Skip to content
Back to search

liveportfolio-MCP

MCP server that lets Claude, ChatGPT and Claude Code edit a portfolio: OAuth 2.1 + PKCE, versioned edits with undo, human-in-the-loop approval for risky changes. Architecture, security model and tool reference.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/cliffordnwanna--liveportfolio-mcp/badge.svg)](/listings/cliffordnwanna--liveportfolio-mcp)

HTML

<a href="/listings/cliffordnwanna--liveportfolio-mcp"><img src="/api/listings/cliffordnwanna--liveportfolio-mcp/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 91 lines
  1. # LivePortfolio MCP
  2. A production Model Context Protocol (MCP) server that lets AI agents (Claude, ChatGPT, Claude Code) read and edit a user's [LivePortfolio](https://liveportfolio.site) portfolio safely. It is a worked example of agent tool design with security built in: OAuth 2.1 with PKCE and per-user consent, assistant tokens accepted only at the MCP endpoint, versioned edits with compare-and-set and undo, and human-in-the-loop approval for risky changes, so an agent never gets unchecked control.
  3. **Engineering highlights**
  4. - Human-in-the-loop approval for removals, whole-list replacements, batches and new links (out of band: the agent cannot approve for itself)
  5. - Versioned, undoable edits with optimistic concurrency
  6. - Prompt-injection stance: user content is data, never instructions
  7. - 56 tools covering every editable field, each tested end to end, and a documented threat model
  8. [Architecture](docs/architecture.md) · [Security model](docs/security.md) · [Tool reference](docs/tools.md) · [Design decisions](docs/design-decisions.md)
  9. This repository holds the documentation and the registry listing (`server.json`) for the hosted server. The application itself is not open source.
  10. - Server address: `https://liveportfolio.site/mcp` (Streamable HTTP, OAuth 2.1 with PKCE)
  11. - Requires a LivePortfolio **Pro** account
  12. - Docs: https://liveportfolio.site/docs/mcp · Connect guides: https://liveportfolio.site/connect · Agent index: https://liveportfolio.site/llms.txt
  13. <!-- DEMO VIDEO: 30 to 45 seconds. Link or embed it here once it is posted. -->
  14. ## How it works
  15. ```mermaid
  16. flowchart LR
  17. A[Assistant client<br/>Claude, ChatGPT, Claude Code] -->|HTTPS + OAuth token| M[MCP endpoint<br/>/mcp]
  18. A -. sign-in .-> O[OAuth 2.1 server]
  19. O --> C[Consent page<br/>liveportfolio.site]
  20. M --> G{Checks on every call:<br/>token, app connection,<br/>access, rate limit}
  21. G --> T[Tools]
  22. T --> P[Operation layer<br/>validation + limits]
  23. P --> V[Versions and history<br/>compare-and-set]
  24. V --> D[(Database)]
  25. T -->|needs approval| Q[Approval request]
  26. Q -. link .-> U[You, on the website]
  27. U --> R[Approval page<br/>owner only]
  28. R --> P
  29. D --> S[Public portfolio site]
  30. ```
  31. ## Documentation
  32. - [Architecture](docs/architecture.md): system overview and sequence diagrams for sign-in, a direct edit and an approval-gated edit
  33. - [Security](docs/security.md): what an assistant can and cannot do, controls, prompt injection, pictures, known limits
  34. - [Tools](docs/tools.md): all 56 tools grouped by section, with limits and what the public page shows
  35. - [Design decisions](docs/design-decisions.md): trade-offs, and the known gap after approval
  36. - [Changelog](CHANGELOG.md)
  37. ## Assistants
  38. Tested: Claude (web and desktop), ChatGPT and Claude Code. Gemini CLI, Cursor and VS Code may work (any client that speaks MCP over HTTP with OAuth) but may need manual setup.
  39. ## Connect
  40. 1. Add `https://liveportfolio.site/mcp` as a custom connector. Steps with screenshots: [Claude](https://liveportfolio.site/docs/connect/claude), [ChatGPT](https://liveportfolio.site/docs/connect/chatgpt), [others](https://liveportfolio.site/docs/connect/other).
  41. 2. Sign in, choose what the assistant may do, and approve. Then start a new chat with the first message shown on the connect page.
  42. Claude Code:
  43. ```bash
  44. claude mcp add --transport http liveportfolio https://liveportfolio.site/mcp
  45. ```
  46. ## What it can do (56 tools)
  47. - **Read:** the portfolio (by section), the editing rules, designs, a completeness checklist, history, a preview of a change, a health check.
  48. - **Profile and contact:** name, role, headline, About, location, email, phone, WhatsApp, social links; add to a field or replace one phrase.
  49. - **Skills:** flat list and groups (add, remove, rename, reorder, replace).
  50. - **Experience and projects:** roles and highlight lines; projects with stack, links, pictures and an optional Power BI dashboard.
  51. - **Education, certifications, testimonials:** add, fix one, remove one, reorder, replace.
  52. - **Photos and design:** upload link or a public https picture; template, look, Pro design version.
  53. - **Safety net:** undo, restore an earlier version, and one batch of up to 60 operations for a whole-CV update.
  54. ## Security model
  55. - OAuth 2.1 with PKCE. Assistant tokens work only on `/mcp`; browser tokens do not work there, and assistant tokens are refused on every other API route.
  56. - Access is per person and checked on every call. Disconnect at https://liveportfolio.site/oauth/connections and the next call fails.
  57. - Direct edits are undoable. Removals, whole-list replacements, batches, new links and large cuts need your approval on liveportfolio.site; the assistant cannot approve for itself.
  58. - Your email, phone and WhatsApp are hidden from the assistant unless you allow them. You can connect read-only.
  59. - We do not store your chat messages. See the [privacy policy](https://liveportfolio.site/privacy).
  60. ## Limits
  61. 60 operations per batch, 30 flat skills, 1500 characters per project text field, 120 calls per 10 minutes, 60 edits per hour. The public page shows the first 5 stack items, the first testimonial and, when you have skill groups, the groups rather than the flat list; the assistant is warned when something is saved but not shown.
  62. ## It cannot
  63. Publish your portfolio, change your address or domain, touch billing, change your sign-in, delete anything, or create your first portfolio.
  64. ## Contact
  65. nwannachumaclifford@gmail.com

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.