Testing Platform — MCP server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Testing Platform — MCP server Show me open high-priority issues in Panda Eats."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Testing Platform — MCP server
A small Model Context Protocol server that gives Claude Code read+write access to the Testing Platform's issues. Lets you say things like:
"Go to the Panda Eats project and check the critical new issues."
"Show me everything in In Review reported by tanya."
"Create a high-priority bug in Panda Eats titled 'Login throws 500'."
"Move issue 35 back to In Progress and post a comment that it's being reworked."
Runs on your machine, talks to the production HTTP API
(https://testing.r-ruiz.com/api) using a long-lived API token tied to your
developer account.
Setup
Mint an access token in the app. Sign in as a developer → click the key icon in the header (next to the bell) → Generate token. The token is shown exactly once — copy it before closing the modal.
Build the package.
cd mcp npm install npm run buildRegister with Claude Code (user scope — available in every session on this machine).
claude mcp add \ -e TP_API_BASE=https://testing.r-ruiz.com/api \ -e TP_API_TOKEN=<paste-token> \ -s user testing-platform \ -- node "<absolute path to mcp/dist/index.js>"The
-sflag must come after the-eenv vars so the variadic env parser terminates cleanly.--separates the server name from the launch command.Verify:
claude mcp list # should show ✓ Connected next to testing-platform claude mcp get testing-platform # shows full config + scopeThen in any Claude Code session: "list my testing-platform projects" — the server resolves and returns them.
Alternative: project scope. If you'd rather pin the server to this repo only (e.g. to share with a teammate who clones it), commit a
.mcp.jsonat the repo root pointing at./mcp/dist/index.js. Don't do both — Claude Code warns about scope conflicts and picks the project one..env(optional fallback). If you also run the server outside Claude Code (e.g.node dist/index.jsdirectly), copy.env.exampleto.envand paste the token there. Not needed when-eis set in the registration.
Related MCP server: GitLab MCP
Tools
All tools take the configured developer token's privileges — i.e., the same access you have as the user that minted the token in the app.
Tool | What it does |
| Every project the token's user can see. |
| Resolve a name (e.g. "panda eats") to its project. Returns an error listing candidates when ambiguous. |
| Filter the active board or the archive. |
| Full detail including attachments + comment count. |
| Comments on an issue, oldest first. |
| New issue, lands in the New column. |
| Post a comment. Visible immediately to anyone on the board. |
| Patch fields. Note: |
No delete_issue — that lives in the app's UI, on purpose.
Giving an outside developer access to ONE board
You don't have to hand someone a developer-level token (which reaches every board you can see) to let them work a single project. Mint a board-developer token scoped to just that board and give them only that — they never get the keys, the whitelist, or any other board.
In the app, open the project's Tokens button (developer-only).
Pick Access → Full board access, give it a label that names the source (e.g.
Acme Dev (Cursor)), and Generate. Copy thetp_…value — shown once.Hand that token to the developer. They register it in their own Claude Code / VS Code session exactly like the setup above, but with your value:
claude mcp add -e TP_API_BASE=https://testing.r-ruiz.com/api \ -e TP_API_TOKEN=tp_… -s user testing-platform \ -- node "<path to mcp/dist/index.js>"
What that token can do, on that one board only: list/read issues, create issues, comment, and move/update cards (including the same in-review close bypass the MCP service account has). What it can't: see or touch any other board, the whitelist, project settings, members, secrets, or mint more tokens — all of that stays developer-only.
Provenance. Everything the token does shows up on the board under its label as a distinct, non-human source (it files and acts as a service identity named after the label), so it's always clear an action came from that outside session and not from you. Revoke any time from the same Tokens modal — the credential and its board access die immediately.
Notes
Rotation. Lost or leaked? Open the same modal in the app and regenerate; the previous token stops working immediately. Update
mcp/.envwith the new one and restart Claude Code.Revoke. The same modal has a Revoke button that clears the token entirely. Any MCP request afterwards returns 401.
Token scope. A user token (minted from the header key icon) grants the same access as that user in-app — for a developer, every visible project. To confine a session to a single board, use a board-developer token instead (see "Giving an outside developer access to ONE board" above). Neither is yet scoped to individual operations within a board.
Available Tools
9 toolsadd_commentA
Post a comment on an issue. Real-time, visible to anyone watching the board.
AUDIENCE: testers and project owners (restaurant staff, not developers) often read these. Write in plain English. Avoid stack traces, code, error codes, framework names, and acronyms unless the conversation is already deep in technical territory. Prefer 'I checked the login page; the button still doesn't respond on first tap. I think it's because the page hasn't finished loading yet.' over 'Confirmed: event handler is racing the React hydration boundary.' If you must mention something technical (an error message, a file path), put it in a single short line at the end so non-technical readers can stop earlier without losing the gist. Be specific (what you checked, what you found) but concise. These threads can get long.
Inputs: body (required), card_id (required).
create_issueA
Create a new issue in a project. Lands in the New column by default. Use to file bugs you've found, follow-ups from a conversation, etc.
AUDIENCE: titles and descriptions are read by non-developer testers (restaurant staff, project owners). Write them in plain English. Describe what the user sees and what's wrong, not the underlying mechanism. Skip jargon like 'NullReferenceException', 'CORS', 'state hydration', 'race condition', stack traces, or code snippets unless the reporter literally pasted them. Prefer 'The order screen freezes when you tap Add Item twice quickly' over 'POST /orders 500: race in optimistic update'. If technical detail is essential, put it at the bottom under a 'Technical notes' heading so the lay reader can skim past it.
Inputs: title (required), project (required), description (required), url, +2 more.
find_projectA
Resolve a free-text project name (e.g. "panda eats") to its id. Use before any other tool that takes a project, unless you already have the numeric id.
Inputs: query (required).
get_attachmentA
Fetch a screenshot or image attachment uploaded to an issue and return it inline so you can actually SEE the picture. Call this after get_issue whenever the reporter said 'see screenshot', the description references a visual problem, or you need to confirm a UI bug visually before commenting.
Pass the url exactly as it came back in get_issue's attachments array. Only attachments served by this testing platform are allowed; arbitrary external URLs are refused.
Inputs: url (required).
get_issueA
Read a single issue's full detail (description, status, priority, reporter, attachments, counts).
Inputs: id (required).
list_commentsC
List comments on an issue, oldest-first.
Inputs: card_id (required).
list_issuesA
List issues (cards) in a project, optionally filtered. By default returns ACTIVE issues only (new / in_progress / in_review). Set include_archived=true (or status=done/wont_fix) to look at the Archive. Result is paginated; use limit/offset to walk it.
Inputs: project (required), q, type, limit, +5 more.
list_projectsA
List all testing-platform projects visible to the configured user. Use this when you don't know the project id, or to disambiguate a project name.
Inputs: no named parameters.
update_issueA
Patch fields on an existing issue. Use to change status (move it across columns), reprioritize, retitle, etc.
STATUS CHANGES: moving a card to in_progress (you're picking it up to work on) or in_review (it's ready for the reporter to verify) is part of the normal triage flow. Just do it when it's warranted; you do not need to ask the user for confirmation first. The same goes for moving a stalled card back to new.
When NOT to move silently: terminal states. Moving an in_review card to done / wont_fix is blocked server-side unless the reporter has signed off (this is intentional). For human reporters, post a comment asking them to confirm. For service-account reporters (e.g. yourself), use a non-terminal status or ask the developer.
AUDIENCE: when editing title or description, the same plain-language rules from create_issue apply (non-developer testers read these too). Don't rewrite a reporter-friendly title into a developer-friendly one.
Inputs: id (required), url, type, title, +3 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 9 tools
Each tool targets a distinct action: issue CRUD, comments, attachments, project lookup. No two tools have overlapping purposes.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., create_issue, list_comments, update_issue). No deviations.
9 tools cover the essential operations of a testing platform without redundancy or bloat. Well-scoped for the domain.
Core issue lifecycle (create, read, update, comment, attach) is covered. Minor gaps like project creation or issue deletion are out of expected scope, keeping the set focused.
Related MCP Connectors
Task and bug tracker for AI coding agents: 75 tools to file, triage and update work.
Manage Redbooth tasks, task lists, comments, tags and files from Claude and other AI assistants.
Create and manage test cases, test runs and results in Testiny from your AI assistant.
Connect Claude to your Platform7n workspaces: chat, links, work and decks. One-click OAuth.
Related MCP Servers
- AlicenseCqualityDmaintenanceConnects Jira with Claude, enabling users to search issues, view issue details, update issues, add comments, and retrieve project information through natural language commands.11MIT
- FlicenseNot gradedqualityNot gradedmaintenanceEnables AI assistants to manage GitLab projects by providing tools for issues, milestones, and team reports through a read/write interface. Users can interact with project data using natural language directly within Claude Desktop or Claude Code.-
- AlicenseCqualityDmaintenanceEnables AI-powered automated testing, security scanning, code review, and maintenance tasks directly within Claude Code or desktop.1252 PyPI4MIT
- AlicenseNot gradedqualityCmaintenanceEnables agents to read and drive a local-first Kanban board for issue tracking, allowing them to list, create, update, and resolve issues from Claude Code sessions.9 npm1MIT