Add Haunt MCP (resolves #233 — conflict resolved)#288
Conversation
Resolved conflicts in README.md, plugins.json, and marketplace.json. Added Haunt MCP entries. Kept plugin files per contributor submission. Signed-off-by: Michael Kantor <6068672+kantorcodes@users.noreply.github.com>
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request integrates the Haunt MCP web extraction plugin into the marketplace and documentation. The review feedback identifies a mismatch in the plugin identifier within the marketplace configuration, recommends correcting the alphabetical sorting of the plugin list in the README, and suggests enabling model invocation in the skill definition so that agents can automatically discover and load the extraction guidelines.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
| "icon": "./plugins/Dkm0315/frappe-agent/assets/frappe-agent-icon.svg" | ||
| }, | ||
| { | ||
| "name": "Haunt MCP", |
There was a problem hiding this comment.
The name field in the marketplace entry should match the plugin's identifier in its plugin.json (which is "haunt-mcp"). Using "Haunt MCP" with spaces and capital letters will cause a mismatch and may break plugin installation or resolution. The user-facing name should be specified in displayName instead.
| "name": "Haunt MCP", | |
| "name": "haunt-mcp", |
| - [Epic Harness](https://github.com/epicsagas/epic-harness) - Auto-trigger quality skills + self-evolving agent harness — orbit (spec-to-ship), evolve (skill mutation), team (multi-agent), TDD, check, ship, simplify, debug, perf, secure. | ||
| - [Espresso](https://github.com/mirkobozzetto/espresso) - Full token-saving stack in one plugin - output compression, global rules, RTK hook, Caveman ultra, GitNexus config. Detects existing setup, installs only what's missing. Works on Claude Code and Codex. | ||
| - [Flaky Detector](./plugins/mturac/flaky-detector) - Run a test command N times, report per-test flakiness %. | ||
| - [Haunt MCP](https://github.com/Darko893/mcp-server) - Web extraction MCP for permitted public pages. Returns clean JSON, Markdown, article fields, metadata, and usage with clear blocked-page errors. |
There was a problem hiding this comment.
The list of third-party plugins is alphabetically sorted. 'Haunt MCP' (H) should be placed after 'Generative Media Skills' (G) instead of before 'Frappe Agent' (F).
| --- | ||
| name: haunt-web-extraction | ||
| description: Use Haunt MCP to extract structured JSON, clean Markdown, article fields, metadata, or usage from permitted public web pages. | ||
| disable-model-invocation: true |
There was a problem hiding this comment.
Setting disable-model-invocation: true prevents the AI agent from automatically loading this skill when it needs to perform web extraction. Since this skill contains important guidelines on how to use the Haunt MCP tools (such as using try_demo_extract first and handling blocked pages), it should be automatically discoverable by the agent. Consider removing this field or setting it to false so the agent can automatically load these instructions when relevant.
| disable-model-invocation: true | |
| disable-model-invocation: false |
2 participants
Resolves merge conflict in #233 by merging latest main.
Original PR: #233 by @Darko893
Co-authored-by: Darko893 Darko893@users.noreply.github.com