app does and why people will love it. Appears publicly on the directory page. End it by naming the product categories or intents the app does NOT cover, so ChatGPT learns when not to route to it (define exclusions by absent categories or wrong intents, never by naming other brands). category: enum: - BUSINESS - COLLABORATION - DESIGN - DEVELOPER_TOOLS - EDUCATION - ENTERTAINMENT - FINANCE - FOOD - LIFESTYLE - NEWS - PRODUCTIVITY - SHOPPING - TRAVEL type: string description: Category from the OpenAI taxonomy. developerName: type: string description: Developer name shown publicly on the app's directory page. companyUrl: type: string supportChannel: type: string privacyPolicyUrl: type: string termsOfServiceUrl: type: string demoRecordingUrl: type: string appCommerceAndPurchasing: type: boolean description: >- Checkbox: 'My app links or directs users out of ChatGPT to make purchases.' Defaults to false. Verify the app does not offer digital goods. serverUrl: type: string authentication: enum: - No auth needed - OAuth 2.0 type: string description: >- OAuth 2.0 or No auth. OAuth configuration is auto-discovered from the MCP server's metadata. tools: type: array items: type: object properties: name: type: string title: type: string description: >- Human-friendly name for the tool, used in the UI. If not provided, `name` will be used. description: type: string annotations: type: object properties: readOnlyHint: type: boolean destructiveHint: type: boolean openWorldHint: type: boolean idempotentHint: type: boolean required: - name description: >- List of tools exposed by the MCP server (auto-populated from the production server's manifest). toolJustifications: type: array items: type: object properties: toolName: type: string readOnlyJustification: type: string description: One sentence on what the tool reads and returns. openWorldJustification: type: string description: >- One sentence on whether the tool reaches external systems beyond your own service. destructiveJustification: type: string description: >- One sentence on whether the tool changes data. For a read-only tool, state the full triad: "No data is created, modified, or deleted." required: - toolName - readOnlyJustification - openWorldJustification - destructiveJustification description: >- Per-tool justification of each MCP annotation value (one specific sentence per hint). description: >- Per-tool justification of `readOnlyHint`, `openWorldHint`, and `destructiveHint` annotation values. One short, specific sentence per hint describing what the tool actually does. skills: type: array items: type: object properties: name: type: string required: - name description: A skill discovered on the MCP server, by name. description: >- Skills the server exposes over MCP (auto-populated from the production server's manifest). Left unset when the server does not advertise the skills extension, which is a different state from advertising it and exposing none. testCases: type: array items: type: object properties: scenario: type: string userPrompt: type: string toolTriggered: type: string description: The tool this prompt should invoke. expectedOutput: type: string description: >- What the user should see, including post-interaction behaviour where relevant. negativeTestCases: type: array items: type: object properties: scenario: type: string userPrompt: type: string screenshots: type: array items: type: object properties: prompt: type: string url: type: string required: - prompt - url translations: type: array items: type: object properties: locale: enum: - am - ar - bg-BG - bn-BD - bs-BA - ca-ES - cs-CZ - da-DK - de-DE - el-GR - es-419 - es-ES - et-EE - fi-FI - fr-CA - fr-FR - gu-IN - hi-IN - hr-HR - hu-HU - hy-AM - id-ID - is-IS - it-IT - ja-JP - ka-GE - kk - kn-IN - ko-KR - lt - lv-LV - mk-MK - ml - mn - mr-IN - ms-MY - my-MM - nb-NO - nl-NL - pa - pl-PL - pt-BR - pt-PT - ro-RO - ru-RU - sk-SK - sl-SI - so-SO - sq-AL - sr-RS - sv-SE - sw-TZ - ta-IN - te-IN - th-TH - tl - tr-TR - uk-UA - ur - vi-VN - zh-CN - zh-HK - zh-TW type: string tagline: type: string description: type: string required: - locale - tagline - description allowedCountries: anyOf: - type: object properties: mode: const: all required: - mode - type: object properties: mode: const: allow countries: type: array items: enum: - AD - AE - AF - AG - AL - AM - AO - AR - AT - AU - AW - AX - AZ - BA - BB - BD - BE - BF - BG - BH - BI - BJ - BL - BM - BN - BO - BR - BS - BT - BW - BZ - CA - CD - CF - CG - CH - CI - CL - CM - CO - CR - CV - CY - CZ - DE - DJ - DK - DM - DO - DZ - EC - EE - EG - ER - ES - ET - FI - FJ - FM - FO - FR - GA - GB - GD - GE - GF - GH - GL - GM - GN - GP - GQ - GR - GT - GW - GY - HN - HR - HT - HU - ID - IE - IL - IN - IQ - IS - IT - JM - JO - JP - KE - KG - KH - KI - KM - KN - KR - KW - KY - KZ - LA - LB - LC - LI - LK - LR - LS - LT - LU - LV - LY - MA - MC - MD - ME - MF - MG - MH - MK - ML - MM - MN - MQ - MR - MT - MU - MV - MW - MX - MY - MZ - NA - NC - NE - NG - NI - NL - 'NO' - NP - NR - NZ - OM - PA - PE - PF - PG - PH - PK - PL - PM - PS - PT - PW - PY - QA - RE - RO - RS - RW - SA - SB - SC - SD - SE - SG - SH - SI - SJ - SK - SL - SM - SN - SO - SR - SS - ST - SV - SZ - TD - TF - TG - TH - TJ - TL - TM - TN - TO - TR - TT - TV - TW - TZ - UA - UG - US - UY - UZ - VA - VC - VN - VU - WF - WS - YE - YT - ZA - ZM - ZW type: string required: - mode - countries - type: object properties: mode: const: block countries: type: array items: enum: - AD - AE - AF - AG - AL - AM - AO - AR - AT - AU - AW - AX - AZ - BA - BB - BD - BE - BF - BG - BH - BI - BJ - BL - BM - BN - BO - BR - BS - BT - BW - BZ - CA - CD - CF - CG - CH - CI - CL - CM - CO - CR - CV - CY - CZ - DE - DJ - DK - DM - DO - DZ - EC - EE - EG - ER - ES - ET - FI - FJ - FM - FO - FR - GA - GB - GD - GE - GF - GH - GL - GM - GN - GP - GQ - GR - GT - GW - GY - HN - HR - HT - HU - ID - IE - IL - IN - IQ - IS - IT - JM - JO - JP - KE - KG - KH - KI - KM - KN - KR - KW - KY - KZ - LA - LB - LC - LI - LK - LR - LS - LT - LU - LV - LY - MA - MC - MD - ME - MF - MG - MH - MK - ML - MM - MN - MQ - MR - MT - MU - MV - MW - MX - MY - MZ - NA - NC - NE - NG - NI - NL - 'NO' - NP - NR - NZ - OM - PA - PE - PF - PG - PH - PK - PL - PM - PS - PT - PW - PY - QA - RE - RO - RS - RW - SA - SB - SC - SD - SE - SG - SH - SI - SJ - SK - SL - SM - SN - SO - SR - SS - ST - SV - SZ - TD - TF - TG - TH - TJ - TL - TM - TN - TO - TR - TT - TV - TW - TZ - UA - UG - US - UY - UZ - VA - VC - VN - VU - WF - WS - YE - YT - ZA - ZM - ZW type: string required: - mode - countries description: >- Availability: all supported countries, or an allow-list / block-list of specific ones (ISO alpha-2). releaseNotes: type: string description: Publicly displayed on the app details page. securitySchemes: bearerAuth: type: http description: >- Bearer token for authentication. Use an API key from the API Keys page in team settings, or an OAuth access token. scheme: bearer ````