Replace grid.lightspark.com with docs.lightspark.com#513
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
Preview deployment for your docs. Learn more about Mintlify Previews.
|
✱ Stainless preview builds for gridThis PR will update the cli csharp go kotlin openapi php python ruby typescript
|
jklein24
deleted the
05-27-replace_grid.lightspark.com_with_docs.lightspark.com
branch
Greptile SummaryThis PR performs a global find-and-replace of
Confidence Score: 3/5Safe to merge only if the UMA proxy service and MCP server are simultaneously deployed at the new docs.lightspark.com subdomains; otherwise customers with live integrations will see broken payment address resolution and MCP connection failures. Most changes are benign doc-site URL swaps, but three clusters of changes touch live service infrastructure: the default UMA domain (affects payment address resolution for customers without a custom domain), the UMA proxy target subdomains (affects customer-managed reverse-proxy rules), and the MCP server URL (affects all AI-agent integrations). If the infrastructure migration is in-flight or hasn't shipped yet, merging this doc change ahead of it creates a window where the published docs point to non-existent endpoints. mintlify/global-p2p/getting-started/platform-configuration.mdx and mintlify/snippets/platform-configuration-uma.mdx (UMA proxy targets), and mintlify/platform-overview/building-with-ai.mdx (MCP server URL) — these three files describe live API endpoints that customers configure against, not just hyperlinks to the docs site.
|
| Filename | Overview |
|---|---|
| mintlify/global-p2p/getting-started/platform-configuration.mdx | Changed default UMA domain and UMA proxy target infrastructure to docs.lightspark.com — these are live service endpoints, not just doc links, and correctness depends on infra migration happening in lockstep. |
| mintlify/snippets/platform-configuration-uma.mdx | UMA proxy target subdomains updated to docs.lightspark.com; same risk as platform-configuration.mdx — these point to live infrastructure. |
| mintlify/platform-overview/building-with-ai.mdx | MCP server URL changed from mcp.grid.lightspark.com to mcp.docs.lightspark.com across all integration tabs; breaking change for existing integrators if the server isn't live at the new URL. |
| .claude/skills/grid-api/SKILL.md | Documentation fetch URLs updated from grid.lightspark.com to docs.lightspark.com — straightforward docs-site URL update. |
| components/grid-visualizer/next.config.mjs | CSP frame-ancestors updated; *.lightspark.com wildcard already covers both old and new domain so this change is safe and correct. |
| openapi/openapi.yaml | Description and hyperlinks updated from grid.lightspark.com to docs.lightspark.com — informational text only. |
| .stainless/stainless.yml | Docs link for SDK generation tool updated from grid.lightspark.com to docs.lightspark.com — config metadata only. |
| .gitignore | Adds Claude Code scratch paths (.claude/worktrees/, .claude/scheduled_tasks.lock, error-audit/) to .gitignore — safe housekeeping. |
Flowchart
%%{init: {'theme': 'neutral'}}%%
flowchart TD
A[URL in changed file] --> B{Is it a live service endpoint?}
B -- No --> C[Doc/config link\nREADME, OpenAPI desc,\nSKILL.md, stainless.yml, Footer]
B -- Yes --> D{Which service?}
D --> E[UMA default domain\ngrid.lightspark.com → docs.lightspark.com\nplatform-configuration.mdx line 20]
D --> F[UMA proxy targets\nproxyUmaSubdomain.grid.lightspark.com\n→ proxyUmaSubdomain.docs.lightspark.com\nplatform-configuration.mdx & snippet]
D --> G[MCP server\nmcp.grid.lightspark.com\n→ mcp.docs.lightspark.com\nbuilding-with-ai.mdx]
C --> H[✅ Safe — informational only]
E --> I[⚠️ Affects default UMA payment addresses]
F --> J[⚠️ Affects customer reverse-proxy rules]
G --> K[⚠️ Breaks existing MCP integrations]
Prompt To Fix All With AI
Fix the following 3 code review issues. Work through them one at a time, proposing concise fixes.
---
### Issue 1 of 3
mintlify/global-p2p/getting-started/platform-configuration.mdx:20
**Default UMA domain is now a docs subdomain**
The default UMA domain was changed from `grid.lightspark.com` to `docs.lightspark.com`. This domain appears in real UMA payment addresses (e.g., `$alice@docs.lightspark.com`) for any customer who hasn't configured a custom domain. If the underlying UMA protocol service infrastructure (which handles `.well-known/lnurlp/*`, `.well-known/lnurlpubkey`, etc.) is not also hosted at `docs.lightspark.com`, any default-domain UMA payments would fail at address resolution. Can you confirm the service layer is being co-migrated, not just the documentation site URL?
### Issue 2 of 3
mintlify/global-p2p/getting-started/platform-configuration.mdx:33-35
**UMA proxy target is a live API endpoint, not a docs URL**
The proxy target subdomains (`<proxyUmaSubdomain>.grid.lightspark.com`) are production API infrastructure that customers must route UMA protocol traffic to — they're not documentation links. Changing these to `<proxyUmaSubdomain>.docs.lightspark.com` is correct only if those subdomains are actually served from `docs.lightspark.com`. Any customer who has already configured their reverse proxy rules based on the old documentation will start dropping UMA traffic with DNS/connection errors. The same pattern appears in `mintlify/snippets/platform-configuration-uma.mdx` lines 124–126.
### Issue 3 of 3
mintlify/platform-overview/building-with-ai.mdx:113
**MCP server URL is a live API endpoint**
`mcp.grid.lightspark.com` is an active API endpoint that customers integrate with directly (it appears five times in this file across Claude Desktop, Cursor/Windsurf, and the generic HTTP tab). Changing it to `mcp.docs.lightspark.com` will immediately break any customer who copy-pastes these config blocks, unless the MCP server is actually being served from the new subdomain at the same time this documentation goes live. Is there a corresponding infrastructure deployment accompanying this change?
Reviews (1): Last reviewed commit: "Replace grid.lightspark.com with docs.li..." | Re-trigger Greptile
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
| 3. Define supported currencies and, if you are a regulated institution, the counterparty information you require | ||
|
|
||
| If you do not configure an UMA domain, Grid will use the default domain `grid.lightspark.com`. | ||
| If you do not configure an UMA domain, Grid will use the default domain `docs.lightspark.com`. |
There was a problem hiding this comment.
The default UMA domain was changed from grid.lightspark.com to docs.lightspark.com. This domain appears in real UMA payment addresses (e.g., $alice@docs.lightspark.com) for any customer who hasn't configured a custom domain. If the underlying UMA protocol service infrastructure (which handles .well-known/lnurlp/*, .well-known/lnurlpubkey, etc.) is not also hosted at docs.lightspark.com, any default-domain UMA payments would fail at address resolution. Can you confirm the service layer is being co-migrated, not just the documentation site URL?
Prompt To Fix With AI
This is a comment left during a code review.
Path: mintlify/global-p2p/getting-started/platform-configuration.mdx
Line: 20
Comment:
**Default UMA domain is now a docs subdomain**
The default UMA domain was changed from `grid.lightspark.com` to `docs.lightspark.com`. This domain appears in real UMA payment addresses (e.g., `$alice@docs.lightspark.com`) for any customer who hasn't configured a custom domain. If the underlying UMA protocol service infrastructure (which handles `.well-known/lnurlp/*`, `.well-known/lnurlpubkey`, etc.) is not also hosted at `docs.lightspark.com`, any default-domain UMA payments would fail at address resolution. Can you confirm the service layer is being co-migrated, not just the documentation site URL?
How can I resolve this? If you propose a fix, please make it concise.
| - `https://<umaDomain>/.well-known/lnurlp/*` -> `https://<proxyUmaSubdomain>.docs.lightspark.com/.well-known/lnurlp/*` | ||
| - `https://<umaDomain>/.well-known/lnurlpubkey` -> `https://<proxyUmaSubdomain>.docs.lightspark.com/.well-known/lnurlpubkey` | ||
| - `https://<umaDomain>/.well-known/uma-configuration` -> `https://<proxyUmaSubdomain>.docs.lightspark.com/.well-known/uma-configuration` |
There was a problem hiding this comment.
The proxy target subdomains (<proxyUmaSubdomain>.grid.lightspark.com) are production API infrastructure that customers must route UMA protocol traffic to — they're not documentation links. Changing these to <proxyUmaSubdomain>.docs.lightspark.com is correct only if those subdomains are actually served from docs.lightspark.com. Any customer who has already configured their reverse proxy rules based on the old documentation will start dropping UMA traffic with DNS/connection errors. The same pattern appears in mintlify/snippets/platform-configuration-uma.mdx lines 124–126.
Prompt To Fix With AI
This is a comment left during a code review.
Path: mintlify/global-p2p/getting-started/platform-configuration.mdx
Line: 33-35
Comment:
**UMA proxy target is a live API endpoint, not a docs URL**
The proxy target subdomains (`<proxyUmaSubdomain>.grid.lightspark.com`) are production API infrastructure that customers must route UMA protocol traffic to — they're not documentation links. Changing these to `<proxyUmaSubdomain>.docs.lightspark.com` is correct only if those subdomains are actually served from `docs.lightspark.com`. Any customer who has already configured their reverse proxy rules based on the old documentation will start dropping UMA traffic with DNS/connection errors. The same pattern appears in `mintlify/snippets/platform-configuration-uma.mdx` lines 124–126.
How can I resolve this? If you propose a fix, please make it concise.|
|
||
| ```text | ||
| https://mcp.grid.lightspark.com/ | ||
| https://mcp.docs.lightspark.com/ |
There was a problem hiding this comment.
mcp.grid.lightspark.com is an active API endpoint that customers integrate with directly (it appears five times in this file across Claude Desktop, Cursor/Windsurf, and the generic HTTP tab). Changing it to mcp.docs.lightspark.com will immediately break any customer who copy-pastes these config blocks, unless the MCP server is actually being served from the new subdomain at the same time this documentation goes live. Is there a corresponding infrastructure deployment accompanying this change?
Prompt To Fix With AI
This is a comment left during a code review.
Path: mintlify/platform-overview/building-with-ai.mdx
Line: 113
Comment:
**MCP server URL is a live API endpoint**
`mcp.grid.lightspark.com` is an active API endpoint that customers integrate with directly (it appears five times in this file across Claude Desktop, Cursor/Windsurf, and the generic HTTP tab). Changing it to `mcp.docs.lightspark.com` will immediately break any customer who copy-pastes these config blocks, unless the MCP server is actually being served from the new subdomain at the same time this documentation goes live. Is there a corresponding infrastructure deployment accompanying this change?
How can I resolve this? If you propose a fix, please make it concise.
…es (#518) ## Summary Updates the Stainless config to reflect the current API surface and adds five new SDK targets. ### New SDK targets - `ruby` (gem `grid`, rubygems publish disabled) - `go` (package `grid`) - `csharp` (package `grid`, nuget publish disabled) - `php` (package `grid`, composer `org-placeholder/grid`) - `cli` (binary `grid`) All have `production_repo: null` for now. ### Resource sync - `config.models`: add `embedded_wallet_config` - `sandbox`: add `cards.simulate` subresource (authorization/clearing/return) with associated card simulation models - `$shared`: add `slv_external_account_create_info`, `slv_beneficiary`, `swift_external_account_create_info` - `customers.methods`: add `generate_kyc_link: post /customers/{customerId}/kyc-link` - `agents`: add `agent_account_restrictions`, `agent_approval_thresholds` - Add top-level `cards` resource (issue / list / retrieve / update) ### Auth credential verify-request aliases — intentionally omitted `email_otp_credential_verify_request`, `oauth_credential_verify_request`, and `passkey_credential_verify_request` are **not** added as model aliases. The auth verify-request schemas are defined as: ```yaml allOf: - $ref: AuthCredentialVerifyRequest - $ref: <X>CredentialVerifyRequestFields ``` Two existing `openapi.transforms` (lines stripping `type` from `AuthCredentialVerifyRequest.properties`, then removing the `$ref` from `allOf[0]` of each verify variant) reduce these to structurally equivalent to their `Fields` siblings. Stainless collapses the allOf and emits only the `<X>CredentialVerifyRequestFields` types in `credentials.ts`, so a model alias pointing to `<X>CredentialVerifyRequest` would generate broken imports (`TS2724` / `TS2552` in `auth.ts`, `credentials.ts`, `index.ts`). The create-side aliases work because `AuthCredentialCreateRequest` retains a non-`type` field (`accountId`) after the transform, so its variants don't collapse. ### Other - `organization.docs`: `docs.lightspark.com` → `grid.lightspark.com`. **Worth a second look** — recent commits (#513, #514) migrated *away* from `grid.lightspark.com` in the docs. Confirm this is intentional before merging. ## Notes Stacked on top of #517 (document request body refactor); no code dependency between the two — sequencing is just chronological.
2 participants
No description provided.