AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

webmcpify

WebMCP agent skill for curated core coverage or route-by-route parity — inventory an existing web app, integrate approved tools, then inspect, verify and heal them in a real browser. Use for "webmcpify", "add WebMCP", or "expose app actions to AI agents".

Original source
github
License
MIT
Pinned revision
5269f7ea2d6c3a9bfda5ed878b6726cf509121fd

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read github--webmcpify at revision 5269f7ea2d6c3a9bfda5ed878b6726cf509121fd, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 12

tests/README.md

Workbench regressions

Validated with Node.js 24.18 and jsdom 28.1.0 for the DOM tests. JSDOM 28 requires Node.js 20.19, 22.12, or 24+; the command below also requires a Node.js release supporting --test-isolation=none. The CLI fixtures use Python 3 and POSIX pseudo-terminals for interactive consent. Do not install test dependencies without approval. If JSDOM is installed outside this checkout, set NODE_PATH to that installation's node_modules directory:

NODE_PATH=/path/to/existing/node_modules node --test --test-isolation=none skills/webmcpify/tests/*.test.mjs

No Playwright package or browser engine is downloaded by these tests. CLI tests copy the launcher into isolated temporary directories and use a fake npm command, fake Playwright module and fake browser installer. They verify refusal without writes and successful interactive/explicit consent for each separate scope.

DOM tests execute the Workbench template, including form events and handlers. Native transport is modeled; these checks do not claim browser API conformance, layout/accessibility verification, or GitHub Copilot testing. The test-only source copy exposes the private dispatch boundary to prove that it also refuses calls when the UI disabled state is bypassed; the shipped API is unchanged.